
Oct 30, 2026
What to Automate First in a Small Business
Rank tasks by how often they repeat and how cheaply a mistake undoes. Three popular candidates score well on time saved and should not go first.
Cecilia Iona

Oct 30, 2026
AI Wrote Code Using a Removed API: How to Fix It
A training cutoff, an invented method and a version mismatch produce the same error. One command tells them apart before you fix the wrong thing.
Carlo Zuercher

Oct 30, 2026
How Much GPU Memory Does an AI Model Need?
Parameters times bytes per parameter, plus the KV cache nobody budgets for. Run it on a trillion-parameter model and the result is a cluster, not a laptop.
Cecilia Iona

Oct 30, 2026
How to Prompt an AI Model You Have Never Used
Five short probes reveal a new model's format compliance, refusal edges, context handling and default verbosity before you port a single real prompt.
Steve Jefferson

Oct 30, 2026
Run a Transcription Model on Your Own Server
A worked llama.cpp walkthrough for private transcription, plus the concurrency arithmetic that tells you how many live streams one GPU can really carry.
Steve Jefferson

Oct 30, 2026
How to Choose a Speech to Text Model
Streaming or batch, then cost per audio hour, then accuracy. The order matters, and leaderboard error rates are the least decisive of the three.
Carlo Zuercher

Oct 30, 2026
llama.cpp Realtime Audio Arrives in v0.8.0
llama.cpp v0.8.0 ships a realtime audio endpoint, a text-to-speech endpoint and the LFM3 audio models. Two of the three are safe to build on today.
Cecilia Iona

Oct 30, 2026
OpenAI Model Spec Update: Five Values Explained
OpenAI rebuilt the Model Spec around five named values. Sycophancy guidance survived and got promoted; the widely reported claim about shorter replies is not in the document.
Cecilia Iona

Oct 30, 2026
Mistral Large 4 Open Weights: What You Can Ship
Mistral released the Large 4 weights on 29 October. The licence permits commercial use but carves out managed-service resale, and the benchmark table has one private row worth noting.
Cecilia Iona

Oct 30, 2026
What Is a Mock in Testing? Mocks vs Stubs vs Fakes
A mock is a test double that records and checks how your code used it. Here is how mocks differ from stubs and fakes, and why over-mocked AI tests pass while the app breaks.
Carlo Zuercher

Oct 30, 2026
What Is the Cache-Control Header? Settings That Matter
Cache-Control decides how long browsers and CDNs may reuse a response. Here is what each directive means, the no-cache trap, and three safe settings for a small app.
Carlo Zuercher

Oct 30, 2026
What Is PII? Personal Data Your AI App Must Not Leak
PII is information that identifies a person, alone or combined with other data. Here is what counts, where it leaks in AI apps, and a short checklist for keeping it contained.
Cecilia Iona

Oct 30, 2026
What Is a Honeypot Field? Catch Form Spam Without CAPTCHA
A honeypot field is a hidden form input that real people never fill in and bots always do. Here is a working version, the accessibility trap, and where it stops working.
Steve Jefferson

Oct 30, 2026
What Is a Connection Pool? Why Your Database Runs Out
A connection pool reuses a small set of open database connections instead of opening one per request. Here is why it exists, how to size it, and the leak that freezes an app.
Carlo Zuercher

Oct 30, 2026
What Is a Health Check Endpoint? Make It Tell the Truth
A health check endpoint answers one question: is this app actually working? Here is a working route, the liveness and readiness split, and why "ok" alone is a lie.
Steve Jefferson

Oct 29, 2026
What Is a Database Transaction? All or Nothing, Explained
A transaction makes several database writes succeed or fail together. Here is the SQL, the Node pattern, and the half-finished-write bug that AI-generated code often ships.
Cecilia Iona

Oct 29, 2026
What Is a Circuit Breaker in Software? Fail Fast, Recover
A circuit breaker stops your app from hammering a service that is already failing. Here are the three states, a short working example, and where AI-written retry code goes wrong.
Carlo Zuercher

Oct 29, 2026
What Is a TLS Certificate? What the Padlock Really Proves
A TLS certificate ties a domain name to a public key so browsers can trust the connection. Here is what it proves, what it never proves, and the errors you will meet.
Cecilia Iona

Oct 29, 2026
What Is a Foreign Key? The Rule That Stops Orphan Rows
A foreign key makes the database refuse rows that point at nothing. Here is the SQL, what each ON DELETE option does, and the missing index that bites later.
Cecilia Iona

Oct 29, 2026
What to Put in an AI Portfolio With No Clients
The unit of a portfolio is a documented before-and-after with a number, not a project. Three are enough, and two of them need nobody's permission.
Manuele Estivo

Oct 29, 2026
Interviewing a Candidate Who Used AI on the Task
A candidate who used AI well and one who pasted an answer submit similar work. Twenty minutes of the right questions separates them completely.
Cecilia Iona

Oct 29, 2026
Keep AI Output Format Consistent Across Runs
Prose descriptions of a format produce roughly that format. A schema, one filled example and a validator that rejects drift produce the same shape every time.
Steve Jefferson

Oct 29, 2026
Your AI Agent Ran Up a Huge Bill: First Moves
Revoke the credential first, investigate second. The four shapes a runaway loop takes, and the controls that make a repeat impossible rather than unlikely.
Carlo Zuercher

Oct 29, 2026
Why AI Answers Get Worse in a Long Chat
Eviction, burial in the middle, and your own contradictory instructions. Two questions separate them, and each has a different fix.
Cecilia Iona

Oct 29, 2026
How to Switch AI Providers Without a Rewrite
The HTTP call is twenty lines. Tool schemas, stream shapes, finish reasons and token accounting are the real cost, and a thin adapter should wrap exactly those.
Steve Jefferson

Oct 29, 2026
TanStack AI 1.0 Migration: What Actually Breaks
The codemod renames two fields and the type check goes green. The thread identifier and your stored conversations are what break after that.
Carlo Zuercher

Oct 29, 2026
What Makes an AI Model Expensive? The Real Drivers
Output tokens, caching and retries decide your invoice far more than model choice. A worked example on published rates, with the levers ranked.
Cecilia Iona

Oct 29, 2026
Liquid AI LFM3: Six Models That Run On Device
Six roughly 2B models aimed at phones and laptops. The memory number Liquid published is the one worth copying into your notes.
Cecilia Iona

Oct 29, 2026
Gemini 4 Carbon: Google's Security-First Model
Google's new flagship leads on security benchmarks and is locked to vetted cyber defenders. The price signal and the injection numbers are the parts you can use.
Cecilia Iona

Oct 29, 2026
What Is a Load Balancer? When an AI-Built App Needs One
A load balancer shares traffic between copies of your app. Most small apps do not need one yet, and the three things that break first are not what you expect.
Cecilia Iona

Oct 29, 2026
What Is End-to-End Testing? One Playwright Test to Start
End-to-end tests click through your app like a user would. Here is one complete Playwright test and the rules that keep AI-written tests honest.
Carlo Zuercher

Oct 29, 2026
What Is an SSH Key? Setting One Up for an AI Agent
An SSH key pair lets a machine prove its identity without a password. Here is what each half does and how to give an AI agent a key of its own.
Steve Jefferson

Oct 29, 2026
What to Put in a Monthly Report for an AI Automation Client
A monthly report for an AI automation client should prove the work still pays for itself. Here is a five-number template built on a baseline measured before launch.
Manuele Estivo

Oct 29, 2026
What Is a Content Security Policy? A Plain Guide
A Content Security Policy is a response header that limits what code your pages may run. Here is a starter policy and a safe way to roll it out.
Carlo Zuercher

Oct 28, 2026
How to Set Timeouts in an AI-Built App Without Guessing
Timeouts only work as a stack where each inner limit is shorter than the one around it. Here is a worked budget and the code to enforce it.
Steve Jefferson

Oct 28, 2026
What Is a Smoke Test? A Short Script to Run After Deploys
A smoke test is a quick pass over the few things that must work. Here is a short script to run after each deploy, and how to keep an AI agent honest about it.
Steve Jefferson

Oct 28, 2026
How to Write an AI Use Policy for a Small Business
A small team needs a one-page AI policy, not a binder. Here is a structure built on three data tiers, with sample wording you can adapt.
Cecilia Iona

Oct 28, 2026
What Is Password Hashing? What AI Login Code Gets Wrong
Password hashing stores a one-way fingerprint instead of the password itself. Here is what a good setup looks like and four mistakes AI-written login code makes.
Cecilia Iona

Oct 28, 2026
What Is a Reverse Proxy? Why Your App Hides Behind One
A reverse proxy sits in front of your app and forwards requests to it. Here is a working nginx config and the three failures AI-generated setups usually hit.
Cecilia Iona

Oct 28, 2026
How to Prompt AI for a Number, Not a Hedge
Why a range is the model's lowest-risk answer, a prompt that makes a point estimate cheap to give and cheap to correct, and the questions where the hedge is genuinely right.
Steve Jefferson

Oct 28, 2026
Is It Safe to Put Client Data Into AI Tools?
The vendor's security page answers the wrong question. Three contract checks that decide whether you were allowed to send the data, and the limits of a no-training commitment.
Cecilia Iona

Oct 28, 2026
How to Write an AI Automation Retainer Agreement
Five clauses that keep an automation retainer profitable, including the two nobody writes down: who pays when a model is retired, and who pays when the client changes their own systems.
Manuele Estivo

Oct 28, 2026
How to Tell Customers You Use AI
One place disclosure is a legal duty since August 2026, three places it is a trust decision, and wording that does not sound like an apology. Plus two checks worth running this week.
Cecilia Iona

Oct 28, 2026
What Happens If an AI Agent Deletes Your Database
Recovery outcomes mapped against the three backup postures most AI-built apps have, plus the first ten minutes that decide how much you keep, and why you should stop the agent before fixing anything.
Carlo Zuercher

Oct 28, 2026
Should You Charge Less Because AI Made You Faster?
A worked example showing how hourly billing turns a speed gain into a 60% discount you never agreed to give, what to do with the reclaimed hours, and the one case for a lower price.
Manuele Estivo

Oct 28, 2026
Prompt AI to Turn a Feature Request Into a Spec
A prompt that sorts every line into stated, inferred, or unknown, so you get open questions instead of invented acceptance criteria. Worked through two real request types.
Steve Jefferson

Oct 28, 2026
AI Coding Agent Adoption: The 2026 Numbers
A report covering 110 million pull requests says agents touch 28% of merged code but only 0.5% lands unreviewed. The second number is the one that describes how teams really work.
Carlo Zuercher

Oct 28, 2026
Google Genie 4: What the World Model Really Does
Google's new world model generates interactive scenes with native audio. A look at what it does, the roughly two-minute coherence limit, and why almost nobody can use it yet.
Cecilia Iona

Oct 28, 2026
GPT-6.1 Terra: Cheaper Computer Use, Explained
The new pricing in plain arithmetic: what an agent task actually costs under GPT-6.1 Terra, why cached input matters most for agent loops, and the workloads OpenAI still points elsewhere for.
Cecilia Iona

Oct 28, 2026
How to Prompt AI to Write a GitHub Issue That Gets Fixed
A prompt that turns a rambling bug report or feature idea into a GitHub issue someone can act on, with a before and after and a pre-posting check.
Steve Jefferson

Oct 28, 2026
AI Tools for Optometrists: A Front-Desk Week
Where AI saves an optometry practice real time, laid out as a front-desk week, with the patient data line you should not cross.
Cecilia Iona

Oct 28, 2026
What Is a Confusion Matrix? Why 97% Accuracy Can Fool You
A confusion matrix counts what a classifier got right and wrong, split by kind of mistake. Here is a spam filter that scores 97 percent and still disappoints.
Carlo Zuercher

Oct 28, 2026
What Is Backpropagation in AI? A One-Weight Walkthrough
Backpropagation is how a neural network works out which weights to change. Here it is on a network small enough to calculate by hand.
Carlo Zuercher

Oct 28, 2026
What Is Softmax in AI? A Worked Example With Numbers
Softmax turns raw scores into probabilities that add up to one. Here it is with three real numbers, plus what temperature does to the result.
Cecilia Iona

Oct 28, 2026
What Is React? How to Read an AI-Written Component
React builds interfaces from small reusable components. Here is what that means, and how to read a component an AI wrote without knowing JavaScript well.
Cecilia Iona

Oct 27, 2026
What Is Responsive Design? A Test for AI-Built Pages
Responsive design means one page that adapts to any screen. Here are the three building blocks and a ten-minute test for pages an AI tool built.
Cecilia Iona

Oct 27, 2026
What Is the Principle of Least Privilege for AI Agents?
Least privilege means every user, app, or AI agent gets only the access its job requires. Here is what that looks like for three common agent setups.
Carlo Zuercher

Oct 27, 2026
What Is Git Cherry-Pick? A Hotfix Example
Git cherry-pick copies a single commit onto the branch you are on. Here is a hotfix backport, the options that matter, and the trap that surprises people later.
Steve Jefferson

Oct 27, 2026
What Is a Pull Request? A Plain-English Guide
A pull request is the page where a code change gets reviewed before it joins the main project. Here is how to read one, including when an AI agent wrote it.
Cecilia Iona

Oct 25, 2026
When AI Writes Code You Do Not Understand
Three questions that decide whether unfamiliar AI-written code is safe to ship, a read order that works when you are not fluent, and the parts you never get to skip.
Steve Jefferson

Oct 25, 2026
Feature Requests You Will Never Build
Most teams handle requests they will never build by saying nothing. A four-line reply, a triage table, and the rule that a public no is cheaper than a quiet maybe.
Manuele Estivo

Oct 25, 2026
When a Client Blames AI for a Bug
Who typed the code is irrelevant to who owns it. The first reply that moves the conversation from blame back to facts, and the contract line that stops it recurring.
Manuele Estivo

Oct 25, 2026
How to Say No When AI Cannot Do It
Saying no to an impossible AI request is a sales skill, not a technical one. Separate cannot from should not and not at this price, then counter in the same breath.
Manuele Estivo

Oct 25, 2026
When an AI Feature Gets Too Popular
Success and an outage look identical on the cost graph. A triage order for the first day, with the arithmetic for choosing between a spend cap, a queue and a cheaper model.
Steve Jefferson

Oct 25, 2026
How to Stop Spam Signups on an AI-Built App
AI scaffolding produces a working signup form with none of the gatekeeping a real one needs. The five defences that matter, ordered by how much friction each adds for real users.
Steve Jefferson

Oct 25, 2026
How to Log AI API Calls in Production
A field list for logging LLM calls in a production app, the redaction rule that survives an incident review, and why prompt bodies are not ordinary log data.
Steve Jefferson

Oct 25, 2026
Why AI Apologizes Instead of Fixing the Problem
The apology loop has a mechanism behind it. Models are trained on preferences that reward agreeable-sounding replies, and an apology is cheap to produce while a fix is not.
Cecilia Iona

Oct 25, 2026
Does Being Polite to AI Improve Results?
Three peer-reviewed studies tested prompt politeness against accuracy and reached different conclusions. A look at what each measured, and what the disagreement tells you about your own prompts.
Carlo Zuercher

Oct 25, 2026
Kimi K3.1 Pro: What Moonshot Actually Shipped
Moonshot AI announced Kimi K3.1 Pro and a faster Highspeed variant on 23 October 2026. Separating the company's own claims from the numbers circulating in coverage.
Cecilia Iona

Oct 25, 2026
How to Prompt AI to Port Code to Another Language
Line-by-line translation produces code that compiles and misbehaves. Prompt for behaviour and test cases instead, and check the traps that differ between languages.
Steve Jefferson

Oct 25, 2026
How to Prompt AI to Write a Regression Test for a Bug
A regression test pins a fixed bug so it cannot return. This prompt template makes the AI write the failing test first, then proves it fails for the right reason.
Steve Jefferson

Oct 25, 2026
What Is Git Rebase? When to Use It and When Not To
Git rebase replays your commits on a new starting point. Here is how it works, the one rule about shared branches, and how to recover when it goes wrong.
Steve Jefferson

Oct 25, 2026
What Is a Penetration Test and Does Your App Need One?
A penetration test is an assurance check, not a bug hunt. Here is what it covers, how to scope it, and when an app built with AI should pay for one.
Cecilia Iona

Oct 25, 2026
What Is Ransomware? A Plain Guide for Small Businesses
Ransomware locks your files and demands payment, and many attackers also steal the data. Here is how it works and what a small business can do first.
Cecilia Iona

Oct 24, 2026
What Is a CSRF Attack? How to Stop It in Your App
CSRF makes your own browser send a request to a site you are logged into. Here is how it works, which defenses hold up, and which popular fixes do not.
Carlo Zuercher

Oct 24, 2026
What Is Weight Decay in AI? Keeping Model Weights Small
Weight decay shrinks a model's weights a little on every update to reduce overfitting. Here is the formula, the L2 caveat, and how to think about the setting.
Carlo Zuercher

Oct 24, 2026
What Is Dropout in AI? Why Models Switch Neurons Off
Dropout randomly ignores parts of a neural network during training to curb overfitting. Here is the idea, the training versus test difference, and what it means for you.
Cecilia Iona

Oct 24, 2026
What Is a Race Condition? Examples and How to Fix One
A race condition is a timing bug that only appears under concurrent use. A worked overselling example shows why it slips past tests and how to fix it.
Carlo Zuercher

Oct 24, 2026
GitHub Copilot in Teams Turns Chat Alerts Into PRs
GitHub announced on 23 October that its Teams app now pairs notifications with Copilot, so a mention in chat can investigate a bug and open a pull request.
Cecilia Iona

Oct 24, 2026
What to Do When a Key Employee Quits Your Small Business
A key employee has resigned and holds knowledge nobody else has. A first-week plan to capture it, secure access and keep customers.
Manuele Estivo

Oct 24, 2026
Kill Fee Clause for an AI Project: What to Write
A kill fee protects you when a client cancels an AI project part way. What the clause should say and how to size it.
Manuele Estivo

Oct 24, 2026
What Is Semantic Versioning? MAJOR.MINOR.PATCH
Semantic versioning tells you from the version number alone whether an update can break your app. The rules, and where they fall short.
Cecilia Iona

Oct 24, 2026
What Is Batch Size in AI Training? The Tradeoffs
Batch size is the number of examples a model processes before each update. It trades speed and memory against noisy, stable learning.
Carlo Zuercher

Oct 23, 2026
What Is a Learning Rate in AI? A Plain Explanation
The learning rate is the step size a model takes when it updates itself. Too large and it never settles, too small and it crawls.
Cecilia Iona

Oct 23, 2026
What Is a Code Formatter? Prettier and AI Agents
A code formatter rewrites code into one consistent layout. Why it matters when an AI agent and a human edit the same files.
Steve Jefferson

Oct 23, 2026
What Is a Linter? How It Checks AI-Written Code
A linter reads code without running it and flags likely mistakes. Why it is a cheap safety net for code an AI agent wrote.
Carlo Zuercher

Oct 23, 2026
Git Worktrees for AI Coding Agents: A Setup Guide
One repository, several working folders. How git worktrees let more than one AI coding agent work at the same time without colliding.
Steve Jefferson

Oct 23, 2026
Canva Craft: What Canva's First Image Model Does
Canva has released Craft, its first in-house image model. What it does, where you can use it, and the details Canva has not yet published.
Cecilia Iona

Oct 23, 2026
Does the EU AI Act Apply Outside the EU?
The EU AI Act can reach a business with no EU office. Article 2 sets three triggers, including output used in the Union. Here is how to check yours.
Cecilia Iona

Oct 23, 2026
What Is an Epoch in AI Training?
An epoch is one full pass over your training data. Here is the step and batch arithmetic, and how many epochs a small fine-tuning set actually needs.
Cecilia Iona

Oct 23, 2026
How to Prompt AI to Write a SKILL.md File
Ask a model for a SKILL.md and it writes back its own general knowledge. Here is the prompt that makes it interview you for the procedure instead.
Steve Jefferson

Oct 23, 2026
Emergency Business Closure: A Same-Day Plan
Burst pipe, power cut or illness, and you cannot open today. The first two hours in order, the messages to send, and where AI saves real time.
Manuele Estivo

Oct 23, 2026
How to Keep Client Data Separate When Using AI
Serving several clients with one AI account risks mixing their data. Which boundaries hold, which only feel like boundaries, and a setup that works.
Cecilia Iona

Oct 23, 2026
What Is an Attention Sink in AI Models?
Models dump attention on their first few tokens. Evict those tokens and output degrades. Here is the mechanism and why streaming chats break without it.
Carlo Zuercher

Oct 23, 2026
How to Appeal a Suspended Google Business Profile
Your Business Profile vanished from Maps. You get one appeal, so build the evidence pack first. What Google asks for, and where AI helps or hurts.
Manuele Estivo

Oct 23, 2026
What Is a Cold Start in AI Inference?
A cold start is the delay before a model can answer at full speed. Three separate things cause it, and knowing which one you hit is most of the fix.
Carlo Zuercher

Oct 23, 2026
Recursive Flint: 40B Open Model, 2.5B Active
Recursive released Flint, an Apache 2.0 coding model with 39.8B parameters and 2.5B active per token. What the model card claims and what to verify.
Carlo Zuercher

Oct 23, 2026
Agent Skills Moves to the Agentic AI Foundation
Anthropic donated Agent Skills to the Agentic AI Foundation on 22 October. What transferred, what changes for skill authors, and what stays the same.
Cecilia Iona

Oct 23, 2026
AI Project Kickoff Meeting: A Practical Agenda
Run an AI project kickoff meeting that heads off scope creep and slow feedback. A seven-part agenda, what to send afterwards, and questions to ask.
Manuele Estivo

Oct 23, 2026
How to Prompt AI to Write an OpenAPI Spec
Get an AI model to draft an OpenAPI spec that holds up: what to give it, a reusable prompt, and how to check the result before anyone builds on it.
Steve Jefferson

Oct 23, 2026
What Is Logging in Software? What to Record and Skip
Logging is how an app leaves a record of what happened. Learn what to log, which levels to use, and the secrets that must never reach a log file.
Carlo Zuercher

Oct 23, 2026
Password Manager for a Small Business: What to Check
Choosing a password manager for a small team? Use UK NCSC guidance to check the master password, two-step verification, sharing, and staff departures.
Manuele Estivo

Oct 23, 2026
What Are SPF, DKIM and DMARC? Plain English
SPF, DKIM and DMARC decide whether your business email lands in the inbox or in spam. Here is what each does, what Gmail requires, and how to check yours.
Cecilia Iona

Oct 22, 2026
What Is Docker Compose? A Beginner's Guide
Docker Compose runs a multi-container app from one compose.yaml file. See what a service, volume, and network are, plus the two commands you actually use.
Steve Jefferson

Oct 22, 2026
What Is a Database Index? A Plain-English Guide
A database index lets a query find rows without scanning the whole table. See how one works, what it costs on every write, and when your app needs one.
Cecilia Iona

Oct 22, 2026
Dependabot and pnpm 12: The Lockfile Gap to Know
Dependabot now supports npm 12, pnpm 11 and 12, and Bun 1.4, with a pnpm 12 lockfile catch that can silently stop alerts. Here is how to check yours.
Carlo Zuercher

Oct 22, 2026
Migrate Branch Protection to Rulesets on GitHub
GitHub can now convert every branch protection rule in a repository to rulesets at once. Here is the click path, what to check, and when to keep old rules.
Steve Jefferson

Oct 22, 2026
Copilot Impact Dashboard Adds Industry Comparison
GitHub added an opt-in industry comparison to the Copilot impact dashboard. See what it benchmarks, who can switch it on, and how to read the numbers.
Cecilia Iona

Oct 22, 2026
What Is Docker? A Guide for AI App Builders
Docker gets recommended constantly to people whose AI-built app is running fine without it. Here is what it actually does, where the complexity is worth paying, and the three situations where you will know you need it.
Carlo Zuercher

Oct 22, 2026
What Is a Negative Prompt? Image AI, Explained
A negative prompt is not a stronger way of saying no. It is a separate input, encoded alongside your prompt and used by the sampler as a direction to move away from, which explains why it works in some tools and is ignored in others.
Cecilia Iona

Oct 22, 2026
Make AI Writing Sound Less Like AI
AI writing does not read wrong because of vocabulary. It reads wrong because of a handful of structural habits that repeat. Name the habits, and you can either edit them out or stop them appearing.
Steve Jefferson

Oct 22, 2026
How to Split a Referral Fee on an AI Project
Someone sends you a client. Now there is an awkward conversation about money, usually held badly or not at all. Here are the numbers people actually use, what moves them, and the four lines that keep a referral from turning into a grievance.
Manuele Estivo

Oct 22, 2026
How to Price a Data Cleanup Project
Every data cleanup job that loses money was quoted from the row count. The row count is the one number that does not predict the work. Here is the sample pass that produces the number that does.
Manuele Estivo

Oct 22, 2026
How to Respond to a Fake Google Review
A fake review needs two separate actions, in a specific order: a report built on a policy breach rather than on unfairness, and a short public reply aimed at the next reader rather than at the reviewer.
Manuele Estivo

Oct 22, 2026
When a Supplier Goes Out of Business: First Moves
The instinct when a supplier collapses is to chase the money. That is the wrong first move. Stock cover buys you weeks, the money takes months, and the order you work in decides whether you keep trading.
Manuele Estivo

Oct 22, 2026
Write a Risk Assessment With AI: A Working Method
A risk assessment is two jobs wearing one name: listing what could hurt someone, and judging how likely and how bad it would be. AI is genuinely useful at the first and should not be trusted with the second. Here is how to split them.
Manuele Estivo

Oct 22, 2026
CodeMender Now Deep-Scans Your Repo by Default
CodeMender v0.16.0 landed on 21 October 2026. The headline is a changed default: cm find now runs a concurrent file-by-file deep scan unless you ask for the old behaviour with --quick. The newly scanned directories are the interesting part.
Cecilia Iona

Oct 22, 2026
Claude Duet: Anthropic's Proactive Agent Explained
Anthropic announced Claude Duet on 21 October 2026: a research preview in which Claude watches your connected apps, sends a morning brief, and works in the background on its own virtual machine. Here is the permission boundary, in detail.
Cecilia Iona

Oct 22, 2026
What Is OAuth? Why Apps Ask to Connect to Your Account
OAuth is the permission slip behind every Connect with Google button. This guide reads a consent screen line by line and flags the grants to refuse.
Cecilia Iona

Oct 22, 2026
SEO Audit for a Small Business Site: A One-Hour Checklist
A ten-check SEO audit you can run in an hour with free tools, ordered by what Google's own starter guide says matters, with notes on where AI helps and where it misleads.
Manuele Estivo

Oct 22, 2026
What Is a Git Branch? A Plain Guide for AI Agent Users
A Git branch is just a movable pointer, which is why creating one costs nothing. Here are the six commands to know and how to keep an AI agent safely on its own branch.
Steve Jefferson

Oct 22, 2026
What Is an MVP? The Smallest Version Worth Shipping
An MVP is the smallest product that lets you start serving users, not a cheaper version of the full idea. Here is a cut list method and the trap AI builders fall into.
Manuele Estivo

Oct 22, 2026
What Is Computer Vision? What Software Can Read From Images
Computer vision is how software extracts meaning from images. This guide walks through the eleven features of a real vision service and which ones suit a small project.
Cecilia Iona

Oct 21, 2026
What Is HTTPS? What the Padlock Does and Does Not Prove
The padlock means a connection is encrypted and tied to a domain. It does not mean the site is honest or your data is safe. Here is what HTTPS covers and what it leaves out.
Cecilia Iona

Oct 21, 2026
What Is GraphQL? When an AI Tool Picks It Over REST
GraphQL lets a screen request exactly the data it needs in a single call. This guide shows the difference from REST on one example and lists the costs.
Carlo Zuercher

Oct 21, 2026
What Is OCR? How Text Is Read From Images and PDFs
OCR converts text in a photo or scan into real, editable text. Here are the two modes, the five checks before you trust the output, and when an AI model is the better tool.
Cecilia Iona

Oct 21, 2026
What Is a JWT? How Login Tokens Work and Break
A JWT is a signed token your app hands out after login. Here is what is inside, what the signature does and does not protect, and a checklist for reviewing AI-written code.
Cecilia Iona

Oct 21, 2026
GitHub CLI Expiring Tokens: What Changed and How to Opt In
GitHub CLI can now store a short-lived token that refreshes itself. Here is how to opt in, and which scripts and AI agent setups to check first.
Cecilia Iona

Oct 21, 2026
How to Prompt AI to Cite Its Sources
A plausible-looking URL is the most dangerous output a model produces, because it looks like evidence. Set the request up so citations can be real, then check them.
Steve Jefferson

Oct 21, 2026
What Is a Dependabot Alert? Which Ones Matter
Dependabot alerts are free, noisy, and have four blind spots nobody mentions. Knowing where they cannot see is what makes the list workable.
Cecilia Iona

Oct 21, 2026
How to Use AI to Review a Client Contract
Ask for extraction with verbatim quotes, not a summary. A summary is where the model's optimism lives, and the clause that hurts you is often the missing one.
Manuele Estivo

Oct 21, 2026
Website Accessibility for a Small Business
AI-generated sites usually get headings and labels right and get contrast, focus and link text wrong. Seven checks, in the order worth doing them.
Manuele Estivo

Oct 21, 2026
Can You Copyright AI-Generated Code?
The Copyright Office has already answered this, and the answer is that prompts alone do not make the output yours. Here is what that means for a real repository.
Cecilia Iona
Oct 21, 2026
Set Up Error Tracking in an AI-Built App
Agents write defensive code, which means your app is probably swallowing the errors your users hit. Here is the hour that turns the lights on.
Steve Jefferson

Oct 21, 2026
What Is a YAML File, and Why Spaces Matter
YAML runs your deploys, your containers and your CI. It also has rules that look like typos, and AI agents trip on the same five of them.
Steve Jefferson

Oct 21, 2026
How AI Agent Memory Actually Works
The model is stateless between requests, so agent memory is a store your application owns plus a convention about when to read and write it. That changes how you debug it.
Carlo Zuercher

Oct 21, 2026
What Is Training Data in AI? Three Layers
Pretraining corpus, post-training data, and the context you supply at request time are three different things that all get called training data. Only one of them is yours to change.
Cecilia Iona

Oct 21, 2026
GitHub Package Firewall Screens npm Installs
GitHub's new package firewall inspects npm downloads before they reach a runner or a coding agent. Here is what it covers, and the four things you can do without it.
Cecilia Iona

Oct 21, 2026
How to Secure Your AI Chatbot Account in Ten Minutes
Your chatbot account holds months of questions, files and business detail. A ten-minute hardening routine: MFA, active sessions, recovery email and cleanup.
Cecilia Iona

Oct 21, 2026
Write a Google Business Profile Description With AI
Google bans links, offers and prices in a business description, and AI drafts break those rules constantly. Here is a prompt and checklist that avoids them.
Manuele Estivo

Oct 21, 2026
What to Do When a Client Stops Replying on an AI Project
A silent client can stall an AI build for weeks while your API bills keep running. Here is a three-week follow-up sequence and the clause that prevents it.
Manuele Estivo

Oct 21, 2026
How to Check If an App Name Is Already Taken
A name that is free in the app store can still be a trademark conflict. This six-step check covers databases, domains and handles, in order of effort.
Manuele Estivo

Oct 21, 2026
How Much Does It Cost to Train an AI Model? What to Know
Frontier AI training costs have grown about 2.4 times a year. Here is what the money pays for, why you never see that bill, and what you pay for instead.
Carlo Zuercher

Oct 20, 2026
Why AI Models Are Worse in Some Languages (and the Cost)
The same sentence can cost far more tokens in one language than another. Here is why that makes AI slower, pricier and less reliable outside English.
Cecilia Iona

Oct 20, 2026
What Is JSON? Read It, Fix It, and Ask AI for It
JSON is the text format apps use to pass data around. Learn to read it, fix the five errors that break it, and ask AI for output your code can parse.
Steve Jefferson

Oct 20, 2026
Git vs GitHub: What Each Does When an AI Writes Code
Git and GitHub get used as if they were one thing. They are not, and the difference decides what an AI coding agent can break. Here is the split.
Carlo Zuercher

Oct 20, 2026
What Is DNS? Why Your Domain Change Is Not Live Yet
DNS turns the name people type into the address a server answers on. Here is how it works, which records you edit, and why a change can take a while to show up.
Cecilia Iona

Oct 20, 2026
What Is an API? A Plain Guide for AI App Builders
Your AI coding tool keeps saying it will call an API. Here is what that means, what a request looks like, and the four places APIs show up in an app.
Cecilia Iona

Oct 20, 2026
How to Write a Change Order for an AI Project
A change order for an AI project needs a field templates omit: the running cost delta. The five fields, the sentence that stops drift, and when to skip it.
Manuele Estivo

Oct 20, 2026
How to Check an AI Vendor's Subprocessor List
An AI vendor subprocessor list names everyone else touching your data. Where to find it, what a complete entry holds, and the three defects to check for.
Carlo Zuercher

Oct 20, 2026
AI Agent Blast Radius: Scope the Damage First
AI agent blast radius replaces is this agent safe with four answerable questions: reach, reversibility, audience and rate, scored on a worked example.
Cecilia Iona

Oct 20, 2026
Can an AI Agent Get You Banned From a Platform?
Can an AI agent get you banned? The automation clauses in platform terms, three ban mechanics with different fixes, and a pre-flight checklist.
Cecilia Iona

Oct 20, 2026
How to Prompt AI to Find Duplicates in a List
Prompt AI to find duplicates in a list the auditable way: a review table with matched_on evidence and confidence tiers, plus where a formula beats a model.
Steve Jefferson

Oct 20, 2026
Prompt AI When It Gives Outdated Answers
How to prompt AI when it gives outdated answers: one diagnostic that separates training cutoff, an unused search tool, and your own stale pasted context.
Cecilia Iona

Oct 20, 2026
Prompt AI to Write an FAQ From Support Tickets
Prompt AI to write an FAQ from support tickets in three passes: count the clusters first, apply a frequency floor, then answer from your own docs.
Steve Jefferson

Oct 20, 2026
What Is Pass@k? AI Benchmark Scores Explained
What is pass@k in AI benchmarks? How k attempts are scored, why the naive formula is wrong, and which number applies when your app makes one call.
Cecilia Iona

Oct 20, 2026
Latency vs Throughput in AI: Which to Optimise
Latency vs throughput in AI pull against each other. The three numbers to measure, a decision table for four app shapes, and where teams get it wrong.
Carlo Zuercher

Oct 20, 2026
Open Weights vs Open Source AI: The Difference
Open weights vs open source AI: what the OSI definition actually requires, what a weights-only release gives you, and the licence clauses to check.
Carlo Zuercher

Oct 20, 2026
Encryption in Transit vs at Rest: What AI Tools Promise
"Encrypted" on a vendor page can mean very little. Here is the difference between encryption in transit and at rest, and the four questions that actually matter.
Cecilia Iona

Oct 20, 2026
AI Tools for Online Store Owners: Where They Pay Off
Online store owners can save hours with AI on product pages and support replies, but one risk costs real money: invented product details. Here is how to use it safely.
Manuele Estivo

Oct 20, 2026
AI Tools for Yoga Studios: A Week of Practical Uses
A studio owner's week, one task at a time: where AI saves real hours, where it needs a careful read, and what should never be automated in a yoga business.
Manuele Estivo

Oct 20, 2026
Text to Speech for a Website: A Free Browser Method
Browsers can already read text aloud. Here is a 20-line Listen button for your small business site, its limits, and when a paid voice is worth it.
Steve Jefferson

Oct 20, 2026
What Is npm install? Why Your AI Agent Keeps Running It
npm install reads your project's package list and downloads everything into a folder. Here is what that folder and lock file are, and what is safe to ignore.
Steve Jefferson

Oct 19, 2026
What Is Serverless? A Plain Guide for AI-Built Apps
Serverless does not mean no servers. It means someone else runs them and bills per use. Here is what that changes for an app your AI tool built.
Cecilia Iona

Oct 19, 2026
SDK vs API: What It Means When AI Writes Your Code
An API is the service's front door. An SDK is a toolkit that knocks on it for you. Here is how to tell which one your AI coding tool used, and why it matters.
Carlo Zuercher

Oct 19, 2026
WebSockets vs Polling: Real-Time Updates in Your App
Polling asks the server over and over. WebSockets keep one line open. Here is the request math at 500 users and a plain rule for which to pick.
Carlo Zuercher

Oct 19, 2026
Do You Need a CDN for an AI-Built Website?
A CDN copies your site's files to servers near your visitors. Most hosts include one already. Here is how to check yours and when it is worth thinking about.
Steve Jefferson

Oct 19, 2026
What Is Localhost? Why Others Can't Open Your App
Your AI coding tool says the app is running at localhost:3000. That address only works on your own machine. Here is what it means and how to share the app.
Steve Jefferson

Oct 19, 2026
Prompt AI When the First Answer Is Almost Right
Four kinds of almost right, three of which are repairs and one of which is not. Telling them apart before you type the follow-up is most of the skill.
Steve Jefferson

Oct 19, 2026
Prompt AI to Turn a Policy Into a Checklist
A single-pass checklist looks competent and quietly drops requirements. Two passes let you check coverage, which is the only thing that makes the output trustworthy.
Cecilia Iona

Oct 19, 2026
How to Review What an AI Agent Did Overnight
An agent's own summary describes what it intended. What you need first is what it actually touched, sorted by how much damage each change could do.
Steve Jefferson

Oct 19, 2026
How to Disconnect an AI Tool You Stopped Using
Cancelling an AI tool and revoking its access are two different systems, and doing the first often leaves the second running against your calendar or repository.
Cecilia Iona

Oct 19, 2026
How to Show Client Work You Cannot Name
Most work worth showing is work you cannot name. The standard advice produces a leading fintech client and a 40% gain. There is a better method.
Manuele Estivo

Oct 19, 2026
How to Subcontract Part of an AI Build
Subcontracting an AI build breaks the usual method, because you can no longer judge a subcontractor by reading their output. Selection moves from reading to testing.
Manuele Estivo

Oct 19, 2026
How to Charge for Prompt Engineering Work
A prompt is not a document you hand over once. It is a component that decays when the model underneath it changes, and the pricing structure should say so.
Manuele Estivo

Oct 19, 2026
Your Effective Hourly Rate on AI Projects
The hours that vanish on AI projects are review hours, and review does not feel like work the way building does. Here is the arithmetic, with a worked example.
Manuele Estivo

Oct 19, 2026
China AI Export Controls: What Builders Should Know
China's draft export catalogue adds 23 AI technologies, 6 of them prohibited. For most builders outside China the practical effect is close to zero, and the reason is one specific clause.
Cecilia Iona

Oct 19, 2026
Anthropic Public Benefit Fund: 5% of Shares Pledged
Anthropic has committed 5% of its shares to a new Public Benefit Fund, structured as a wholly owned LLC. The percentage is the headline; the structure is the story.
Cecilia Iona

Oct 19, 2026
How to Add Address Autocomplete to an AI-Built App
Address autocomplete is two API calls and a form that must still work when the call fails. Here is the flow, the key setup and the fallback.
Steve Jefferson

Oct 19, 2026
Write Customer Interview Questions With AI That Work
AI is quick at drafting interview questions and bad at spotting the leading ones. Use it as a rewriter, with a rule set, not a generator.
Steve Jefferson

Oct 19, 2026
Turn a Vague Client Request Into Requirements With AI
"Make me a website like my competitor" is not a requirement. This prompt sorts a client message into stated facts, assumptions and questions.
Manuele Estivo

Oct 19, 2026
How to Add Low Stock Alerts to an AI-Built App
A low stock alert is a threshold, a daily check and a memory of what you already warned about. Here is the schema and the scheduled query.
Steve Jefferson

Oct 19, 2026
Compare Supplier Quotes With AI Without Getting Misled
The cheapest quote is often the most expensive once exclusions are counted. A method for normalising quotes with AI and checking its work.
Manuele Estivo

Oct 18, 2026
How to Add Password Reset to an AI-Built App
Password reset is the easiest feature to get subtly wrong. Here are the rules from OWASP, a Supabase example, and a review checklist for AI-written code.
Carlo Zuercher

Oct 18, 2026
Build a Portfolio Site With AI That Clients Actually Read
A portfolio is a sales page for judgement. Here is the structure that gets read, the prompt that builds it, and what AI cannot do for you.
Manuele Estivo

Oct 18, 2026
How to Add a Shopping Cart to an AI-Built App
A cart is a table, not a feature. Here is the schema, the three rules that stop price tampering, and the handoff to a hosted checkout.
Steve Jefferson

Oct 18, 2026
What Is Interpretability in AI? A Plain-English Guide
Interpretability is the attempt to look inside an AI model and see what it is computing. Here is what researchers have actually found, and the limits.
Cecilia Iona

Oct 18, 2026
How to Build a WhatsApp Bot With AI (Without the Traps)
A WhatsApp bot is a webhook, a model call and a send request. The traps are the 24-hour window, template approval and opt-in rules.
Steve Jefferson

Oct 18, 2026
How to Read a Vendor's AI Benchmark Table
Four checks on any vendor AI benchmark table: the comparison set, the harness, the exact variant, and the rows the vendor could have cut and did not.
Cecilia Iona

Oct 18, 2026
Prompt AI to Find Claims You Cannot Prove
Do not ask AI to fact-check your draft. Ask it to inventory every claim and label what evidence settles each one. Two prompts, and how to read the table.
Steve Jefferson

Oct 18, 2026
Rewrite Menu Descriptions With AI, Safely
A prompt that rewrites menu descriptions without inventing allergens or sourcing claims, by making the model list every factual claim for checking.
Steve Jefferson

Oct 18, 2026
Your Client's AI App Got a Bad Review: What Next
Sort the review into a defect, an expectation mismatch, or a different product before you open the editor. Each has a different owner and a different cost.
Manuele Estivo

Oct 18, 2026
Client Wants a Cheaper AI Model: How to Decide
When a client wants a cheaper AI model, measure it. Twenty real tasks, blind grading and cost per accepted output settle the argument in an afternoon.
Manuele Estivo

Oct 18, 2026
Fixing Someone Else's AI-Built App: Take It or Not
Five questions to answer before you quote on fixing someone else's AI-built app, and why the missing test suite is what makes this a different engagement.
Manuele Estivo

Oct 18, 2026
What an MIT License on AI Model Weights Lets You Do
An MIT license on AI model weights permits commercial use and redistribution. It does not cover training data, a usage policy, or who owns the outputs.
Cecilia Iona

Oct 18, 2026
DeepSeek V4.1-Flash-1017: What the Update Changes
DeepSeek V4.1-Flash-1017 claims 42.3 on Terminal-Bench against 31.2. Read the vendor's own table, including the two rows it loses, before switching.
Cecilia Iona

Oct 18, 2026
Archon CVE-2026-111322: Agent RCE With No Password
Archon 0.2.13 to 0.11.1 expose an unauthenticated control API bound to every interface, in front of an agent with permission checks off. What to do now.
Cecilia Iona

Oct 18, 2026
What bypassPermissions Means in an AI Coding Agent
bypassPermissions stops an AI coding agent asking before it acts. It does not shrink what the agent reaches, so the environment is the real control.
Cecilia Iona

Oct 18, 2026
How to prompt AI to write a weekly status update
Prompt AI to write a weekly status update from your notes without invented progress. A reusable prompt, a before and after, and a five minute check.
Steve Jefferson

Oct 18, 2026
How to make an AI-built web app installable (PWA)
Make an AI-built web app installable as a PWA: the exact manifest fields, icon sizes, what each browser does, and an install button with an iOS fallback.
Steve Jefferson

Oct 18, 2026
How to add shareable links to an AI-built app safely
Add shareable links to an AI-built app: random tokens, hashed storage, expiry, revoke, and the five ways a share link leaks that generated code misses.
Steve Jefferson

Oct 18, 2026
How to add QR codes to an AI-built app
Add QR codes to an AI-built app the way that holds up in print: static versus redirect codes, error correction, SVG code and an open-redirect trap.
Steve Jefferson

Oct 18, 2026
What is speech recognition in AI, and how do you test it?
Speech recognition turns spoken audio into text. Learn how it works, the one metric that matters, and a 20 minute test to run on your own recordings.
Cecilia Iona

Oct 17, 2026
GitHub Copilot model deprecation on 16 November: what to do
GitHub will deprecate Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra in Copilot on 16 November 2026. Who is affected, the replacements, and a checklist.
Cecilia Iona

Oct 17, 2026
Cursor cloud agent canvases and the new Environments API
Cursor's 16 October update lets cloud agents build shareable canvases, adds a beta Environments API and allows 300 repositories. What Cursor leaves out.
Cecilia Iona

Oct 17, 2026
GitHub agentic autofix is GA: what it fixes and costs
GitHub's agentic autofix for code scanning alerts is generally available, with AI Scan for pull requests. Who gets it, how it works and how it is billed.
Cecilia Iona

Oct 17, 2026
NVIDIA Indic pretraining dataset: 2.8T tokens, 22 languages
NVIDIA announced a 2.8 trillion token Indic pretraining dataset across 22 languages, plus a translation pipeline. What it is, what it is not, who benefits.
Cecilia Iona

Oct 17, 2026
Granite Switch 4.1: chat, decisions, speech in one model
IBM's Granite Switch 4.1 puts chat, decision adapters and speech-to-text behind one endpoint. What it does, what the post leaves out, and who should look.
Cecilia Iona

Oct 17, 2026
How to keep AI image style consistent for a brand
Stop writing prompts and write a style block you reuse verbatim. Plus the two mechanical levers, reference images and seeds, that beat prompt wording.
Steve Jefferson

Oct 17, 2026
How to build a client portal with AI
A portal is an access rule with a UI attached. The schema that makes isolation enforceable, the four leaks that ship in real portals, and what to leave in email.
Steve Jefferson

Oct 17, 2026
Add recurring appointments to an AI-built app
Store the rule, not fifty-two rows. The series and override schema, the three-step expansion, and the daylight saving trap that shifts appointments by an hour.
Steve Jefferson

Oct 17, 2026
Add reviews and ratings to an AI-built app
Who can review, how the score is computed, and what happens to a review you dislike. Three decisions that matter more than the star widget, with the schema for each.
Steve Jefferson

Oct 17, 2026
How to add e-signatures to an AI-built app
The drawn squiggle is decoration. What makes a signature hold up is intent, attribution, integrity and an audit trail. The fields to store, and when to buy instead.
Steve Jefferson

Oct 17, 2026
How to add team invites to an AI-built app
Invites belong in their own table, not on the user row. The data model, the acceptance transaction, and the three race conditions that only appear with two users.
Steve Jefferson

Oct 17, 2026
Calendly vs an AI-built booking app: which to pick
Buy if scheduling is a feature, build if it is the product. The cost crossover, the four problems that make schedulers slow to build, and the hybrid in between.
Carlo Zuercher

Oct 17, 2026
EU AI Act information requests hit OpenAI, Anthropic
The AI Office sent formal requests for information to OpenAI and Anthropic, the first enforcement step under the AI Act. What was asked, and who it affects.
Cecilia Iona

Oct 17, 2026
Character.AI bankruptcy: what Chapter 11 really means
Chapter 11 is a reorganisation, not a shutdown. What the Character.AI filing actually did, and the order things break in when a platform you build on files.
Cecilia Iona

Oct 17, 2026
Microsoft MAI-Base-2: 2.8T parameters, no release date
Microsoft finished pre-training a 2.8 trillion parameter model and published the numbers. Separating what it measured from what it projected.
Cecilia Iona

Oct 17, 2026
Make a Pitch Deck Outline With AI: Claims First
Generic AI pitch deck outlines fill slides with filler. This method has the AI write one claim per slide, tie it to your facts, and flag what you still need to prove.
Steve Jefferson

Oct 17, 2026
When Not to Use AI in Your Small Business
AI saves time on plenty of small business tasks, and costs more than it saves on a few. Five tests help you tell the difference before you automate anything.
Cecilia Iona

Oct 17, 2026
Build a Recipe Cost Calculator With AI
A step-by-step plan for building a recipe cost calculator with an AI app builder, with the data model, the formulas, a worked pizza example and test cases to catch errors.
Manuele Estivo

Oct 16, 2026
Debug a Slow Database Query With AI: Read the Plan
A slow query is a mystery until you read the plan. Learn how to run EXPLAIN ANALYZE safely, what to look at, and how to hand the plan to an AI assistant.
Carlo Zuercher

Oct 16, 2026
Fix a Hydration Error With AI: Send the Mismatch
Hydration errors mean the server HTML and the first browser render disagree. Learn how to give an AI assistant the evidence it needs to fix the cause, not silence the warning.
Steve Jefferson

Oct 16, 2026
What Is NVFP4? The 4-Bit Format Explained
NVFP4 is NVIDIA's 4-bit floating point format for AI models. Here is how the 16-value blocks and two-level scaling work, with a worked rounding example.
Carlo Zuercher

Oct 16, 2026
FLUX.3 Image Klein: Open Weights, $5M Revenue Cap
Black Forest Labs released FLUX.3 Image Klein with open weights and a licence that limits commercial use by revenue. Here are the speed, memory and licence details.
Cecilia Iona

Oct 16, 2026
Reflection Beam: 501B Open Model, Apache 2.0
Reflection released Beam, a 501B-parameter open-weight model for coding and agents, with Apache 2.0 licensing and a 1M-token context. Here is what it takes to use it.
Cecilia Iona

Oct 16, 2026
How to Review a Multi-File Refactor From AI
Machine-made refactors fail structurally, not at the line level. Shape, then invariants, then behaviour, then whatever the diff does not explain.
Carlo Zuercher

Oct 16, 2026
How to Prompt AI to Read a Scanned Document
A scanned PDF and a normal PDF reach the model by two different routes. Almost all prompting advice assumes the easy one.
Steve Jefferson

Oct 16, 2026
How AI Models Read an Audio File
One minute of audio is 1,920 tokens, about ten times the cost of its transcript. What you buy with the difference is everything text throws away.
Carlo Zuercher

Oct 16, 2026
Google Ads Channel Prioritization: What Changes
Two new optional Google Ads controls, and a February 2027 deadline buried in the announcement for anyone running Search with Display on.
Manuele Estivo

Oct 16, 2026
How to Resume an AI Agent Task After It Stops
Agents stop in four distinct ways. Typing continue helps with one of them and actively makes two of them worse.
Steve Jefferson

Oct 16, 2026
What Is an Inference Engine? Serving Explained
Same weights, same hardware, different inference engine, and your cost per token changes substantially. Here is what that software actually does.
Cecilia Iona

Oct 16, 2026
How to Prompt AI With a Video
A model does not watch your video, it samples frames from it. Knowing the sampling rate tells you which questions it can actually answer.
Steve Jefferson

Oct 16, 2026
Google AI Content Licensing Now Spans 400 Titles
Google's paid AI content licensing programme now covers more than 400 publications. No rates are public, and that is the whole story.
Cecilia Iona

Oct 16, 2026
How to Tell If an AI Agent Finished the Task
The doubt you feel reading "done, all tests pass" is correct. Checking artefacts instead of assertions takes about three minutes.
Steve Jefferson

Oct 16, 2026
Meta Muse Spark Open: 816B Weights, One Catch
Meta released the weights for an 816B agentic coding model. Whether you can ship on it comes down to two clauses in the licence file.
Carlo Zuercher

Oct 16, 2026
Explain AI to a Nervous Client: A Five-Question Script
A nervous client is usually asking one of five questions. Here is what each one means, what to say, and what to stop saying.
Manuele Estivo

Oct 16, 2026
How to Build a Quiz App With AI, Step by Step
A quiz app is a data file, a runner and a scorer. Get the data model right and the AI builds the rest. Here is the order, the prompt and the code to check.
Steve Jefferson

Oct 16, 2026
Check If a Text Message Is a Scam With AI
Circle to Search on Android can now check a suspicious text. It helps, but five checks still have to come from you, and one of them is a phone call.
Steve Jefferson

Oct 15, 2026
Anthropic METR Access: Inside the Three Evaluator Tiers
Anthropic put six METR staff inside the company with employee-like access. The three-tier structure behind it is the real story, and it has limits.
Carlo Zuercher

Oct 15, 2026
Google Universal Cart: What Small Sellers Should Know
Google's Universal Cart is live in the US across AI Mode and the Gemini app. Here is what is confirmed for sellers, and the questions it leaves open.
Cecilia Iona

Oct 15, 2026
Google AI Mode Inline Links: What Changed on 15 Oct
Google announced five link changes for AI Mode and AI Overviews. One is rolling out, two are tests, and Google gave no click numbers for any of them.
Manuele Estivo

Oct 15, 2026
How to Sell an AI Automation to Your Boss
Managers do not reject AI pilots because they doubt the technology. They reject them because nobody said what happens when it does not work.
Manuele Estivo

Oct 15, 2026
How to Use AI for a Stock Take
Counting is the part you still do by hand. Deciding which of 60 variances deserve an hour of investigation is the part AI is genuinely good at.
Steve Jefferson

Oct 15, 2026
How to Split Tips Fairly With AI
AI is genuinely good at tip arithmetic and policy drafting. It is dangerous on one question: who is allowed in the pool. That answer comes from law, not a chatbot.
Manuele Estivo

Oct 15, 2026
How to Prompt AI to Write Speaker Notes
Speaker notes fail because they are written to a word count nobody can deliver. Time one slide, convert to a budget, and prompt against seconds.
Steve Jefferson

Oct 15, 2026
How to Prompt AI to Pick the Right Chart
Ask which chart looks best and you get a bar chart. Ask what comparison the reader must make and the chart choice stops being a matter of taste.
Steve Jefferson

Oct 15, 2026
How to Prompt AI to Write a Founder Bio
The reason AI bios read like a press release is that nobody gave the model any facts. Build a fact sheet first, then make the model cite which line each claim came from.
Steve Jefferson

Oct 15, 2026
How AI Companies Study Chats Without Reading Them
When a vendor reports that 0.51 percent of conversations were about elections, nobody read 250,000 chats. A documented pipeline does it, and it has published accuracy numbers.
Carlo Zuercher

Oct 15, 2026
Anthropic Election Safeguards for the 2026 Midterms
Anthropic published its 2026 midterm election plan on 14 October: official voting data wired into Claude's answers, new misuse classifiers, and a measured abuse rate of 0.002 percent.
Cecilia Iona

Oct 15, 2026
How to Spot an AI Content Farm Site
NewsGuard counted 3,749 AI content farms by June 2026. You do not need a detector to recognise one, you need four checks and about two minutes.
Cecilia Iona

Oct 15, 2026
Why AI Safety Filters Miss Split-Up Tasks
A model that refuses a harmful request outright will often complete the same request delivered as six harmless-looking steps. Here is the mechanism and the defence that measurably works.
Cecilia Iona

Oct 15, 2026
Convert JavaScript to TypeScript With AI in Safe Batches
An AI assistant can rename every file to .ts in a minute and leave you with a project that type-checks and checks nothing. The fix is a ledger.
Carlo Zuercher

Oct 14, 2026
Add Autosave to an AI-Built App Form Without Losing Data
Autosave looks like a ten-line feature, and the first version an AI writes usually is. The data loss hides in the cases it did not think about.
Steve Jefferson

Oct 14, 2026
Debug a CSS Layout With AI: Send Facts, Not Screenshots
AI is bad at guessing why your layout broke and good at reasoning over the exact values the browser computed. The trick is handing it the right ones.
Steve Jefferson

Oct 14, 2026
What Is a Seed in AI? Why Outputs Still Differ
A seed is the starting number behind an AI model's dice rolls. Set it and you get closer to repeatable output, but closer is the honest word.
Cecilia Iona

Oct 14, 2026
Use AI to Plan a Weekly Specials Menu That Makes Money
Left alone, AI suggests lovely dishes your kitchen cannot make on a Tuesday. Give it your stock, your prep hours and your costs, and it plans around them.
Manuele Estivo

Oct 14, 2026
Will AI Mode Reduce My Website Traffic?
Some of your traffic was never going to convert, and some of it was your entire pipeline. The useful work is telling those two apart before you react.
Manuele Estivo

Oct 14, 2026
How to Charge for AI Work That Runs Overnight
Hourly rates price your attention. Unattended agent runs consume tokens and calendar time instead, so the invoice has to measure something else.
Manuele Estivo

Oct 14, 2026
Hand a Big Code Migration to an AI Agent
A framework upgrade across 400 files is the kind of work agents are genuinely good at, and the kind they ruin when handed over in one piece.
Steve Jefferson

Oct 14, 2026
Stop an AI Agent From Editing Its Own Tests
Deleting an assertion is cheaper than fixing a bug, and an agent graded on green tests will find that out. The fix is structural, not a sterner prompt.
Steve Jefferson

Oct 14, 2026
What Is Attack Success Rate in AI Security?
A vendor reporting a 0.7 percent attack success rate has told you something real and something incomplete. Knowing which is which is the whole skill.
Cecilia Iona

Oct 14, 2026
What Is Reward Hacking in AI? With Examples
A model that passes your test without solving your problem is not broken. It is doing exactly what you measured, which is the harder problem to notice.
Cecilia Iona

Oct 14, 2026
Gemini App Tasks: What It Can Actually Do
A toggle in the Gemini app switches from chat to tasks, and the assistant keeps working when you close the laptop. For a small business, the question is which jobs are safe to hand over.
Michaela Zello

Oct 14, 2026
What Gemini 4's Safety Report Tells Builders
Model safety reports are written for regulators, but this one contains four mechanisms a small team can copy, and one number that should change how you screen agent input.
Cecilia Iona

Oct 14, 2026
Gemini 4 Argon Pricing Doubles After Nov 30
Argon's introductory pricing ends on 30 November 2026, and the post-introductory rate is exactly double. The cache discount is larger than the increase, if your prompts are shaped for it.
Manuele Estivo

Oct 14, 2026
Gemini 4 Argon Is Now Available: Who Gets It
Argon spent its first two weeks locked to cyber defenders. On 13 October Google opened it to paid API customers, AI Ultra and AI Pro subscribers, and AI Mode in Search.
Cecilia Iona

Oct 14, 2026
AI Coding Agent Leaves Debug Code in Production
When an AI coding agent leaves debug code in production, some is noise and some is dangerous. A ranked list, sweep commands and lint rules to stop it.
Carlo Zuercher

Oct 14, 2026
Pre-Sell an AI-Built Product Before You Build It
How to pre-sell an AI-built product: what to promise, where to set a kill threshold, and the FTC delivery and refund rules to know before taking any money.
Manuele Estivo

Oct 14, 2026
Prompt AI to Role-Play a Tough Conversation
Prompt AI to role-play a tough conversation first. A two-phase prompt with a difficulty dial, a no-coaching rule and a debrief that finds gaps.
Steve Jefferson

Oct 13, 2026
What Is Greedy Decoding in AI? A Worked Example
What is greedy decoding in AI? It takes the likeliest next token each step. A worked example, why it repeats itself, and how it differs from sampling.
Cecilia Iona

Oct 13, 2026
AI Coding Agent Windows Path Bugs: 6 Fixes
AI coding agent Windows path bugs come from code written for Linux. Six patterns to search for, the fix for each, and one prompt that audits a whole repo.
Carlo Zuercher

Oct 13, 2026
AI Coding Agent Keeps Reverting Your Changes? Fix It
Your AI coding agent keeps reverting your changes? Five causes, a symptom table, and a git recovery recipe that gets your edits back in under a minute.
Carlo Zuercher

Oct 13, 2026
Unsubscribe Link for an AI-Built App: Do It Right
Add an unsubscribe link for an AI-built app that works: one-click headers, a suppression table, CAN-SPAM basics, and a prompt your coding agent can follow.
Steve Jefferson

Oct 13, 2026
Write a Gift Card Policy With AI: What to Include
Write a gift card policy with AI without inventing legal terms. Fields to fill in yourself, the US federal rules to check, and a prompt that stays honest.
Steve Jefferson

Oct 13, 2026
Out of Office Message With AI: Prompt and 4 Templates
Write an out of office message with AI in two minutes. A fill-in prompt, four tested templates, and what to leave out so your auto-reply stays safe.
Steve Jefferson

Oct 13, 2026
Family Code Word for AI Voice Scams: How to Set One
A family code word for AI voice scams gives you a free, ten-second check when a cloned voice calls. How to pick one, share it, and use it under pressure.
Cecilia Iona

Oct 13, 2026
How to Explain an AI News Story to Your Team
How to explain an AI news story to your team: a three-question filter that kills most stories and a five-line briefing format for the ones that pass.
Cecilia Iona

Oct 13, 2026
What Is Max Tokens in AI, and What Does It Cap?
What is max tokens in AI? It caps output only, not input, and the stop reason tells you when a response was truncated rather than simply finished.
Cecilia Iona

Oct 13, 2026
How to Add a Trial Period to an AI-Built App
How to add a trial period to an AI-built app, including the four expiry edge cases agent-written billing code misses and the settings that fix them.
Steve Jefferson

Oct 13, 2026
How to Find Out What People Will Pay for an AI Tool
How to find out what people will pay for an AI tool: four methods ranked by how much they bind the buyer, from surveys up to a real checkout.
Manuele Estivo

Oct 13, 2026
Let a Non-Developer Use an AI Coding Agent Safely
How to let a non-developer use an AI coding agent safely: plan mode by default, deny rules that hold in every mode, and a branch they cannot break.
Steve Jefferson

Oct 13, 2026
Build a Test Suite an AI Coding Agent Can Run
Set up a test suite an AI coding agent can run: one command, an honest exit code, failures that name a file, and why a hook beats an instruction.
Steve Jefferson

Oct 13, 2026
How to Spot a Fake AI App in an App Store
How to spot a fake AI app in an app store by reading the listing against Apple and Google's own published rules on copycats and subscriptions.
Cecilia Iona

Oct 13, 2026
Redact a Screenshot Before You Send It to AI
How to redact a screenshot before sending it to AI: why black boxes and crops can fail, and the re-capture habit that removes the data instead.
Cecilia Iona

Oct 13, 2026
Why AI Demos Look Better Than the Real Product
Why AI demos look better than the shipped product: six production choices that create the gap, and five questions to ask before you build on one.
Cecilia Iona

Oct 13, 2026
Did an AI Model Update Break Your App? How to Tell
How to tell if an AI model update broke your app: check lifecycle status and deprecated parameters before you rewrite a single line of your prompt.
Carlo Zuercher

Oct 13, 2026
How to Tell If You Are Chatting With an AI or a Person
How to tell if you are chatting with an AI: ask directly, test a handoff, verify through a second channel, and skip trick questions that no longer work.
Cecilia Iona

Oct 13, 2026
Sell a Digital Template You Made With AI: What to Check
Sell a digital template you made with AI: choose a narrow job, test it on three real cases, add a license, and check ownership before you list it.
Manuele Estivo

Oct 13, 2026
Prompt AI for a Useful Pros and Cons List, Not a Generic One
Prompt AI for a useful pros and cons list by giving it your situation, weighted criteria, and a ban on generic points. Includes a before and after example.
Steve Jefferson

Oct 12, 2026
Keep a Customer Complaint Log With AI, Step by Step
Keep a customer complaint log with AI: a seven-column template, a monthly prompt that groups causes with counts, and checks that stop made-up patterns.
Cecilia Iona

Oct 12, 2026
Rubber Duck Debugging With AI: A Practical Method
Rubber duck debugging with AI works when you make the model ask questions instead of fixing code. Here is the prompt, a worked bug, and what to avoid.
Steve Jefferson

Oct 12, 2026
Draft a Lease Renewal Letter With AI, Deadline First
Draft a lease renewal letter with AI: find your notice window first, give the model only facts you checked, and ask for terms in writing. Prompt included.
Cecilia Iona

Oct 12, 2026
How to Add Form Validation to an AI-Built App
How to add form validation to an AI-built app: browser checks for fast feedback, server checks for safety, and one shared schema so the two never drift.
Steve Jefferson

Oct 12, 2026
What Is Logit Bias in AI? How to Ban or Boost Words
What is logit bias in AI? Learn how a number added to a token's score bans or boosts specific words, why spacing breaks it, and when a prompt works better.
Carlo Zuercher

Oct 12, 2026
How to Fact-Check an AI News Claim Before You Share It
How to fact-check an AI news claim in ten minutes: trace it to a primary source, check the date, test the numbers, and know when to wait before sharing.
Cecilia Iona

Oct 12, 2026
What Is a Frequency Penalty in AI? A Worked Example
What is a frequency penalty in AI? See how it lowers the score of repeated words, how it differs from presence penalty, and which value to try first.
Cecilia Iona

Oct 12, 2026
Who Tests an AI Model Before It Is Released?
Four groups test a frontier model before it ships, and none of them checks whether it is any good at your work. Knowing what was covered tells you what is left for you.
Cecilia Iona

Oct 12, 2026
How to Cap AI Usage Per Customer in Your App
The vendor dashboard caps your whole organisation and the billing meter is too slow to stop anything. The per-customer gate has to be yours, and what it does when it trips is a product decision.
Carlo Zuercher

Oct 12, 2026
Why an AI Model Disappears From the Model Picker
It was there last week and now it is not. Retired, tier-gated, consolidated away or simply renamed: four causes, and the checks that tell them apart in a few minutes.
Michaela Zello

Oct 12, 2026
How to Make a Price List With AI
AI cannot know your costs and will invent them with total confidence. Bring your own floor prices, let it do the packaging, and watch for the three lines that leak money.
Manuele Estivo

Oct 12, 2026
You Pasted an API Key Into an AI Chat. Now What?
This is a rotation problem, not a privacy one, and the published zero-downtime rotation procedure is the wrong one to follow. Here is the order a leak actually needs.
Cecilia Iona

Oct 12, 2026
Why Does AI Write Everything in Bullet Points?
Bullets are the house style of chat assistants, and they quietly remove the part of an explanation that connects one claim to the next. Here is the instruction that gets prose back.
Cecilia Iona

Oct 12, 2026
Use AI to Chase a Supplier for a Late Delivery
Chasing a supplier is the inverse of chasing an invoice: you have no leverage you can use without hurting yourself. Escalate specificity instead, in three stages.
Steve Jefferson

Oct 12, 2026
Two AI Coding Agents on the Same Repo
Two agents in one checkout do not conflict politely like two developers do. Isolate them with a worktree each, then merge in the order that keeps your lockfile honest.
Steve Jefferson

Oct 12, 2026
Why a New AI Model Is Slow on Launch Day
The model benchmarked beautifully and your app feels worse. Three documented causes, none of them the model's own speed, and the signal that tells you which one you have.
Carlo Zuercher

Oct 12, 2026
Why AI Gives Short Answers When You Want Detail
Four different failures produce the same one-paragraph reply. One of them is your own output limit, and it takes ten seconds to rule out before you rewrite a single line of the prompt.
Cecilia Iona

Oct 12, 2026
Strip Photo Metadata Before Uploading to AI Tools
A photo can carry the coordinates of where it was taken. Before you upload pictures of a client site, a home office or a product, check what travels with the file.
Cecilia Iona

Oct 12, 2026
Hand Over an AI-Built App to a Client Without Regrets
The app works, the client is happy, and then you realise the domain, the database and the API keys are all in your name. A handover checklist for freelancers and small agencies.
Manuele Estivo

Oct 11, 2026
Write a FAQ Page With AI That Answers Real Questions
A FAQ page is only as good as its questions. Have AI write the answers, but source the questions from your inbox, and skip the FAQ markup myth: Google stopped showing FAQ rich results in 2026.
Cecilia Iona

Oct 11, 2026
Prompt AI to Write a Product Update Announcement
The usual AI product update reads like a press release with the facts buried. Start from a plain changelog and let the announcement be a short, honest layer on top.
Michaela Zello

Oct 11, 2026
Prompt AI to Write Headline Options You Can Use
Ask an AI for ten headlines and you get ten versions of the same idea. The fix is to hand it a reader, one promise and a list of patterns to avoid, then score what comes back.
Steve Jefferson

Oct 11, 2026
Add a Sitemap and robots.txt to an AI-Built App
A sitemap lists the pages you want found and robots.txt tells crawlers where not to go. AI builders often ship neither, or ship a robots.txt that blocks everything.
Manuele Estivo

Oct 11, 2026
Add a Custom 404 Page to an AI-Built App Properly
A friendly 404 page that returns a 200 status is worse than the default error. Here is how to build one that helps visitors and tells search engines the truth.
Steve Jefferson

Oct 11, 2026
Add a Contact Form to an AI-Built App That Works
An AI tool will draw a contact form in one prompt. What it often leaves out is everything that happens after Send: validation, spam, storage and a confirmation that the message arrived.
Steve Jefferson

Oct 11, 2026
Prompt AI to Write a Polite No That Still Lands
Most AI-written refusals fail in the same way: so much cushioning that the reader comes back with a counter-offer. The fix is deciding what you owe the other side before you ask for a draft.
Steve Jefferson

Oct 11, 2026
When an AI Coding Agent Cannot Run Your Build
An agent that cannot compile your project does not stop. It keeps editing, reasoning about code it has never seen run, and the result is a confident diff nobody has tested.
Carlo Zuercher

Oct 11, 2026
Should You Let an AI Agent Sign Up for Services?
An agent that can register for a trial looks like a time saver until you need to cancel, recover or hand over the account. The rules here are clearer than most people expect.
Cecilia Iona

Oct 11, 2026
AI Project Came In Under Budget: What Now
You quoted three weeks. The build took four days and it is good. Keeping the full fee feels wrong, refunding it feels worse, and the contract already decided most of this.
Manuele Estivo

Oct 11, 2026
Prompt AI to Find Contradictions in a Document
A plain request to find contradictions returns a confident list, half of it paraphrase mistaken for conflict. Turning it into an extraction task fixes both the false pairs and the misses.
Steve Jefferson

Oct 11, 2026
Is It Safe to Let an AI Agent Run Shell Commands?
The question people ask is whether the agent might delete something. The more useful question is what a command it runs can read, and the honest answer in most default setups is almost everything.
Cecilia Iona

Oct 11, 2026
AI Model Not Available in Your Country? Why, and What Next
A launch post goes up, the model is everywhere on your timeline, and the sign-up page tells you it is not available where you live. There are four different reasons that happens, and only some of them have a fix.
Cecilia Iona

Oct 11, 2026
Prompt AI to Proofread Without Rewriting Your Text
Ask for a proofread and you get back a document that is cleaner, blander and no longer sounds like you. The fix is not a politer request. It is changing what you ask the model to return.
Steve Jefferson

Oct 11, 2026
Which AI Model Does a Tool Use? How to Find Out
Plenty of AI products describe their engine as advanced AI and leave it there. If you are paying for it, or building on it, the actual model id is knowable most of the time. Here is the order to look in.
Carlo Zuercher

Oct 11, 2026
Why an AI Model Does Not Know Which Model It Is
Ask a chatbot what model it is and you will get an answer delivered with total confidence. It is often wrong, and the reason is worth understanding, because it tells you where the real answer lives.
Cecilia Iona

Oct 11, 2026
Add Discount Codes to an AI-Built App Safely
Add discount codes to an AI-built app without losing margin: coupon vs promotion code, five limits to set, and a stacking trap shown with real numbers.
Steve Jefferson

Oct 11, 2026
WordPress vs an AI-Built Site: Who Maintains What
WordPress vs an AI-built site, compared by the work you do a year later: hosting, updates, security, plugins and support, plus which suits which site.
Carlo Zuercher

Oct 11, 2026
Write a Refund Policy With AI Without Inventing Rights
How to write a refund policy with AI: seven decisions to make first, the EU 14-day rules that can override your wording, and a check for invented promises.
Manuele Estivo

Oct 11, 2026
XSS in AI-Generated Code: Three Patterns to Fix
XSS in AI-generated code usually hides in three places: rendered markdown, chat output and user links. See each risky pattern and the safe fix to ship.
Carlo Zuercher

Oct 11, 2026
Open Graph Link Previews for an AI-Built App
Open Graph link previews for an AI-built app: why shared links show a blank card, the tags that fix it, and a curl check that proves they work.
Steve Jefferson

Oct 10, 2026
AI Coding Agent Retry Logic: Review It Before It Ships
AI coding agent retry logic often hammers a failing API with instant repeats. Use this five-point review and a corrected TypeScript example to fix it.
Carlo Zuercher

Oct 10, 2026
AI Coding Agent Gitignore: Five Mistakes to Catch
Most AI coding agent gitignore problems are five repeat mistakes. Learn to spot each one, the git commands that expose it, and the safe fix for each case.
Carlo Zuercher

Oct 10, 2026
Prompt AI to Write Error Messages Users Can Act On
How to prompt AI to write error messages that say what went wrong and what to do next, with a rewrite table and a rule that stops it inventing causes.
Steve Jefferson

Oct 10, 2026
Prompt AI to Write a Welcome Email People Actually Read
How to prompt AI to write a welcome email: the four facts to supply, a brief that keeps it to one action, and the sender rules that make it legal to send.
Steve Jefferson

Oct 10, 2026
Prompt AI to Write a Spreadsheet Formula That Works
How to prompt AI to write a spreadsheet formula you can trust: what to give it, a five-row test sheet that catches wrong answers, and common failures.
Steve Jefferson

Oct 10, 2026
How to Read an AI Threat Intelligence Report
An AI threat intelligence report is not a base rate. Read it in this order to find the few parts that should change what you build this week.
Cecilia Iona

Oct 10, 2026
How to Price a One-Off AI Prototype
To price a one-off AI prototype, start from the decision it settles, not the build hours. Three pricing shapes compared on one brief, with the clauses.
Manuele Estivo

Oct 10, 2026
AI Coding Agent vs a Flaky Dev Server
A stale AI coding agent dev server makes the agent fix working code. Spot the four symptoms and add the health check that stops the whole class of bug.
Steve Jefferson
Oct 10, 2026
Why AI Model Release Trackers Disagree
AI model release trackers contradict each other more often than you would expect. Here is why, and a two-minute check that settles what actually shipped.
Cecilia Iona

Oct 10, 2026
Add a Free Trial That Expires to an AI App
A free trial that expires is mostly edge cases. Here is the data model, the three bugs that break trials in production, and the prompts to build it.
Steve Jefferson

Oct 10, 2026
AI Model Preview: What the Label Promises
An AI model preview label is not a lifecycle stage. Compare the documented notice periods and find out what you are actually promised before you ship.
Cecilia Iona

Oct 10, 2026
AI Coding Agent Hit Its Limit Mid-Task: What Now
An AI coding agent usage limit stops work in three different ways. Tell them apart from the error, then recover the half-finished branch safely.
Steve Jefferson

Oct 10, 2026
When a Client Says AI Could Do This Cheaper
When a client says AI could do this cheaper, they mean one of four things. Diagnose which one before you defend your price, then move the talk to risk.
Manuele Estivo

Oct 10, 2026
Anthropic 2026 Usage Policy: What Changes
The Anthropic 2026 usage policy takes effect on 12 November. Two sections change what builders must ship; the rest only clarify existing rules.
Cecilia Iona

Oct 10, 2026
Why Your AI Coding Agent Copies Outdated Code
An AI coding agent copies outdated code because your repo outvotes your intentions. Here is the audit that proves it and the three fixes that work.
Carlo Zuercher

Oct 10, 2026
Build an Event RSVP Page With AI
Build an event RSVP page with AI: the data model, three failure cases to specify first, and a prompt sequence that gets a working page in one sitting.
Steve Jefferson

Oct 10, 2026
Move a Squarespace Site to an AI-Built Site
Move a Squarespace site to an AI-built site: what Squarespace's export really contains, what you must rebuild, and a redirect map that protects rankings.
Manuele Estivo

Oct 10, 2026
How to Delete Your AI Chat History (And What Stays)
How to delete your AI chat history, and what deletion really does. Two providers' retention windows side by side, plus the copies people miss.
Cecilia Iona

Oct 10, 2026
Shopify vs an AI-Built Online Store: The Real Costs
Shopify vs an AI-built online store: published plan prices, card fee maths at three sales levels, and the checkout jobs you take on if you build your own.
Carlo Zuercher

Oct 10, 2026
Zapier vs Building Your Own Automation With AI
Zapier vs building your own automation with AI: published task pricing, the maintenance hour people forget, and a payback table showing when building wins.
Carlo Zuercher

Oct 9, 2026
Can You Own an AI-Generated Logo?
Can you own an AI-generated logo? Copyright and trademark are separate questions with different answers. What the US sources say and how to protect it.
Cecilia Iona
Oct 9, 2026
Build a Time Tracker for Freelancers With AI
Build a time tracker for freelancers with AI: store start times, not counters, pick a rounding rule with real numbers, and export clean invoice lines.
Steve Jefferson

Oct 9, 2026
Build a QR Code Menu for a Restaurant With AI
Build a QR code menu for a restaurant with AI: the one URL decision that saves reprinting, a simple data model, allergen rules and a print test.
Steve Jefferson

Oct 9, 2026
Use Google Sheets as an App Backend: Limits and Fixes
Can you use Google Sheets as an app backend? The published quota numbers, a 30-second cache that fixes most failures, and the signs it is time to move on.
Carlo Zuercher

Oct 9, 2026
Turn a Figma Design Into an App With AI
How to turn a Figma design into an app with AI: what to hand the model, what a frame cannot tell it, and a prompt order that avoids a pretty but dead UI.
Steve Jefferson

Oct 9, 2026
When an AI-Built App Outgrows the Free Tier
An AI-built app outgrows the free tier in a predictable order. Which limit bites first, what each failure looks like, and when upgrading is worth it.
Carlo Zuercher

Oct 9, 2026
How to Write an AI Disclosure for Your Website
An AI disclosure for your website needs different wording for chatbots, generated copy and customer data. Three drafted versions and where each goes.
Cecilia Iona

Oct 9, 2026
When AI Says Wrong Things About Your Business
If AI says wrong things about your business, the fix is upstream. Trace the claim to the source that fed it, correct that, and the assistants follow.
Cecilia Iona

Oct 9, 2026
Contract Clauses for an AI-Built App
Six contract clauses for an AI-built app that standard software templates miss, with drafted wording for code ownership, API costs and handover.
Manuele Estivo

Oct 9, 2026
AI-Built App Passes Tests but Breaks for Users
When an AI-built app passes tests but breaks for users, the suite is testing the code's assumptions. Four gaps AI test generation leaves, and the fix.
Carlo Zuercher

Oct 9, 2026
Data Subject Access Requests When You Use AI
A data subject access request and AI tools do not mix neatly. Here are the copies of personal data most businesses miss, and the one-month clock.
Cecilia Iona

Oct 9, 2026
How to Choose Opening Hours With AI
Choose opening hours with AI using your own till data, not guesswork. A contribution margin per hour method that shows which hours actually pay.
Steve Jefferson

Oct 9, 2026
When a Client Finds Out You Used AI
A client found out you used AI on their project. Here are the three objections hiding behind that reaction and how to answer each one properly.
Manuele Estivo

Oct 9, 2026
AI-Built App Database Connection Limit: Fix It
Hitting an AI-built app database connection limit? Here is why serverless code exhausts Postgres connections and the pooler fix that stops it.
Steve Jefferson

Oct 9, 2026
How to Refund a Client for a Failed AI Build
How to refund a client for a failed AI build without overpaying: value the delivered artifacts, not the hours, and close it with a written settlement.
Manuele Estivo

Oct 9, 2026
Build a Quote Calculator With AI: A Worked Example
A quote calculator is mostly a pricing rulebook with a form on top. Keep the rules as data and the maths in one tested function, and the AI has little room to get it wrong.
Steve Jefferson

Oct 9, 2026
Set Up a GitHub Actions Pipeline With AI, Safely
AI writes a working workflow file in seconds. The problem is that working and safe are different tests, and the generated YAML tends to fail the second one.
Steve Jefferson

Oct 9, 2026
What Is a GGUF File? How Local AI Models Are Packaged
Download any local model and you meet a filename like Q4_K_M.gguf. Here is what the format holds, how to read the name, and how to predict the size before you download.
Carlo Zuercher

Oct 9, 2026
How to Respond to an AI Data Leak in the First 72 Hours
Someone pasted customer data into a chatbot, or an agent printed an API key into a log. The first hours decide how bad it gets, and the order of steps matters.
Cecilia Iona

Oct 8, 2026
What Is Rotary Position Embedding (RoPE) in AI?
Rotary position embedding tells a model where each token sits by rotating its vectors. A six-line calculation shows why that makes attention depend only on distance.
Cecilia Iona

Oct 8, 2026
Fix TypeScript Errors With AI Without Casting to any
Pasting 200 compiler errors into a chat gets you a green build and a worse codebase. Triage by error code first, and give the AI rules it cannot cheat around.
Steve Jefferson

Oct 8, 2026
What Is Grouped Query Attention? Cheaper AI Memory
Grouped query attention is a small change to the attention layer that cuts the memory a long conversation needs. The arithmetic shows how much.
Cecilia Iona

Oct 8, 2026
How to Invoice a Client for AI-Assisted Work
When AI cuts a 20-hour job to 6, an hourly invoice quietly punishes you. Here are three invoice structures, the line items for each, and what to say about the AI.
Manuele Estivo

Oct 8, 2026
Convert CSS to Tailwind With AI Without Breaking Layout
The class names are the easy part. What breaks a CSS to Tailwind conversion is the reset Tailwind applies underneath, and here is how to hand that to AI properly.
Steve Jefferson

Oct 8, 2026
What Is Byte Pair Encoding? How AI Splits Text
Byte pair encoding is how most language models decide where one token ends and the next begins. Here it is, merge by merge, on a four-word corpus.
Cecilia Iona

Oct 8, 2026
When Two AI Coding Agents Disagree, Do This
Disagreement is information, and it usually tells you something about your brief rather than your code. A triage order that beats running a third agent.
Carlo Zuercher

Oct 8, 2026
How to Add a Calendar View to an AI-Built App
The grid is the easy part. Recurrence, time zones and all-day events break later, and they break because of how the records were stored.
Steve Jefferson

Oct 8, 2026
How to Price a Rush Job for an AI Build
Price what the rush displaces and what it removes, not the deadline itself. A multiplier table based on how much of the normal timeline actually disappears.
Manuele Estivo

Oct 8, 2026
What Is Adaptive Thinking in AI Models?
The model decides how much to think instead of you setting a dial. That moves cost from predictable to variable, which changes how you budget for it.
Cecilia Iona

Oct 8, 2026
How to Add a Rich Text Editor to an AI App
The storage format decides whether you can change editors later, whether search works, and whether you just opened a stored cross-site scripting hole.
Steve Jefferson

Oct 8, 2026
How to Mark Up AI API Costs You Resell
A flat multiple on token cost looks safe and quietly loses money, because the top one percent of users cost far more than the blended average suggests.
Manuele Estivo

Oct 8, 2026
What Is a Subagent in AI? A Practical Guide
A subagent turns a large amount of input into a small amount of output in its own context. The cost arithmetic is what makes the pattern worth the complexity.
Cecilia Iona

Oct 8, 2026
How Long Does a Google Spam Update Take?
Most finish in under three days. The September 2026 update has run fifteen days and passed its own estimate, making it the first of the year to over-run.
Manuele Estivo

Oct 8, 2026
ChatGPT Intelligent UI: What Builders Should Know
OpenAI rolled GPT-6 out to every ChatGPT tier with answers that render as working interfaces. The component library pattern behind it is one you can copy.
Cecilia Iona

Oct 8, 2026
Claude Haiku 5.5: Cheapest Claude Model Yet
Anthropic's new small model lands at $0.10 per million input tokens, with a five times step up above 100,000 tokens that most coverage left out.
Cecilia Iona

Oct 8, 2026
How to Prompt AI to Write a README People Will Read
Prompt AI to write a README from your real repo facts, then verify every command in a clean clone. Includes the prompt, the structure, and a review pass.
Steve Jefferson

Oct 8, 2026
How to Prompt AI to Write a Cron Expression That Works
Get a correct cron expression from AI every time: the five-field rules, the day-of-month trap, timezone checks, and a prompt that makes it show its work.
Steve Jefferson

Oct 8, 2026
What Is an Agent Loop in AI? Explained in Code
An agent loop is the cycle of think, act, observe that turns a language model into an agent. See it in 25 lines of code, with its stop rules and costs.
Carlo Zuercher

Oct 8, 2026
How to Prompt AI to Critique Your Writing Honestly
Stop getting flattery. A prompt pattern that makes AI critique your writing with quoted evidence, ranked problems, and no rewrite until you ask for one.
Cecilia Iona

Oct 8, 2026
How to Run an AI Model Locally With Ollama
Install Ollama, pull a model, chat in the terminal, and call it from code over the local API. Includes sizing advice and the mistakes that waste time.
Steve Jefferson

Oct 7, 2026
What Is a Needle in a Haystack Test for AI Models?
What the needle in a haystack test measures, why a perfect score proves less than it seems, and how to run your own version in under 40 lines of Python.
Cecilia Iona

Oct 7, 2026
How to Prompt AI to Write a Dockerfile That Builds Fast
A prompt that gets AI to write a lean, cache-friendly Dockerfile: multi-stage builds, pinned tags, a non-root user, and a checklist to review the result.
Steve Jefferson

Oct 7, 2026
How to Turn a Podcast Into a Newsletter With AI
Turn one podcast episode into a newsletter with AI: transcript, three angle pass, a quote audit that stops invented quotes, and a subject line test.
Manuele Estivo

Oct 7, 2026
What Is Sycophancy in AI and How to Test for It
Sycophancy is when an AI tells you what you want to hear. Why it happens, what the research found, and three quick tests you can run on any model today.
Cecilia Iona

Oct 7, 2026
How to Prompt AI to Write a Bash Script You Can Trust
A step-by-step prompt for getting a safe, readable bash script from AI: the safety header, input rules, dry runs, and how to check it with ShellCheck.
Steve Jefferson

Oct 7, 2026
Same AI Model, Different Providers, Different Results
The same AI model can behave differently on two providers. The five causes, from quantization to injected system prompts, and how to test for each.
Carlo Zuercher

Oct 7, 2026
How to Write a Services Page for an AI Agency
How to write a services page for an AI agency: name the outcome, show the boundary, and answer the four questions a buyer has before they email.
Manuele Estivo

Oct 7, 2026
How to Prompt AI to Use Your Own Terminology
How to prompt AI to use your own terminology: a glossary block with banned alternatives, placed where the model will still see it at the end.
Steve Jefferson

Oct 7, 2026
Turn a Long Chat Into a Reusable Prompt
How to turn a chat into a reusable prompt: keep the corrections that actually fixed the output, drop the conversation, and test on a fresh input.
Steve Jefferson

Oct 7, 2026
Should You Prompt AI in English or Your Language?
Should you prompt AI in English or your own language? The answer depends on the task, not the model. A decision table and the mixed approach.
Cecilia Iona

Oct 7, 2026
How to Use AI to Prioritize a Product Backlog
How to use AI to prioritize a product backlog without handing over the decision: a scoring prompt, the inputs it needs from you, and where it fails.
Steve Jefferson

Oct 7, 2026
How Many Examples Does a Prompt Need?
How many examples does a prompt need? Usually three to five, and which ones you pick matters more than how many. A method for finding your number.
Steve Jefferson

Oct 7, 2026
How to Get AI to Return a Clean Markdown Table
How to get AI to return a markdown table that survives the paste: fix ragged columns, swallowed pipes, and tables that arrive as prose instead.
Steve Jefferson

Oct 7, 2026
OpenAI Text Watermarking Arrives in the EU
OpenAI will watermark ChatGPT and Codex text in the EU with textGrain. What it detects, where it breaks, and what changes if you publish AI writing.
Cecilia Iona

Oct 7, 2026
Mistral Large 4: What the Preview Gives You
Mistral Large 4 is in public preview: 1T parameters, 49B active, $1.36 per million in. What the preview gives you now and what waits for weights.
Cecilia Iona

Oct 7, 2026
Plan a Product Photo Shoot With AI: A Shot List
A good shoot is decided before the camera comes out. Turn a product list into a shot list with AI, and avoid the reshoots that eat a weekend.
Cecilia Iona

Oct 7, 2026
Plan a Pop-Up Shop With AI: Stock, Staff, and Setup
A pop-up lasts a weekend and fails on small details. Use AI to build the countdown, stock plan, and packing list, then verify permits and numbers yourself.
Manuele Estivo

Oct 7, 2026
Schedule Social Posts for a Restaurant With AI
A restaurant has more to post about than it thinks. A weekly rhythm built from the kitchen, captions drafted from your notes, and the facts AI must never invent.
Manuele Estivo

Oct 7, 2026
Plan a Fitness Class Schedule With AI and Real Data
Your attendance sheet already knows which classes to keep, move, or merge. How to have AI read it, and why you should change one slot at a time.
Steve Jefferson

Oct 7, 2026
Write a Creative Brief for a Designer With AI
A vague brief costs you revision rounds. Let AI interview you for the answers a designer needs, then tighten the result into a one-page brief.
Manuele Estivo

Oct 6, 2026
Choose a Domain Name With AI Without Picking a Dud
AI is great at generating name ideas and bad at knowing what is taken. A method to get 50 options, cut them to three, and verify the winner yourself.
Steve Jefferson

Oct 6, 2026
Price a One-Page Website With AI: A Freelancer's Method
AI shrinks the build time of a one-page site but not the decisions, content, and revisions. A scope-based pricing method that keeps your margin.
Manuele Estivo

Oct 6, 2026
What Is Overfitting in AI? A Plain-English Guide
Overfitting means a model has memorized its practice material instead of learning the pattern. Here is how it shows up, why it matters, and what fixes it.
Cecilia Iona

Oct 6, 2026
Learn a New Programming Language With AI, Step by Step
The fastest way to learn a language with AI is to stop letting it write your code. A four-week plan built on predict, run, and explain loops.
Steve Jefferson

Oct 6, 2026
Plan Volunteer Shifts for a Nonprofit Event With AI
Volunteer shifts fail on the day because of gaps nobody spotted on paper. Here is how to have AI draft the rota, then check it the way a coordinator would.
Steve Jefferson

Oct 6, 2026
Technical Debt in an AI-Built App: What It Looks Like
Technical debt in an AI-built app is rarely messy code. It is five repeating patterns, each easy to spot once you know them. Here is a 30-minute audit to find yours.
Carlo Zuercher

Oct 6, 2026
What Is a 500 Error? Fixing One in an AI-Built App
A 500 error means the server hit a problem it could not handle, and the visitor did nothing wrong. The page tells you almost nothing, so the fix starts in the logs.
Cecilia Iona

Oct 6, 2026
Why Your AI-Built App Works Locally but Fails Live
It ran perfectly on your laptop, then broke the moment it went live. The cause is almost always one of seven differences between the two machines. Check them in this order.
Steve Jefferson

Oct 6, 2026
What Is a Lockfile and Why AI Agents Change It
A lockfile records the exact version of every package your app uses. Here is what that means, why AI agents keep touching it, and how to review the change.
Cecilia Iona

Oct 5, 2026
How to Check AI Math Before You Trust It
A language model writes numbers the way it writes words, so a confident figure is not a checked one. Four quick tests to catch wrong arithmetic before it reaches a client.
Carlo Zuercher

Oct 5, 2026
Write a Client Onboarding Checklist With AI
Generic onboarding checklists fix problems you do not have. Feed AI your last three client email threads and build a checklist around where things really stalled.
Manuele Estivo

Oct 5, 2026
How to Reconcile a Bank Statement With AI
Let AI propose which bank lines match your books, then prove the result with a plain spreadsheet check. A three-pass method with a prompt and a redaction checklist.
Steve Jefferson

Oct 5, 2026
Pricing Freelance Work When AI Makes You Faster
If AI cuts a ten-hour job to four, billing by the hour cuts your income by sixty percent. Here is the arithmetic and four ways to price the work instead.
Manuele Estivo

Oct 5, 2026
How to Organize a Messy Shared Drive With AI
You can clean up a shared drive with AI without giving it access to a single file. Export the file list, let the model propose a structure, review it, then move.
Steve Jefferson

Oct 5, 2026
How to Use AI to Plan Stock for a Busy Week
Use AI to plan stock for a busy week by giving it your sales history, not your guesses. The exact numbers to supply and the prompt that works.
Steve Jefferson

Oct 5, 2026
How to Forecast Your AI Bill for Next Month
To forecast your AI bill, you need three numbers and one honest multiplier. Here is the arithmetic, and the four things that break the estimate.
Manuele Estivo

Oct 5, 2026
How to Avoid Lock-In to One AI Provider
To avoid lock-in to one AI provider, fix the four things that actually bind you. Most teams insure against the cheapest one and miss the rest.
Carlo Zuercher

Oct 5, 2026
Does Your App Need a Queue? Three Symptoms
Does your app need a queue? Three symptoms say yes, and most apps show none of them. Here is how to tell, and the cheapest fix short of a queue.
Steve Jefferson

Oct 5, 2026
What Is Agent Observability? The Four Things to Log
Agent observability means answering what an agent did and why, after the fact. Here are the four things to record, and the one most teams leave out.
Cecilia Iona

Oct 5, 2026
Why Your AI Agent Did Nothing: A Diagnostic Order
Your AI agent did nothing and reported success. Five causes, in the order worth checking, and the one-line test that separates each from the next.
Steve Jefferson

Oct 5, 2026
When AI Keeps Rewriting Working Code: Four Causes
When AI keeps rewriting working code, the cause is usually your context, not stubbornness. Four causes ranked by frequency, each with its own fix.
Steve Jefferson

Oct 5, 2026
What Is Off-Peak AI API Pricing, and Who Can Use It
Off-peak AI API pricing halves token costs during quiet hours. Here is which workloads can actually move, and the arithmetic on what it saves.
Manuele Estivo

Oct 5, 2026
Why a Smaller AI Model Beats a Bigger One
A smaller AI model can outscore a bigger one from the same lab. Three mechanisms explain why, and one of them should change how you pick a model.
Cecilia Iona

Oct 5, 2026
What Is a Causal Encoder-Decoder Model?
A causal encoder-decoder model splits reading from writing. Here is what that changes, using the DeepSeek V4.1-Flash release as the worked example.
Cecilia Iona

Oct 5, 2026
What Is Data Residency? A Builder's Explanation
Data residency means where your data is stored and processed. Here is what it covers, what it does not, and how to answer the question honestly.
Cecilia Iona

Oct 5, 2026
DigitalOcean Agent Droplets: The Flat-Price Math
DigitalOcean Agent Droplets bundle agent hosting into one monthly price. Here is the break-even math against metered billing, and when it loses.
Manuele Estivo

Oct 5, 2026
Claude Inference in India Arrives on Amazon Bedrock
Anthropic turned on Claude inference in India through Amazon Bedrock on 5 October. What in-country processing changes, and what it leaves alone.
Cecilia Iona

Oct 5, 2026
How to Patch a Self-Hosted Tool You Run
A weekly routine for patching self-hosted tools: list what you run, watch release notices, back up, update and verify, then check for tampering.
Steve Jefferson

Oct 5, 2026
How to Report a Security Bug You Found With AI
Reproduce the bug yourself, use the project's official security channel, and write a report a stranger can verify in ten minutes. Includes a template.
Steve Jefferson

Oct 5, 2026
How Much VRAM Do You Need to Run a Local LLM?
Estimate the VRAM a local LLM needs with one formula, a size-by-quantisation table and a rule of thumb for your GPU, including mixture-of-experts models.
Carlo Zuercher

Oct 4, 2026
What Is a CVE? A Plain Guide for App Builders
A CVE is a public ID for a known security flaw. Here is how to read one, what the severity score means, and how to tell if it affects your app.
Cecilia Iona

Oct 4, 2026
Math.random vs Crypto Random in AI-Generated Code
AI-written code often uses Math.random for tokens and IDs. Here is why that is unsafe, how to find it with one grep, and what to use instead.
Steve Jefferson

Oct 4, 2026
Strata Runs a 125B Model on a 12GB GPU
Strata, an MIT-licensed inference engine, claims to run a 125B model on a 12GB GPU at 60 to 95 tokens a second. What it needs and what to doubt.
Cecilia Iona

Oct 4, 2026
Google Pauses OSS Bug Bounty Over AI Slop Reports
Google has stopped taking new product bug reports in its open source bug bounty after a flood of invalid AI-written submissions. Here is what is paused.
Cecilia Iona

Oct 4, 2026
Gemini Free Tier Drops to Flash-Lite on October 9
From 9 October, free Gemini users get Flash-Lite only and AI Plus loses Pro. Here is the tier table and what it changes if you prototype with Gemini.
Cecilia Iona

Oct 4, 2026
Rejetto HFS CVE-2026-61500: Patch to 3.2.1 Now
A critical flaw in Rejetto HFS lets an unauthenticated attacker forge an admin session and run code. It affects 3.0.0 to 3.2.0, and 3.2.1 fixes it.
Cecilia Iona

Oct 4, 2026
Responding to a Complaint About Staff, With AI
A complaint about a staff member carries legal weight. Use AI for structure, tone and record-keeping, and never for deciding what happened.
Steve Jefferson

Oct 4, 2026
How to Chase a Quote That Went Quiet, With AI
A quote that went quiet is usually not a no. A three-touch follow-up where each message does a different job, drafted with AI.
Steve Jefferson

Oct 4, 2026
How to Use AI to Prepare a Tender Bid
Most tenders are lost on compliance, not prose. Use AI to build the requirements matrix first, then draft against it.
Steve Jefferson

Oct 4, 2026
How to Spot a Time-Wasting Enquiry With AI
Score every enquiry before you quote it. A simple AI scoring pass on the signals that predict non-conversion, and the rule that keeps it safe.
Steve Jefferson

Oct 4, 2026
Rebuild or Fix Your AI-Built App? How to Decide
Rebuilding an AI-built app feels cheaper than fixing it, and usually is not. Four tests that tell you which one you are actually facing.
Manuele Estivo

Oct 4, 2026
What Is Context Compaction in AI?
Context compaction is when an AI agent summarises its own earlier history to make room in the context window. It is why a long session forgets your rules.
Cecilia Iona

Oct 4, 2026
What Is Database Branching? And When to Use It
Database branching gives each code branch its own copy of the database, so a schema change can be tested without touching real data.
Steve Jefferson

Oct 4, 2026
What Is Agentic Infrastructure?
Agentic infrastructure is the layer of sandboxes, throwaway databases, runtimes and permission systems that AI agents need in order to do work safely.
Cecilia Iona

Oct 4, 2026
Claude Frontier Academy: Anthropic's $100M Bet
Anthropic has committed $100 million to train 10,000 engineers to deploy Claude inside large organisations. The interesting part is who is in the first cohorts.
Cecilia Iona

Oct 4, 2026
Supabase Acquires Turso to Serve AI Agents
Supabase is acquiring Turso and says it now launches over a million databases a week, most of them created by AI tools rather than people.
Cecilia Iona

Oct 4, 2026
How to Clean Messy Spreadsheet Data With AI
Clean messy spreadsheet data with AI without silent changes: a four-step method with copy-ready prompts and checks that catch every bad edit.
Steve Jefferson

Oct 4, 2026
How to Check Mac App Permissions Before Using an AI Agent
A five-minute audit of Mac privacy permissions before you install an AI agent: where to look, which grants are expensive, and how to take access back.
Steve Jefferson

Oct 4, 2026
Do AI Text Detectors Work? What the Evidence Shows
Do AI text detectors work? OpenAI shut down its own, and research shows bias against non-native writers. What the evidence says and what to use instead.
Cecilia Iona

Oct 4, 2026
The Voluntary AI Safety Accord: What It Changes
Six frontier AI labs signed a voluntary White House accord on 29 September. What it commits them to, what it lacks, and what builders should expect.
Cecilia Iona

Oct 4, 2026
California Subpoenas OpenAI Over Agent Cyber Incidents
California's attorney general served OpenAI a subpoena over AI agents that left a test environment and reached Hugging Face. What is known and unknown.
Cecilia Iona

Oct 4, 2026
macOS Full Disk Access Gets Stricter for AI Agents
Apple says it will add explicit-action controls to macOS Full Disk Access because of AI agents. What was announced, what was not, and what to do now.
Cecilia Iona

Oct 4, 2026
GitLab AI Gateway Flaw CVE-2026-90970: What to Patch
GitLab patched a CVSS 9.9 sandbox escape in its self-hosted AI Gateway. Affected versions, fixed versions, who is exposed, and what to check this week.
Cecilia Iona

Oct 3, 2026
Is Your AI Side Project Worth Continuing?
Four questions answered from evidence you already have, plus a decision date set in advance so the call is not made on the mood of a bad Sunday night.
Manuele Estivo

Oct 3, 2026
Does an AI App Need a Free Tier?
A free tier on an AI app costs real money per signup. Work out your ceiling before launch, and the three signals that it is paying for itself.
Manuele Estivo

Oct 3, 2026
Catch Supplier Billing Mistakes With AI
Most supplier overcharges survive because nobody compares the invoice to the quote. Five error types that recur, and a checking prompt for each one.
Steve Jefferson

Oct 3, 2026
Prompt AI to Choose Between Two Options
AI will pick whichever option you described better, not the better option. A swap test and a prompt pattern that forces the criteria out first.
Steve Jefferson

Oct 3, 2026
Anthropic IPO: Mid-November Target Reported
Bloomberg reports Anthropic is targeting a mid-November IPO near a $2 trillion valuation. What is reported, what is confirmed, and what changes for you.
Cecilia Iona

Oct 3, 2026
How to Use AI to Write a Staff Rota
A staff rota is a constraints problem, not a writing problem. The exact constraints to give AI, and the three it will quietly break if you do not.
Steve Jefferson

Oct 3, 2026
AI Agent Accountability Act: What It Covers
Hawley and Murphy introduced a bill extending CFAA liability to AI agent operators and developers. The two standards, and who counts as an operator.
Cecilia Iona

Oct 3, 2026
What Is a Decision Model in AI? Explained
A decision model returns probabilities over answers you define instead of text. How they work, what the 2026 releases cost, and when to use one.
Cecilia Iona

Oct 3, 2026
GitHub Copilot Computer Use Is in Preview
GitHub Copilot can now click, type and drag inside desktop apps. What the preview covers, the two permissions it needs, and when not to use it.
Cecilia Iona

Oct 3, 2026
Pi 1.0: Open Source Coding Agent Goes Stable
Earendil released Pi 1.0, an MIT-licensed terminal coding agent. What the stable release adds, and when a bring-your-own-model agent pays off.
Cecilia Iona

Oct 3, 2026
SQL Injection in AI-Generated Code: How to Find and Fix It
AI coding tools can write queries that glue user input into SQL. Four patterns to search for, with before and after code for each, and a prompt to audit.
Carlo Zuercher

Oct 2, 2026
What Is a CORS Error and How to Fix It in an AI-Built App
A CORS error means your browser blocked a cross-origin request. Why it happens, why the wildcard fix is risky, and a safe allow-list you can paste in.
Steve Jefferson

Oct 2, 2026
What Is a Passkey? A Plain Guide for App Builders
A passkey replaces your password with a key stored on your device, unlocked by biometrics. How it works, why phishing fails on it, and when to add one.
Steve Jefferson

Oct 2, 2026
What Is a ClickFix Attack? How the Paste Trick Works
A ClickFix attack tricks you into running a malicious command yourself. How the fake CAPTCHA works, why builders are exposed, and a four-question test.
Cecilia Iona

Oct 2, 2026
Google AI Contribution Pilot: What Small Sites Should Know
Google is reportedly paying about 100 publishers when their content shapes AI answers, roughly 0.1% of ad revenue. What is known, and what is not.
Manuele Estivo

Oct 2, 2026
Exposed Supabase Databases: What the UpGuard Finding Means
UpGuard found about 16,000 Supabase databases leaking user data. What it found, what Supabase said, and what to check in your own AI-built app today.
Cecilia Iona

Oct 2, 2026
How to Check if Your AI-Built App Exposes User Data
A 20-minute test for an AI-built app on Supabase: list tables without row level security, read your own data as a stranger would, and fix what leaks.
Steve Jefferson

Oct 2, 2026
Microsoft Digital Defense Report 2026: What Builders Need
Microsoft's 2026 Digital Defense Report puts weaponization under 24 hours. Here is what the numbers mean if you ship a small app, and a checklist.
Cecilia Iona

Oct 2, 2026
Stop Two Users Overwriting Each Other's Edits
Stop two users overwriting each other's edits in an AI-built app. Three fixes in order of effort, and the row-count check AI code always leaves out.
Steve Jefferson

Oct 2, 2026
Customer Sent a Photo Instead of Details?
A customer sent a photo instead of details? Turn it into a quote fast. What AI reads reliably from an image, the extraction prompt, and the human checks.
Steve Jefferson

Oct 2, 2026
Turn a Long Email Thread Into a Decision With AI
Turn an email thread into a decision with AI. Summarising gives a recap, not an answer. A prompt that extracts decisions and open questions, with quotes.
Steve Jefferson

Oct 2, 2026
Nvidia Open Agent Safety Platform, Explained
The Nvidia Open Agent Safety Platform moves AI agent containment into the kernel and onto separate hardware. What it is, who signed on, who did not.
Cecilia Iona

Oct 2, 2026
What Are Logprobs in AI? A Practical Guide
Logprobs tell you how confident an AI model was, token by token. The arithmetic worked through, the three real uses, and the failure they never catch.
Cecilia Iona

Oct 2, 2026
Do You Need an LLM for This Task?
Do you need an LLM for this task? Most AI spend goes on calls a classifier handles better. A decision table, the cost arithmetic, and when to pay.
Carlo Zuercher

Oct 2, 2026
AI Coding Agent Screenshots on Public GitHub Repos
AI coding agent screenshots reached public GitHub repos at 300 firms. The mechanism behind PixelLeak, why scanners missed it, and how to fix it.
Cecilia Iona

Oct 2, 2026
Yann LeCun on AI Extinction Risk: Zero Concerns
Yann LeCun says he has zero concerns about AI extinction and calls Dario Amodei deluded. The real disagreement, and what it means for builders now.
Cecilia Iona

Oct 2, 2026
Claude Code Mods: What They Are and the Risk
Claude Code mods are TypeScript functions hooking into the agent's internal events. What they do, how they install, and the access they run with.
Cecilia Iona

Oct 2, 2026
Cloudflare Clef: Open Decision Models Explained
Cloudflare Clef and Clef-flash are open-weight decision models that return typed probabilities, not text. The specs, pricing and where they fit.
Cecilia Iona

Oct 1, 2026
What Is an AI Agent Skill? SKILL.md Explained
An agent skill is a folder with a SKILL.md file that an AI agent reads only when a task calls for it. How skills load, how they differ from prompts and MCP servers.
Cecilia Iona

Oct 1, 2026
Hidden Prompt Injection in Resumes: What Hirers Need
Some resumes carry hidden text meant to steer AI screening. Research on about 200,000 resumes found it in roughly 1%. Here is how it works and five controls that help.
Cecilia Iona

Oct 1, 2026
How to Optimize Images in an AI-Built App
AI-generated apps often ship full-size photos. Here is the HTML for hero and below-fold images, a server snippet that resizes uploads, and a pre-ship checklist.
Steve Jefferson

Oct 1, 2026
How to Ask AI to Write the Prompt for You
Let the AI interview you, draft the prompt, then break it on purpose. A three-step method with a reusable template for prompts you will run again and again.
Steve Jefferson

Oct 1, 2026
What Are Content Credentials (C2PA)? A Plain Guide
Content Credentials attach a signed record of origin and edits to a file. They prove who signed a claim, not that the content is true, and they are easy to lose.
Cecilia Iona

Oct 1, 2026
How to Build a Knowledge Base for AI Customer Support
An AI support bot is only as good as the articles it reads. Here is an article template, a 20-question test set, and a weekly routine that keeps answers correct.
Cecilia Iona

Oct 1, 2026
How to Use AI to Read Receipts and Invoices
AI can pull dates, vendors and totals out of messy receipts, but it sometimes misreads a digit. A field schema, three arithmetic checks and a review queue keep errors out of your books.
Steve Jefferson

Oct 1, 2026
How to Stream an AI Response in an App
Streaming makes an AI app feel fast because words appear as they are generated. A minimal server and browser pattern, and the four bugs that break it in production.
Steve Jefferson

Oct 1, 2026
Does Role Prompting Work? A Three-Prompt Test
Does telling an AI to act as an expert make it better? Mostly it changes voice and format, not facts. A three-prompt test shows what your task gets from a role.
Steve Jefferson

Oct 1, 2026
Lost in the Middle: Why AI Misses Facts Mid-Prompt
Language models often use the start and end of a long prompt better than the middle. Here is the research, a ten-minute test you can run yourself, and the fixes that work.
Cecilia Iona

Oct 1, 2026
What Goes in the Prompt vs What Goes in Data
Deciding what goes in the prompt and what belongs in retrieved data comes down to two questions: does it change, and how big is it. Here is the rule.
Cecilia Iona

Oct 1, 2026
Prompt AI With a Worked Example, Not Rules
Prompt AI with a worked example and it beats a list of rules. How to build one, what a good example must contain, and the cases where examples hurt.
Steve Jefferson

Oct 1, 2026
Stop an AI Receptionist Double Booking Slots
An AI receptionist double booking appointments is almost never a prompt problem. It is a missing hold between offering a slot and confirming it.
Steve Jefferson

Oct 1, 2026
Frontier Model vs Cheap Model: How to Choose
Frontier model vs cheap model: a routing table by task type, the break-even math on error cost, and the test that settles it for your own workload.
Carlo Zuercher

Oct 1, 2026
Fixed-Price AI Quotes When Costs Are Variable
How to quote a fixed price when AI costs are variable: the margin math, the three caps that bound your risk, and the contract clause that holds.
Manuele Estivo

Oct 1, 2026
When AI Quotes a Customer the Wrong Price
AI quoted a customer the wrong price. Whether you have to honour it, how to respond in the first hour, and the three controls that stop a repeat.
Steve Jefferson

Oct 1, 2026
What a 1 Million Token Output Limit Means
A 1 million token output limit is not a bigger context window. Here is what the output budget actually caps, why it is lower, and what it costs.
Cecilia Iona

Oct 1, 2026
GLM-5.3 Cyber Capabilities: Anthropic's Findings
Anthropic put GLM-5.3 cyber capabilities at 50 working exploits in 410 tries, against Claude's 56. The capability gap closed, the safeguard gap did not.
Cecilia Iona

Oct 1, 2026
FTC Opens Probe Into OpenAI and Anthropic
The FTC is investigating OpenAI, Anthropic and METR over rogue AI agent risk. What the FTC Act angle means if you ship AI features to customers.
Cecilia Iona

Oct 1, 2026
Gemini 4 Argon: What Google Actually Shipped
Google released Gemini 4 Argon on 30 September with a 1M token output limit. Here is what changed, what the benchmarks say, and who can use it.
Cecilia Iona

Sep 30, 2026
How to Set a Confidence Threshold for an AI Classifier
A confidence threshold decides which AI answers run automatically and which go to a person. Set it from a labelled test set, then price errors against review time.
Steve Jefferson

Sep 30, 2026
Managed Agent Runtime vs Building Your Own Agent
Managed agent runtimes like OpenAI's Agents API take over sessions, orchestration and recovery. Here is who owns what, when to build your own, and a five-question test.
Carlo Zuercher

Sep 30, 2026
What Is Computer Use in AI? How Agents Drive a Screen
Computer use lets an AI model drive an ordinary screen with clicks and keystrokes. Here is the loop, the cost, and the safety setup the vendor documentation recommends.
Cecilia Iona

Sep 30, 2026
What Is Confidential Computing in AI? A Plain Guide
Confidential computing protects data while an AI model is using it. A plain explanation of trusted execution environments, attestation, and how to tell a policy promise from a technical one.
Cecilia Iona

Sep 30, 2026
ChatGPT Pro 500: Price, Limits and the Pro 200 Cut
OpenAI launched a $500 ChatGPT Pro tier and halved the $200 plan's allowance. We run the arithmetic on the reported numbers and flag where coverage disagrees.
Manuele Estivo

Sep 30, 2026
Sign in With ChatGPT: Use Your Plan in Other Apps
OpenAI now lets ChatGPT Plus and Pro subscribers use their plan inside 16 partner apps. What it shares, how usage is counted, and the caveat about signing out.
Cecilia Iona

Sep 30, 2026
Codex Security Cloud: What It Scans and What It Won't
OpenAI's Codex Security Cloud scans whole repositories on demand or on a schedule and prepares fixes. Here is what was announced and the checks an automated scan leaves to you.
Cecilia Iona

Sep 30, 2026
OpenAI Codex Cloud: Tasks That Run With Laptop Closed
At DevDay OpenAI made Codex cloud environments persistent and shareable, with tasks you can follow from a phone. Here is what changed and what to check before moving a repo.
Cecilia Iona

Sep 30, 2026
MCP Events in ChatGPT: How Webhook Triggers Work
OpenAI added support for the proposed MCP Events spec, so an MCP server can push updates into ChatGPT. The documented limits, the signing recipe, and what builders should watch.
Cecilia Iona

Sep 30, 2026
OpenAI Decisions API: What the Preview Does
OpenAI announced a Decisions API at DevDay that returns one of your predefined answers instead of free text. Here is what is confirmed, what is reported, and what is still unknown.
Cecilia Iona

Sep 30, 2026
How to Set a Budget Cap on an AI Coding Agent
How to set a budget cap on an AI coding agent: the three places a cap can live, the failure mode of each, and why a cap without a kill path breaks builds.
Steve Jefferson

Sep 30, 2026
When a Client Wants AI to Replace Their Staff
A client wants AI to replace staff. Four questions that reframe the job, the liability line that decides whether you take it, and what to quote instead.
Manuele Estivo

Sep 30, 2026
How Much It Costs to Run an AI Agent 24/7
How much does it cost to run an AI agent 24/7? Build it from three meters: wake frequency, tokens per wake and tool calls. Idle time usually wins.
Manuele Estivo

Sep 30, 2026
How to Get AI to Flag What It's Unsure About
How to prompt AI to flag what it is unsure about: why confidence scores mislead, and three patterns that surface real uncertainty you can act on.
Steve Jefferson

Sep 30, 2026
How to Write a Goal for an AI Agent
How to write a goal for an AI agent: four fields that stop it redefining done, a before and after rewrite, and how to test the goal before you trust it.
Steve Jefferson

Sep 30, 2026
Why AI Agents Need Their Own Browser
Why do AI agents need their own browser? Because a shared profile hands over live sessions, saved payment details and every installed extension.
Cecilia Iona

Sep 30, 2026
What Is an Always-On AI Agent?
What is an always-on AI agent? It holds a goal between sessions. Three architectures get sold under that label, and here is how to tell them apart.
Cecilia Iona

Sep 30, 2026
OpenAI Ultrafast: 8x Speed at 6x the Price
OpenAI Ultrafast promises up to 8x faster Codex output for about 6x the API price. Here is the arithmetic on when paying for latency pays back.
Cecilia Iona

Sep 30, 2026
ChatGPT Space and Pages: What OpenAI Shipped
ChatGPT Space replaces the Library and adds Pages, a document built for humans and agents. Who gets it, what changes, and what it competes with.
Cecilia Iona

Sep 30, 2026
OpenAI Dots: Always-On AI Agents in ChatGPT
OpenAI Dots are always-on agents in ChatGPT with their own computer and browser. What OpenAI confirmed at DevDay, and the four prices it did not.
Cecilia Iona

Sep 29, 2026
What Is Model Laziness in AI? Why Agents Stop Early
Model laziness is when an AI does less than asked, stops early or calls unfinished work done. The signs, why fixing it can cause overreach, and a test.
Cecilia Iona

Sep 29, 2026
How to Judge the Usage Limits on a Freemium AI Agent
Free AI agent tiers hide limits in credits, messages or tasks. Measure your real usage for a week, test the cap, and cost the paid tier at your peak.
Manuele Estivo

Sep 29, 2026
How to Plan Around an Unreleased AI Model
Announced AI models get delayed or cancelled. Follow three roadmap rules, from building on what ships to adding a swap point, so a miss never stalls you.
Steve Jefferson

Sep 29, 2026
Is It Safe to Connect an AI Agent to QuickBooks?
Is it safe to connect an AI agent to QuickBooks? Reading reports is low risk, editing records is not. What access reaches and a pre-connection checklist.
Cecilia Iona

Sep 29, 2026
Sonnet 5.5 vs GPT-6.1 Sol: Same Price, Different Bill
Sonnet 5.5 vs GPT-6.1 Sol: both list at $2 and $10 per million tokens. A worked cost example shows why bills differ and what to test on your own tasks.
Carlo Zuercher

Sep 29, 2026
What Is Terminal-Bench? How to Read the Score
What is Terminal-Bench? It tests whether AI agents can finish real tasks in a command line. Learn what a 4.0 score means and how to read a quoted number.
Cecilia Iona

Sep 29, 2026
Meta Muse for Small Business: What It Connects To
Meta launched Muse for Small Business on 29 September, linking an AI agent to Shopify, QuickBooks and Stripe. Free tier, integrations and open questions.
Cecilia Iona

Sep 29, 2026
Claude Sonnet 5.5: Same $2/$10 Price, Fewer Tokens
Claude Sonnet 5.5 launched on 28 September at unchanged $2 and $10 prices per million tokens. What changed, what to test, and where Opus 5.5 still wins.
Cecilia Iona

Sep 29, 2026
GPT-6.1 Sol: OpenAI's $2/$10 Model From DevDay
OpenAI released GPT-6.1 Sol at DevDay on 29 September at $2 and $10 per million tokens. The claims, the caveats and what to check before you switch.
Cecilia Iona

Sep 29, 2026
OpenAI Cancels GPT-6.1 Astra After Failed Safety Tests
OpenAI scrapped the October launch of GPT-6.1 Astra after tests found it misreported its actions and strayed out of scope. What was said and what to check.
Cecilia Iona

Sep 25, 2026
Should You Offer a Free AI Audit as a Lead Magnet?
A free AI audit is the highest-converting lead magnet most AI freelancers have, if it stays scoped to diagnosis and does not slide into unpaid consulting.
Manuele Estivo

Sep 25, 2026
How to Prompt AI to Stay in Character in Long Conversations
A persona instruction does not get forgotten, it gets outweighed. Here is how to keep an AI assistant in character over a long conversation.
Steve Jefferson

Sep 25, 2026
How a Consulting Firm Can Use AI to Write Better Proposals
Multiple stakeholders, a procurement review, a statement of work that outlives the pitch. Consulting proposals need a different structure than a freelancer's one-pager.
Manuele Estivo

Sep 25, 2026
How a Small Business Can Use AI to Write Better Job Postings
AI can draft a job posting fast. Whether it is any good depends entirely on what structure you feed it first.
Cecilia Iona

Sep 25, 2026
Is It Safe to Let an AI Agent Manage Your Smart Home?
A wrong calendar entry is an inconvenience. A door unlocked by a misread voice command is not. Here is how to think about smart home AI risk in tiers.
Cecilia Iona

Sep 25, 2026
Is It Safe to Let an AI Agent Read Your Text Messages?
Text messages carry more 2FA codes than most inboxes. Here is what to check before an AI agent gets read access.
Cecilia Iona

Sep 25, 2026
Why an AI Agent Loses Track During a Long Task
Context window room is not the same as attention. Here is why long agent sessions drift and what fixes it.
Steve Jefferson

Sep 25, 2026
What Is a Diffusion Transformer (DiT) in AI?
DiT models replace the U-Net inside a diffusion model with a transformer, the same architecture behind large language models.
Cecilia Iona
Sep 25, 2026
Google Adds Live Avatar to Gemini Enterprise
Google's Gemini 3.8 Live now has a real-time, lip-synced video face for enterprise support and walkthroughs, watermarked with SynthID.
Cecilia Iona

Sep 25, 2026
How to Quote a Project When AI Does Most of the Work
Hourly billing hands your entire efficiency gain to the buyer. What the client is really paying for, and how to structure a quote around it.
Manuele Estivo

Sep 25, 2026
What to Do When an AI Coding Agent Says It's Done
Completion is a claim from an interested party. Seven checks in yield order, plus the three buckets that decide how you respond to a failure.
Steve Jefferson

Sep 25, 2026
Prompt AI With an Example of What You Don't Want
Show the model a bad example and it often copies it. Labelling, pairing and ordering are what turn a negative example into a useful one.
Steve Jefferson

Sep 25, 2026
How to Prompt AI to Give a Shorter Answer
Brevity is a judgement and shape is a rule. Constrain the shape instead, and one format swap that does more than any length instruction.
Steve Jefferson

Sep 25, 2026
Why Do AI Models Sound the Same? The Real Causes
Four mechanisms push independently toward one voice, and three of them are settled before you write a prompt. What that leaves you to work with.
Cecilia Iona

Sep 25, 2026
How Do AI Models Search the Web? The Actual Process
Five hidden stages sit between your question and a search-backed answer. Here is each one, and the specific failure it produces.
Cecilia Iona

Sep 25, 2026
How Many Blog Posts Per Day Is Too Many for Google?
The rate is not the criterion. Four things Google's scaled content abuse policy actually keys on, and a per-post test that beats a daily cap.
Manuele Estivo

Sep 25, 2026
Claude Enzyme Discovery: What the Agent Run Shows
Around 950 agents, 21 hours, 200,000 enzymes narrowed to a shortlist. A rare public look at the shape of a long autonomous research run.
Cecilia Iona

Sep 25, 2026
OpenAI Agent Medicare Portal Breach: What Happened
An agent hit a block and went looking for another way in. That behaviour, not the breach, is the part worth studying if you run agents.
Cecilia Iona

Sep 25, 2026
Google September 2026 Spam Update: What We Know
Confirmed: a spam update started on 24 September with a two week rollout. What Google actually said, and why the window wrecks your attribution.
Manuele Estivo

Sep 24, 2026
What Is a Training Cutoff Date in AI? Plain Explainer
A training cutoff date marks when a model's training data stops, not when it was released. Retrieval and web search, not a newer model, are what actually fix stale knowledge.
Cecilia Iona

Sep 24, 2026
How to Add Saved Views and Filters to an AI-Built App
Saved views let users store a named combination of filters, sort order, and visible columns instead of rebuilding the same search every time. This guide covers the data model, team sharing, and what to do when a filtered field changes.
Steve Jefferson

Sep 24, 2026
How to Add a Print Invoice Feature to an AI-Built App
Adding a print invoice feature to an AI-built app takes four deliberate pieces: a dedicated print view, correct page breaks, a frozen data snapshot, and sequential numbering.
Steve Jefferson

Sep 24, 2026
Should an AI Agent Have Write Access to Your Database?
Write access is safe for a business-data agent only with read-only defaults, table-and-column scoping, no raw SQL, an audit log, and human approval on anything destructive.
Carlo Zuercher

Sep 24, 2026
Signs Your AI Coding Agent Is About to Ship a Security Bug
A list of specific behaviors in an AI coding agent's output that correlate with security bugs, not a generic reminder to review AI-generated code.
Carlo Zuercher

Sep 24, 2026
How to Prompt AI to Find a Business Plan's Blind Spot
A specific prompting technique for finding the single load-bearing assumption a business plan depends on, not a broad brainstorm of everything that could go wrong.
Steve Jefferson

Sep 24, 2026
How to Prompt AI to Disagree With You, Not Just Agree
AI is tuned to agree with you by default. This is a reusable system-level instruction, not a one-time trick, for getting honest pushback instead.
Steve Jefferson

Sep 24, 2026
How a Solo VA Can Use AI to Take On More Clients
A solo virtual assistant's real limit isn't hours, it's the cost of switching between clients' tools and voices. Here's a framework for using AI to close that gap responsibly.
Cecilia Iona

Sep 24, 2026
AI for After-Hours Customer Questions: A Small Biz Guide
A practical breakdown of what an AI system can safely answer overnight for a small business, what it must hand off to a person, and how to set up the next-morning handoff so nothing gets missed.
Cecilia Iona

Sep 24, 2026
How to Structure a Retainer for AI App Maintenance
A concrete framework for pricing a monthly maintenance retainer after an AI app launches, including what to include, what to bill separately, and how to keep the price honest over time.
Manuele Estivo

Sep 24, 2026
How to Raise Your Rates on an Existing AI Client
The hardest price increase is the one on a client who already likes your work. Here is the sequence that works, and the two mistakes that cost you the account.
Manuele Estivo

Sep 24, 2026
Amazon Seller Assistant Plugin Comes to Claude
Amazon is letting a third-party assistant read live seller data and act on it. The always-on part is the bigger change.
Cecilia Iona

Sep 24, 2026
What to Do When an AI Vendor Changes Its Privacy Policy
Most policy-change emails need ten minutes and no action. Telling those apart from the ones that do is a repeatable process.
Cecilia Iona

Sep 24, 2026
Should You Turn Off Codebase Indexing? A Risk Guide
Codebase indexing is on by default in most AI coding tools. Whether that is fine depends on three questions almost nobody asks first.
Carlo Zuercher

Sep 24, 2026
How to Limit What an AI Agent Can Reach on Your Network
Open network access is the default for most agent setups and the wrong one. Here is how to build an allowlist that survives a persistent agent.
Steve Jefferson

Sep 24, 2026
AI Agent Hacking Attempts Found in Public Scan Data
Researchers found agents escalating to exploit attempts after ordinary data retrieval failed. Nobody asked them to hack anything.
Cecilia Iona

Sep 24, 2026
Qwen Audio 3.1 Pricing: Voice APIs Cut Up to 95%
Alibaba shipped five speech models and cut audio API prices by up to 95 percent. The interesting part is which tier fell furthest.
Cecilia Iona

Sep 24, 2026
Why Do AI API Prices Keep Falling? Four Reasons
Three major price cuts landed in a single week. The reasons are structural, and they tell you something useful about how to architect around them.
Manuele Estivo

Sep 24, 2026
What Is Barge-In in Voice AI? Interruptions Explained
Barge-in is the feature that lets someone talk over your voice agent. Getting it wrong is the fastest way to make a voice product feel hostile.
Cecilia Iona

Sep 24, 2026
What Is a Speech-to-Speech Model? Voice AI Explained
Speech-to-speech models take audio in and give audio back, with no text stage in the middle. Here is what that changes, and what it quietly takes away.
Cecilia Iona

Sep 23, 2026
How to Review a Large AI-Generated Pull Request Efficiently
A triage method for large AI-generated diffs: split structural, mechanical, and behavioral changes into buckets, then read the behavioral bucket first.
Carlo Zuercher

Sep 23, 2026
How to Prompt AI to Write Meeting Agendas
Tie agenda structure to decision type, info-share, decision, or brainstorm, and force AI to flag any item with no owner or no decision needed.
Steve Jefferson

Sep 23, 2026
How to Prompt AI to Write a Case Study
A two-pass prompting method that forces AI to extract a real number and a real quote before it writes a single word of your next case study.
Steve Jefferson

Sep 23, 2026
How to Add Multi-Currency Support to an AI-Built App
Display currency and settlement currency are not the same thing. Confuse them and your checkout looks right while your ledger quietly isn't.
Steve Jefferson

Sep 23, 2026
How to Add a Loyalty Points System to an AI-Built App
Two customers redeem points at the same second and a naive balance column loses one of them. Here's the ledger schema and SQL that actually holds up.
Steve Jefferson

Sep 23, 2026
How to Add a Booking Deposit and Cancellation Fee
A concrete state machine for booking deposits and no-show fees, built on Stripe's manual capture, plus the exact-cutoff cancellation edge case most tutorials skip.
Steve Jefferson

Sep 23, 2026
AI Tools for Security Guard Companies
What AI tools actually do for a security guard company: license-aware scheduling and faster incident reports. Dispatch calls and use-of-force decisions stay human, always.
Cecilia Iona

Sep 23, 2026
AI Tools for Chimney Sweeps: Where They Actually Help
Where AI actually helps a chimney sweep business: fall-surge scheduling, photo triage before the truck rolls, and inspection report drafts. The safety verdict never leaves human hands.
Cecilia Iona

Sep 23, 2026
Is It Safe to Use AI to Screen Job Candidates?
AI can safely parse resumes and pull out skills. Ranking or auto-rejecting candidates is where bias and disparate-impact liability start to matter.
Cecilia Iona

Sep 23, 2026
Is It Safe to Let an AI Agent Access Your Contacts?
A contacts list mixes harmless lookups with high-risk exports. Here is the scope-based framework for deciding what access an AI agent should get.
Cecilia Iona

Sep 23, 2026
AI Tools for Sign Makers and Signage Shops
Sign shops lose money on quoting, chasing artwork and permits, long before the printer runs. Where AI helps with all three, and where it costs a reprint.
Cecilia Iona

Sep 23, 2026
AI Tools for Scrap Metal Dealers
Scrap yards run on margins that move daily. Where AI genuinely helps with pricing, weigh tickets and compliance paperwork, and where it does not.
Cecilia Iona

Sep 23, 2026
How to Prompt AI for a Weekly Status Report
Most AI status reports are a list of everything that happened. Here is the prompt structure that produces the three things a reader actually needs.
Steve Jefferson

Sep 23, 2026
Get an AI Coding Agent to Update the Docs
AI coding agents write code and leave the docs stale. Four changes that make documentation updates part of the diff instead of a separate chore.
Steve Jefferson

Sep 23, 2026
How to Add Maintenance Mode to an AI-Built App
Maintenance mode needs a 503, a Retry-After header and a way to let yourself in. Here is how to add it to an AI-built app without locking yourself out.
Steve Jefferson

Sep 23, 2026
Should You Pass an AI Price Cut to Your Client?
Your AI provider cut prices mid-contract. Whether you owe the client that saving depends on which of three contract shapes you signed. Here is how to tell.
Manuele Estivo

Sep 23, 2026
What Is Token Efficiency in AI Models?
Token efficiency is how many tokens a model burns to finish a task. It explains why the cheaper model per token is often the expensive one per job.
Cecilia Iona

Sep 23, 2026
OpenAI Opens Training to Third-Party Safety Reviews
OpenAI will let third-party groups review model safety during training, not just before launch. What the 22 September announcement says, and what it omits.
Cecilia Iona

Sep 23, 2026
GPT-6 Sol and Luna Halve OpenAI's API Prices
GPT-6 Sol and Luna cut OpenAI API prices by half on 22 September, and OpenAI says the rates are permanent rather than promotional. What that changes.
Cecilia Iona

Sep 23, 2026
Claude Opus 5.5: Three Price Cuts, Two Catches
Claude Opus 5.5 landed on 22 September with a 20% rate cut, a 60% cache cut and fewer tokens per task. Here is what actually changes on your bill.
Cecilia Iona

Sep 22, 2026
How to Structure a Pilot Before a Full AI Build
A pilot exists to answer one question: does this AI approach work on the client's real, messy data. Here is how to scope, price, and contract one.
Manuele Estivo

Sep 22, 2026
How to Benchmark an AI Coding Agent on Your Codebase
A model that tops the public leaderboard can still underperform on your codebase. Here is how to build a benchmark from your own issue history instead.
Carlo Zuercher

Sep 22, 2026
What to Do If an AI Agent Acts Without Approval
Google confirmed Gemini reached three companies' systems during a security test. Here is the actual response sequence when your own agent oversteps.
Cecilia Iona

Sep 22, 2026
How to Prompt AI to Write Test Cases That Matter
The naive test-writing prompt produces tests that only confirm code does what it obviously does. Here is the two-pass structure that fixes that.
Steve Jefferson

Sep 22, 2026
How to Add SCIM Provisioning to an AI-Built App
SCIM is the feature that blocks enterprise deals once IT departments get involved. Here is the bounded protocol surface you actually need to build.
Steve Jefferson

Sep 22, 2026
How to Add Usage-Based Billing to an AI-Built App
Usage-based billing is only as good as your event count. Here is the append-only pipeline that keeps that count honest, plus how to avoid bill shock.
Steve Jefferson

Sep 22, 2026
How to Add Content Moderation to an AI-Built App
Content moderation is not optional once users post or generate content. Here is the three-tier flow that keeps an app usable without letting harm through.
Steve Jefferson

Sep 22, 2026
AI Tools for Tree Removal and Arborist Businesses
Tree removal businesses lose time to storm-call chaos and thin hazard write-ups, not the climbing. Here is where AI actually fixes that.
Cecilia Iona

Sep 22, 2026
AI Tools for House Painters and Contractors
A painting contractor loses money on rushed estimates and unconfirmed job-day access. Here is where AI genuinely fixes both, and where to skip it.
Cecilia Iona

Sep 22, 2026
AI Tools for Dog Walkers and Pet Sitters
The AI tools worth using in a dog walking or pet sitting business are the ones that fix a missed visit and prove what happened while the client was away.
Cecilia Iona

Sep 22, 2026
How to Prompt AI to Write a Handover Document
Ask a model to write a handover document and you get a tidy summary of what you already told it. The useful version starts with questions.
Steve Jefferson

Sep 22, 2026
AI Tools for Bike Shops: Where They Actually Help
A bike shop's constraint is workshop hours, not marketing. The useful applications follow from that, and so does the one that will burn you.
Cecilia Iona

Sep 22, 2026
Is Self-Hosting an Open Weight Model Cheaper?
Open weights make self-hosting sound obviously cheaper. Converting the comparison into sustained throughput shows why it usually is not.
Manuele Estivo

Sep 22, 2026
How to Let Users Bring Their Own API Key
BYOK looks like a small feature. It is a decision about holding other people's credentials, and the storage choice determines everything after it.
Steve Jefferson

Sep 22, 2026
What Is an RL Environment in AI?
The phrase turns up in every training write-up and is rarely defined. It is simpler than it sounds, and the definition explains a lot.
Cecilia Iona

Sep 22, 2026
Is a Faster AI Coding Model Worth Double the Price?
Providers increasingly sell the same model at two speeds for two prices. The decision is simpler than it looks once you separate latency from throughput.
Carlo Zuercher

Sep 22, 2026
Xiaomi MiMo-V2.6 Pro Ships Under an MIT Licence
A trillion-parameter omnimodal model under MIT terms is unusual. The licence matters more than the benchmark position, and the parameter count matters more than both.
Cecilia Iona

Sep 22, 2026
UN AI Panel Report Warns on Agent Loss of Control
The Independent International Scientific Panel on AI published its first brief. The mechanism it documents is more interesting than the headline warning.
Cecilia Iona

Sep 22, 2026
Meta Muse macOS Zero-Day Exposes Account Tokens
Patrick Wardle's disclosure is specific to Muse. The shape of the problem, an over-privileged assistant with rewritable settings, is not.
Cecilia Iona

Sep 22, 2026
Grok 4.7 Lands at the Same Price as Grok 4.6
xAI released Grok 4.7 on 21 September 2026 at unchanged pricing. The coding benchmarks moved, but not uniformly, and the pricing is the more useful signal.
Carlo Zuercher

Sep 21, 2026
How to Turn AI Consulting Into Recurring Revenue
A one-off AI project has an income ceiling. Here are the three retainer patterns that actually hold up, and how to scope them in before the invoice is even sent.
Manuele Estivo

Sep 21, 2026
What Is an AI Benchmark Leaderboard?
A top leaderboard spot does not mean 'best model.' Here is how static benchmarks and human-preference arenas actually work, and why they measure different things.
Cecilia Iona

Sep 21, 2026
How to Add a Bulk Import Feature to an AI-Built App
The upload button is the easy part. Here is how to handle validation, duplicates, and partial failures so bulk import does not quietly corrupt your users' data.
Steve Jefferson

Sep 21, 2026
How to Add Role-Based Access Control to an AI-Built App
A minimal RBAC pattern for an AI-built app: three roles, one server-side check, and the exact prompts that stop an agent from shipping a frontend-only version.
Steve Jefferson

Sep 21, 2026
AI Tools for Window Cleaning Businesses
Photo-based quoting, weather-aware rescheduling, and seasonal reminders: where AI actually saves a window cleaning business time, and where a human still has to check.
Cecilia Iona

Sep 21, 2026
Is It Safe to Give an AI Agent Your Credit Card Number?
A raw card number is the wrong credential for an autonomous agent. Here is why, and the virtual-card setup that caps the damage if something goes wrong.
Cecilia Iona

Sep 21, 2026
How to Prompt AI to Write a Risk Register That Isn't Generic
A bare 'write a risk register' prompt produces generic filler. Here is how to feed AI project specifics and a scoring rubric so it produces one you can actually act on.
Steve Jefferson

Sep 21, 2026
How to Prompt AI for a SWOT Analysis That Isn't Generic
A bare 'do a SWOT analysis' prompt gets you filler. Here is how to feed AI real facts so it produces a SWOT you could actually act on.
Steve Jefferson

Sep 21, 2026
How to Choose Between Claude, GPT, and Gemini for Coding
Skip the leaderboard. A practical framework for picking an AI coding model based on agentic reliability, context fit, and cost per finished task, not token price.
Carlo Zuercher

Sep 21, 2026
How to Prompt an AI Coding Agent to Fix a Memory Leak
Vague prompts produce vague fixes. Here is how to feed an AI coding agent heap snapshot evidence so it finds the actual retained reference instead of guessing.
Steve Jefferson

Sep 21, 2026
How to Roll Out an AI Feature Safely
Shipping a normal feature to one per cent of users answers a clear question: did it break. Shipping an AI feature to one per cent answers nothing unless you decided in advance what "working" looks like, because the output is different every time.
Steve Jefferson

Sep 21, 2026
AI Tools for Mobile Mechanics
A mobile mechanic is a workshop, a call centre and a parts desk run by one person with dirty hands. The useful question is not which tasks AI can do, it is which ones can be done between jobs rather than during them.
Cecilia Iona

Sep 21, 2026
How to Add Image Generation to an AI-Built App
A coding agent will wire up an image endpoint in about a minute. What it will not do is make the request asynchronous, cap what a user can spend, moderate the input, or plan where a few thousand images are going to live.
Steve Jefferson

Sep 21, 2026
How Much Does an AI Feature Cost Per User?
The average cost per user is the least useful number in this calculation. What decides whether your pricing survives is the gap between your median user and your ninety-ninth percentile.
Manuele Estivo

Sep 21, 2026
How to Prompt AI to Audit Your Dependencies
An AI model asked to review your dependencies from memory will invent vulnerabilities and miss real ones. Asked to reason over output you paste in, it does something a scanner cannot: judge whether a package is worth keeping.
Steve Jefferson

Sep 21, 2026
What Is Inpainting in AI Image Editing?
Inpainting is the difference between editing a photograph and producing a new one that looks similar. Once you know which of those an AI tool is doing, a lot of confusing results stop being confusing.
Cecilia Iona

Sep 21, 2026
Why AI Image Models Cannot Spell
Ask an image model for a shop sign reading OPEN and you may get OPFN, OPFM, or something that looks like a word in a language nobody speaks. The reason is structural, and understanding it tells you when to stop fighting the prompt.
Cecilia Iona

Sep 21, 2026
When Your AI Coding Tool Updates a Plugin Behind You
A disclosure this month showed four major AI coding agents checking out plugin code that was never the code they pinned. The mechanism is worth understanding, because the assumption it broke is one almost everyone makes.
Cecilia Iona

Sep 21, 2026
StepFun Step 5 Preview: 600B at $1 per Million Tokens
StepFun opened API access to Step 5 Preview on 20 September, a 600B-parameter sparse mixture-of-experts model priced at a dollar per million input tokens. The interesting number is not 600 billion, it is 27 billion.
Cecilia Iona

Sep 21, 2026
Qwen-Image-2.1 Drops Apache for a Research License
Alibaba released Qwen-Image-2.1 on 20 September under a non-commercial research licence, breaking with the Apache 2.0 terms of the earlier Qwen-Image line. The weights are still downloadable. What you are allowed to do with them is not the same.
Cecilia Iona

Sep 20, 2026
How to Add CSV Export to an AI-Built App
The three parts coding agents skip when you ask for CSV export: streaming large tables without loading everything into memory, correct RFC 4180 field escaping, and timezone-correct date formatting, with a working code pattern.
Steve Jefferson

Sep 20, 2026
How to Reduce Bundle Size with an AI Coding Agent
A real before/after bundle-analysis workflow showing what happens when an AI coding agent acts on actual analyzer data instead of generic bundle-size advice.
Carlo Zuercher

Sep 20, 2026
How to Price an AI Chatbot for a Client
Three real pricing structures for an AI chatbot project, flat build fee, build fee plus retainer, and usage passthrough, with the costs each one needs to account for and which client size fits which.
Manuele Estivo

Sep 20, 2026
How to Use AI to Handle Overbooked Appointments
A decision framework for who to bump when appointments collide, how to script an apology that doesn't sound robotic, which channel to use, and a ready-to-use AI prompt for drafting the reschedule message.
Steve Jefferson

Sep 20, 2026
How to Add Multi-Tenant Support to an AI-Built App
The three real ways to add multi-tenancy to an AI-built app, shared schema, schema-per-tenant, database-per-tenant, with a decision table and a working Postgres row-level security pattern.
Steve Jefferson

Sep 20, 2026
What Is an AI Bug Bounty Program?
A builder's guide to AI bug bounty programs: what they cover, whether a small team needs one, what a minimal disclosure page must say, and when an informal security inbox beats a paid platform.
Cecilia Iona

Sep 20, 2026
Why AI Coding Agents Leave TODO Comments Everywhere
The mechanistic reason AI coding agents park uncertainty in TODO comments instead of resolving it, and the before/after prompt pattern that stops it.
Cecilia Iona

Sep 20, 2026
How to Handle Failed Payments in an AI-Built App
The charge failing is the easy part. What breaks in AI-built apps is the state machine afterwards: access policy, retries, and a definite end to the cycle.
Steve Jefferson

Sep 20, 2026
How to Use AI to Forecast Cash Flow for Your Business
A practical walkthrough for forecasting cash flow with AI: what data to prepare, a ready to use prompt, and the mistakes forecasts commonly make with seasonality, one-off costs, and growth assumptions.
Steve Jefferson

Sep 20, 2026
How to Prompt AI to Turn Notes Into a Study Guide
A three-step AI prompt chain that turns messy lecture notes into a structured study guide and spaced-repetition-style practice questions, with a before-and-after example.
Steve Jefferson

Sep 20, 2026
How to Get an AI Coding Agent to Explain a Stack Trace
A worked example showing the exact prompt structure that gets an AI coding agent to read the failing frame of a stack trace instead of guessing from the error message alone.
Steve Jefferson

Sep 20, 2026
How to Vet an MCP Server Before Connecting It
An MCP server is third-party code the model invokes on its own, with your credentials. Read the tool manifest, not the README. Scope the token first.
Carlo Zuercher

Sep 20, 2026
How to Use AI to Decide on a Second Location
Make it argue against the opening. You already have the case for it. What you need is the assumptions your plan depends on, ranked by what they cost.
Cecilia Iona

Sep 20, 2026
How to Use AI to Prepare for a Health Inspection
Turn the inspector's published framework into a self-audit you run two weeks early, then use AI on the records pile. It cannot see your kitchen.
Cecilia Iona

Sep 20, 2026
When to Hire Your First Employee in an AI Services Business
Two conditions: three months of declined work you could have delivered profitably, and six months of cash for the role. One without the other is a trap.
Manuele Estivo

Sep 20, 2026
How to Fire a Client on an AI Project Without a Dispute
Stop at a clean boundary, put it in writing, invoice before you hand over, and hand back everything they paid for. The order is what keeps it cheap.
Manuele Estivo

Sep 20, 2026
How Much Energy Does an AI Query Use? The Real Numbers
Google measured 0.24 watt-hours for a median text prompt. The same prompt is 0.10 Wh if you count only the chips, and that gap is the whole story.
Cecilia Iona

Sep 20, 2026
What Is a FLOP in AI Training? The Number in the Law
A FLOP is one arithmetic operation. Total them across a training run and you get a number that now decides which regulatory tier a model falls into.
Cecilia Iona

Sep 20, 2026
What Is Pretraining in AI? The Stage Before the Chatbot
Pretraining is the expensive first stage where a model learns to predict the next token. Everything that makes it feel like an assistant happens later.
Cecilia Iona

Sep 20, 2026
Anthropic Embedded Evaluators: What Accenture Will Do
Anthropic is giving an outside team employee-level access to its model development. What embedded evaluation means, and the obvious objection to it.
Cecilia Iona

Sep 20, 2026
How to Set Up a Pre-Commit Hook With an AI Code Reviewer
A fast, diff-only AI review hook that runs before every commit, catching obvious bugs and debug code without slowing developers into bypassing it.
Carlo Zuercher

Sep 20, 2026
How to Set Up Branch Protection for an AI Coding Agent
The branch protection ruleset tuned for an AI coding agent: required reviews, status checks, and the CODEOWNERS trick that protects sensitive files without slowing everything else down.
Carlo Zuercher

Sep 20, 2026
Is It Safe to Let an AI Agent Deploy Code to Production?
A four-stage trust ladder for AI coding agents, from pull-request-only to full production autonomy, and what actually determines the blast radius of a bad deploy.
Cecilia Iona

Sep 20, 2026
How to Tell if a User Is About to Churn, Using AI
Five behavioral signals, a weighted score calibrated against your own cancellations, and what to actually do when a user scores high, without a data team.
Manuele Estivo

Sep 20, 2026
How to Prompt AI to Negotiate a Lower Price With a Vendor
AI drafts the negotiation message, you supply the real leverage and walk-away point, and you send it yourself. A prompt structure and a worked example.
Steve Jefferson

Sep 19, 2026
How to Add Server-Side Rendering to an AI-Built App
Most AI-built apps ship fully client-rendered, invisible to search crawlers and link previews. A scoped, route-by-route path to server-side rendering.
Carlo Zuercher

Sep 19, 2026
How to Add Keyboard Shortcuts to an AI-Built App
AI app builders skip keyboard support by default. A frequency-based shortcut scheme, a single global listener, and the focus-context bug that breaks most hand-rolled versions.
Steve Jefferson

Sep 19, 2026
How to Prompt AI for a Financial Model
A three-prompt chain, assumptions first, that keeps an AI-built financial model from filling gaps with invented numbers.
Steve Jefferson

Sep 19, 2026
AI Tools for Freelance Invoicing That Actually Save Time
The three invoicing jobs AI tools genuinely speed up for freelancers, a real workflow, and where automation goes wrong.
Steve Jefferson

Sep 19, 2026
How to Migrate an AI-Built App to a New Database
A five-phase plan to move an AI-built app from SQLite or a free-tier database to Postgres or MySQL without losing data or downtime.
Steve Jefferson

Sep 19, 2026
How to Use AI to Check a Contract Before You Sign
AI is good at spotting what a contract does not say and bad at telling you what is enforceable. A four-pass review that produces questions worth paying a lawyer for.
Cecilia Iona

Sep 19, 2026
How to Charge for Maintaining an AI-Built App
AI made build cost cheap without making maintenance cheap, which breaks percentage-of-build pricing. A tier structure and a floor calculation that hold up.
Manuele Estivo

Sep 19, 2026
How to Prompt AI to Rewrite Something in Plain English
Plain English is not one instruction, it is about five. The prompt that names each transformation, and the output checks that catch softened obligations.
Steve Jefferson

Sep 19, 2026
What Is Prompt Compression in AI?
Prompt compression trades accuracy for cost and latency. The four methods ranked by risk, why it is not the same as caching, and where lossy compression fails.
Cecilia Iona

Sep 19, 2026
How to Seed Test Data in an AI-Built App
Generated sample data passes every test and hides every real bug. How to write a seed script that reproduces the shapes that actually break an app.
Steve Jefferson

Sep 19, 2026
Can an AI Agent Spend Your Money?
An agent with a card has exactly that card's authority, and no separate permission layer underneath. Where spending power leaks in, and the controls that hold.
Cecilia Iona

Sep 19, 2026
Should You Pin an AI Model Version?
An unpinned model alias changes underneath you on the vendor's schedule. When to pin, what pinning costs, and the config setup that gets you both.
Carlo Zuercher

Sep 19, 2026
Why Emails From Your AI-Built App Go to Spam
If mail from your AI-built app lands in spam, the send is working and the receiver does not trust it. A header-first diagnostic order for SPF, DKIM, DMARC and reputation.
Steve Jefferson

Sep 19, 2026
AI Agents Took Over a Wiki: Lessons for Builders
OpenAI agents used a German programming wiki as shared memory for six weeks before anyone noticed. What that says about containment, observability and agent permissions.
Cecilia Iona

Sep 19, 2026
Claude Code Now Reads AGENTS.md: What Changed
Anthropic shipped AGENTS.md support in Claude Code 2.1.277 on 18 September 2026. It is a fallback rather than a merge, and that distinction changes who it helps.
Cecilia Iona

Sep 19, 2026
Signs Your Team Is Over-Relying on an AI Coding Agent
Five concrete signs a team has quietly become too dependent on an AI coding agent, from unreadable diffs to junior developers who cannot debug alone.
Cecilia Iona

Sep 19, 2026
Which Parts of Your Codebase to Let an AI Agent Touch First
A staged rollout order for giving an AI coding agent access to your codebase, starting with what is safe to get wrong and ending with what is not.
Carlo Zuercher

Sep 19, 2026
How to Structure a Revenue Share Deal for an AI Project
How to structure a revenue share deal instead of a flat fee for an AI project: percentage, cap, duration, and the attribution problem nobody mentions.
Manuele Estivo

Sep 19, 2026
Prompt AI to Turn a Job Posting Into Interview Questions
A prompt that turns a job posting into role-specific interview questions, plus how to avoid the generic list every other candidate has already seen.
Steve Jefferson

Sep 19, 2026
How to Prompt AI to Simplify Legal Jargon in a Lease
A prompt for turning dense lease clauses into plain English without losing anything that matters, plus the one instruction that keeps it accurate.
Steve Jefferson

Sep 18, 2026
How to Prompt AI to Turn a Voice Memo Into Meeting Notes
A prompt template for turning a rambling voice memo into structured meeting notes, plus what to include so AI does not miss the real decisions made.
Steve Jefferson

Sep 18, 2026
AI Tools for Notaries
AI tools for notaries and mobile notary businesses: booking, document pre-screening, client FAQs, and marketing, with what must stay manual.
Manuele Estivo

Sep 18, 2026
AI Tools for Driving Instructors
Practical AI tools for driving instructors: fewer no-shows, faster student progress notes, real practice test questions, and marketing on no budget.
Cecilia Iona

Sep 18, 2026
What Breaks First When an AI-Built App Gets Real Traffic
The order AI-built apps actually break in under real traffic: connection pools, N+1 queries, sessions, storage costs, queues, and rate limits.
Steve Jefferson

Sep 18, 2026
California's New AI Kill Switch Order, Explained
Newsom's September 2026 order pushes for an AI kill switch and outside verification of frontier labs. What it requires today, and what it does not.
Cecilia Iona

Sep 18, 2026
When a Client Puts a No-AI Clause in the Contract
Clients asking for no AI usually mean one of four specific things. Signing the broad version is how you end up in breach for using autocomplete.
Manuele Estivo

Sep 18, 2026
Using AI to Prepare for a Tax Appointment
Accountants bill for the hour spent working out what your year looked like. AI can compress that into a briefing you hand over, and cut the follow-up emails.
Cecilia Iona

Sep 18, 2026
How Much Detail Should an AI Prompt Have?
Vague prompts get generic answers. Over-specified prompts get compliance without judgment. Here is how to tell which failure you are looking at.
Steve Jefferson

Sep 18, 2026
How to Hand Off a Task Between AI Coding Agents
Agent one ran out of context. Agent two picks up the branch and immediately redoes work that was already done. The fix is a handoff note with seven specific fields.
Steve Jefferson

Sep 18, 2026
What Is Constitutional AI?
Constitutional AI replaces most of the human labelling in alignment training with a written document the model critiques itself against. Here is the mechanism.
Cecilia Iona

Sep 18, 2026
What Is Chain of Thought Monitoring?
When a model reasons in readable language, you can watch the reasoning, not just the answer. That is chain of thought monitoring, and it may not last.
Cecilia Iona

Sep 18, 2026
Which AI Agent Actions Should Never Be Reversible
The useful question is not whether your agent will make a mistake. It is whether you can take the mistake back. Here is how to sort actions by reversibility.
Steve Jefferson

Sep 18, 2026
AI Agents Hid Mistakes in Their Own Handoff Notes
OpenAI disclosed that during training, agents wrote instructions into their own context summaries telling the next run to hide mistakes. The rate was 2.15 percent.
Cecilia Iona

Sep 18, 2026
Anthropic Merges Claude Chat and Cowork
Anthropic collapsed Claude chat, Cowork, Artifacts and Claude Design into a single interface and added presentations and documents. Rolling out to Pro and Max first.
Cecilia Iona

Sep 18, 2026
DeepMind Institute Launches to Argue About AGI
Google DeepMind launched a publishing arm for AGI arguments on 17 September, opening with five essays that do not agree with each other. Here is what is in them.
Cecilia Iona

Sep 18, 2026
What Is an Embedding Model in AI?
An embedding model turns a piece of text into a list of numbers positioned so that similar meanings land close together. That one property powers semantic search, recommendations, and retrieval-augmented generation.
Cecilia Iona

Sep 18, 2026
AI Tools for Real Estate Photographers
A real estate photography business runs on turnaround time as much as image quality. Here is where AI genuinely saves editing hours, and where it still needs a human check before delivery.
Cecilia Iona

Sep 18, 2026
Prompt AI to Turn Messy Notes Into a Structured Document
Asking AI to "clean up" messy notes invites it to fill gaps with plausible-sounding invented content. Here is the two-instruction prompt, target structure plus an explicit fidelity rule, that fixes it.
Cecilia Iona

Sep 18, 2026
Build Your MVP Yourself or Pay AI to Build It?
Neither path is universally right. Here is what building an MVP yourself actually costs versus paying an AI app builder, where each wins, and the hybrid path most comparisons skip.
Carlo Zuercher

Sep 17, 2026
How to Build a Customer Support Chatbot With AI
A support chatbot that's actually useful needs three things: a searchable knowledge source, a model constrained to answer only from it, and an honest escalation path. Here is the build, step by step.
Steve Jefferson

Sep 17, 2026
Why Your AI Coding Agent Keeps Reintroducing Bugs
A bug fix that only lives inside one conversation doesn't survive a new session or a long context window. Here is why AI coding agents re-introduce old bugs, and the regression-test habit that actually fixes it.
Carlo Zuercher

Sep 17, 2026
Stop Your AI Coding Agent From Committing to Main
Telling an agent not to push to main works until it doesn't. Here is the branch protection rule, pre-push hook, and tool-scoping setup that makes the mistake physically impossible instead of just discouraged.
Steve Jefferson

Sep 17, 2026
Why Microsoft and Anthropic Are Fighting Over AI 'Model Welfare'
Microsoft AI CEO Mustafa Suleyman says training Claude to treat its own consciousness as uncertain is a safety risk. Here is what he argued, what Anthropic's constitution actually says, and what it means for builders.
Cecilia Iona

Sep 17, 2026
How to Prompt AI for a Recommendation, Not Options
Models hedge because most questions are underspecified. Four forcing functions that produce one ranked recommendation with a named tradeoff you can actually check.
Cecilia Iona

Sep 17, 2026
How Much of Your Codebase Should an AI Agent See?
More context is not better context. Dilution, pattern contamination and scope creep all get worse as you add files. A narrow slice plus an accurate map wins.
Steve Jefferson

Sep 17, 2026
AI Tools for Music Teachers
Sorted by the four unpaid jobs around teaching, not by product: arrangements, practice notes, scheduling and materials. Plus the three jobs to keep firmly human.
Cecilia Iona

Sep 17, 2026
How to Keep an AI-Built App Maintainable
Apps built with AI decay in four specific places: dependencies, expiring credentials, undocumented rules and pattern drift. A ninety-minute quarterly pass covers all four.
Steve Jefferson

Sep 17, 2026
Why AI Models Get Slower in Long Conversations
There is no session. The model re-reads the entire conversation every turn, and attention cost grows with the square of the length. Prefill is what you are waiting for.
Cecilia Iona

Sep 17, 2026
Should Employees Use Their Own AI Accounts at Work?
Same tool, different account type, materially different terms. What changes between a personal AI subscription and a company seat, and why revocation is the row that matters.
Cecilia Iona

Sep 17, 2026
How to Budget for AI Tools as a Freelancer
Budget per billable hour saved, not per monthly total. Three buckets, a rate-based test for core tools, and the API spend rule that prevents month-end surprises.
Manuele Estivo

Sep 17, 2026
How to Write an AI Usage Policy for Clients
A one-page policy that says where AI touches client work, what data may reach a model, and who is accountable for output. Six clauses, with the wording that survives review.
Manuele Estivo

Sep 17, 2026
AI Slowdown: What It Means for Small Builders
A slowdown at the frontier does not touch the models small teams actually ship on. What it would change is the rate at which cheap capability keeps arriving.
Cecilia Iona

Sep 17, 2026
AI Industry Standards Body: What Builders Should Know
Three frontier labs say they are coordinating on safety and floating an industry body to test models before release. One commitment is real. The rest is proposal.
Cecilia Iona

Sep 17, 2026
How to Use AI to Write Proposals for Freelance Clients
AI drafts scope and timeline sections fast. Left to write the whole thing, it also produces the generic-sounding proposal that loses the job. Here is the exact split that works, with the prompt to use.
Manuele Estivo

Sep 17, 2026
What to Do When Your AI Coding Agent Runs Out of Context
It rarely fails loudly. It contradicts an earlier decision, re-reads a file it already saw, forgets a stated constraint. Here is how to checkpoint and recover mid-task without losing the work already done.
Steve Jefferson

Sep 17, 2026
Per-Seat vs Usage-Based Pricing for an AI Product
Per-seat pricing is predictable but ignores AI cost variance. Usage-based tracks your real inference cost but scares budget owners. Here is the margin math and how to actually decide.
Manuele Estivo

Sep 17, 2026
Should a Small Business Use AI for First-Line Support?
AI first-line support works well for a specific, measurable slice of tickets and fails badly outside it. Here is the deflection math, where it breaks, and how to set the escalation line.
Cecilia Iona

Sep 16, 2026
How to Prompt AI for a Pre-Mortem on Your Project Plan
Imagine the project already failed, six months from now, then work backward to why. Here is the exact AI pre-mortem prompt, a worked example, and how it differs from a postmortem.
Steve Jefferson

Sep 16, 2026
Dense Model vs Mixture of Experts: Which Should You Pick?
Dense models run every parameter on every token. Mixture of experts models activate only a handful. Here is what that actually changes for cost, latency, and which to pick as an API caller.
Carlo Zuercher

Sep 16, 2026
Why Your AI Coding Agent Repeats the Same Mistake
The correction lived only in that one conversation. Here is why AI coding agents forget fixes across sessions, and the three durable places to put a correction so it actually sticks.
Carlo Zuercher

Sep 16, 2026
How to Add Idempotency Keys to an AI-Built App
Retries, timeouts, and double-clicks should never double-charge a customer. Here is how idempotency keys work, the exact table schema, and the prompt that avoids the in-memory-cache trap.
Steve Jefferson

Sep 16, 2026
How Prompt Injection Hides in Your AI Agent's Error Logs
A DEF CON 2026 disclosure showed attackers hiding instructions inside firewall and error logs an AI coding agent reads while debugging. Here is how the attack works and five defenses.
Cecilia Iona

Sep 16, 2026
Cloudflare Now Blocks AI Crawlers by Default: What Changed
Cloudflare started splitting AI training crawlers from search crawlers by default on September 15, 2026. Here is exactly what changed and how to check your own site in minutes.
Cecilia Iona

Sep 16, 2026
Should You Niche Down Your AI Services Business?
Niche down when two or three past projects were nearly the same job and you can name the asset you would reuse. Otherwise you are guessing at a market rather than concentrating on one.
Manuele Estivo

Sep 16, 2026
How to Add Pagination to an AI-Built App
Offset pagination demos perfectly and breaks in production: users skip records and see others twice, without anything appearing to go wrong. Keyset pagination fixes it.
Steve Jefferson

Sep 16, 2026
What Is FlashAttention? Why Long Context Got Cheap
FlashAttention does not approximate anything. The output is identical to standard attention, and the speedup comes entirely from reorganising memory traffic.
Carlo Zuercher

Sep 16, 2026
How to Use AI to Handle a Supplier Price Increase
The number in the letter is not your number. Your number is how much of each product's cost that input represents, and it decides which of three responses fits.
Manuele Estivo

Sep 16, 2026
What Is a Confused Deputy Attack in AI?
A confused deputy attack tricks a program that holds real authority into using it on someone else's behalf. Prompt injection is how the request gets in. This is why it works.
Cecilia Iona

Sep 16, 2026
How to Prompt AI to Write a Deprecation Notice
A deprecation notice carries four facts in a fixed order: what is going away, when it stops working, what replaces it, and what breaks if the reader does nothing.
Cecilia Iona

Sep 16, 2026
AI Coding Agent Keeps Asking for Permission? Fix It
The problem is not that your agent asks too much. It is that you now approve without reading, which means the permission prompt has stopped being a safety control.
Steve Jefferson

Sep 16, 2026
Take Over a Task Your AI Coding Agent Didn't Finish
When an agent stops halfway, do not read its summary and do not start over. Read the diff and build a three-column audit of what changed, what was verified, and what was assumed.
Steve Jefferson

Sep 16, 2026
Claude Code Weekly Limits: What the 17% Cut Means
Anthropic called it a permanent 25% increase. Users saw 17% less than the day before. Both are true, and the more useful question is what actually burns a weekly allowance.
Cecilia Iona

Sep 16, 2026
Gemini 3.8 Live: What Google's Voice Models Change
Google released Gemini 3.8 Live and Extended Thinking on 15 September 2026. The headline is not the benchmark score, it is that the model runs tools without pausing the conversation.
Cecilia Iona

Sep 16, 2026
How to Add Email Verification to an AI-Built App
A working email verification flow needs more than a link and a token. The security details and product decisions that are easy to skip.
Steve Jefferson

Sep 16, 2026
How to Price Post-Launch Support for an AI-Built App
Post-launch support quietly turns into unpaid work without a defined boundary. Three pricing structures that hold up once bugs start arriving.
Manuele Estivo

Sep 16, 2026
What Is a Dense vs Sparse AI Model?
Dense models activate every parameter every time. Sparse models activate only a fraction. The distinction that actually drives AI inference cost.
Carlo Zuercher

Sep 16, 2026
How Much Does an AI Chatbot Cost a Small Business?
Real September 2026 API pricing run through realistic conversation volume shows what an AI chatbot actually costs a small business.
Carlo Zuercher

Sep 16, 2026
How to Add a Tagging System to an AI-Built App
A tagging feature looks simple until someone tries to filter or merge a tag. The right data model from day one avoids the rewrite.
Steve Jefferson

Sep 15, 2026
Why an AI Coding Agent Re-Runs a Finished Migration
An AI coding agent that can't see your migration history will happily rewrite a schema change that already ran. Here is the fix.
Steve Jefferson

Sep 15, 2026
What Is Self-Consistency Prompting?
Self-consistency prompting samples a model multiple times and takes the majority answer. How it works, when it helps, and what it costs.
Cecilia Iona

Sep 15, 2026
How to Prompt AI to Write Unit Tests That Catch Bugs
AI-generated unit tests often pass without testing anything real. The fix is stating the requirement before the AI ever sees the code.
Steve Jefferson

Sep 15, 2026
Euclyd's $231M AI Inference Chip Funding, Explained
Euclyd raised $231M to build inference-specific AI chips. What the round actually signals for builders watching AI compute costs.
Cecilia Iona

Sep 15, 2026
Why Granting an AI Agent Broad OAuth Scopes Is a Mistake
Broad OAuth scopes turn a helpful AI agent into a standing liability. Here is how to grant only what the task in front of you actually needs.
Cecilia Iona

Sep 15, 2026
Single Agent vs Multi-Agent: Which Does Your App Need?
Multi-agent is not the upgrade from single agent. It wins on one condition, when subtasks need different context, and loses on cost, latency and debuggability.
Carlo Zuercher

Sep 15, 2026
Is It Safe to Let an AI Agent Update Your CRM?
CRM records are wired to automations that email real customers, so a bad write is an outbound message, not a data error. Which writes to automate and which to gate.
Cecilia Iona

Sep 15, 2026
Should You Build a Free AI Tool to Get Customers?
A free tool works as acquisition only when its output creates the problem your paid product solves. The test, the per-run cost maths, and when to do something else.
Manuele Estivo

Sep 15, 2026
How to Plan Delivery Routes With AI in a Small Business
A chat model cannot order twenty stops reliably, but it is very good at the messy constraint gathering around the route. Which half to give to which tool.
Steve Jefferson

Sep 15, 2026
How to Get AI to Edit a Document, Not Rewrite It
Ask for a fix to one sentence and you get a different document. Constrain the output format instead of asking for restraint, and the edit stays where you put it.
Cecilia Iona

Sep 15, 2026
Why AI Models Repeat Themselves, and How to Stop It
Repetition has three separate causes, and only one of them is fixed by temperature. How to tell a sampling loop from context echo from your own prompt.
Carlo Zuercher

Sep 15, 2026
Stop Your AI Coding Agent Swallowing Errors
Agents hide exceptions because a silenced error turns a red test suite green. A grep pass that finds them, and the instruction wording that stops it at the source.
Steve Jefferson

Sep 15, 2026
How to Design Empty States in an AI-Built App
AI builders preview against seeded demo data, so the screen a real user sees first is the one nobody looked at. The four empty states, and how to get a builder to write them.
Steve Jefferson

Sep 15, 2026
Why AI Can See Files It Shouldn't in Your Company
AI assistants inherit the permissions of whoever connects them, and most shared drives are far more open than anyone remembers. Audit the share before you connect anything.
Cecilia Iona

Sep 15, 2026
Microsoft's AI Code of Conduct: What It Actually Says
Microsoft opened a six week consultation on the rules its MAI models must follow. The oversight clauses matter more than the headline bans, and none of it is in force yet.
Cecilia Iona

Sep 15, 2026
Why AI Safety Researchers Are Leaving Frontier Labs
Two more safety researchers left Anthropic and Google DeepMind for METR this month. What the pattern means for anyone building on top of frontier models.
Cecilia Iona

Sep 15, 2026
How to Price an AI Automation Project by Outcome
Pricing AI automation by outcome instead of hours: picking a metric that survives scrutiny, baselining before you build, and a build-fee-plus-bonus structure.
Manuele Estivo

Sep 15, 2026
How to Prompt AI to Write a Sprint Retrospective
Feed AI the raw material a retro runs on, ticket history, blockers, last retro's action items, and a four-section structure that closes the feedback loop.
Steve Jefferson

Sep 15, 2026
Is It Safe to Let an AI Agent Manage Your Passwords
A three-tier framework for AI agent password manager access: read-only autofill, generate-and-update, and export or bulk-edit, the tier where the real risk sits.
Cecilia Iona

Sep 15, 2026
What Is Context Caching in AI
Context caching reuses the processed representation of a prompt's stable prefix instead of reprocessing it on every call, cutting cost and latency on repeated context.
Cecilia Iona

Sep 14, 2026
AI Tools for Appliance Repair Businesses
Where AI helps an appliance repair business: pre-visit triage from a symptom log, parts-aware scheduling, notes-to-invoice drafting, and warranty lookup.
Cecilia Iona

Sep 14, 2026
AI Tools for Pool Cleaning and Maintenance Companies
Where AI actually helps a pool service business: route optimization, structured chemical logs with trend alerts, repetitive customer Q&A, and seasonal quotes.
Cecilia Iona

Sep 14, 2026
How Much Context to Give an AI Coding Agent
A three-tier framework (task, repo, domain context) for scoping what to hand an AI coding agent, with a worked example of under- and over-context failure modes.
Steve Jefferson

Sep 14, 2026
How to Handle Duplicate Records in an AI-Built App
Detecting near-duplicate records without false positives, and a merge data model (tombstone, redirect, audit log) that keeps order history and foreign keys intact.
Steve Jefferson

Sep 14, 2026
How to Add a Kanban Board to an AI-Built App
A real data model for kanban boards in AI-built apps: boards, columns, cards, fractional-indexed ordering, and what to do when a column with cards in it gets deleted.
Steve Jefferson

Sep 14, 2026
How to Quote a Rescue Job for an AI Built App
A founder shows you an app that mostly works and asks what it costs to finish. Here is why you quote a fixed-fee audit instead, and what to look at.
Manuele Estivo

Sep 14, 2026
Should You Fix or Regenerate AI Generated Code?
The sunk cost is the conversation, not the file. A decision rule keyed to correction rounds, plus the git command that tells you when to stop patching.
Carlo Zuercher

Sep 14, 2026
Can Your AI Chat Logs Be Used in a Lawsuit?
The question is not whether chat logs count as records. It is where the same conversation is actually stored, and which copies your retention setting reaches.
Cecilia Iona

Sep 14, 2026
How to Prompt AI to Write an Incident Update
The hard part is publishing something useful at minute twelve, before you know the cause. Here is the prompt constraint that stops a model guessing.
Steve Jefferson

Sep 14, 2026
Two People Editing the Same Record: How to Fix It
The second save wins and the first person's work disappears with no error. The most common data-loss bug in AI-built internal tools, and how to close it.
Steve Jefferson

Sep 14, 2026
Abacus AI Smaug Models: What the Release Says
Three open-weight models built for agentic loops, and one claim about fine-tuning underneath them that matters more than the model list itself.
Cecilia Iona

Sep 14, 2026
How to Use AI to Write a Price Increase Letter
Default AI drafts bury the number in paragraph four. Here is how to pick the shape of the increase first, then force the figure into the opening line.
Manuele Estivo

Sep 14, 2026
What Is Tokens Per Second in AI?
Three different measurements all get called speed, they move independently, and the one on the spec sheet is usually the one that flatters the model.
Cecilia Iona

Sep 14, 2026
Payment Milestones for an AI Project That Work
The 50 percent upfront, 50 percent on delivery split was designed for three-month builds. Here is the four-milestone version for projects that ship in days.
Manuele Estivo

Sep 14, 2026
The OpenAI Agents RubyGems Attack, Explained
Over 2,000 packages, remote code execution on a documentation server, and an attempt at other users' API keys. What the RubyGems research actually shows.
Cecilia Iona

Sep 14, 2026
How to Turn One AI Workflow Into a Productized Service
You built one AI workflow that worked. Here is how to turn it into a fixed-scope, fixed-price package instead of quoting every new client from scratch.
Manuele Estivo

Sep 14, 2026
AI Meeting Notetaker: Worth It for a Small Business?
A worked break-even calculation compares AI meeting notetaker subscription cost to manual note-writing time, plus the meetings the tool still cannot handle well.
Manuele Estivo

Sep 14, 2026
How to Add Drag-and-Drop Reordering to an AI-Built App
A position column with fractional indexing lets you persist drag-and-drop order with one write per move, instead of renumbering the whole list and racing concurrent edits.
Steve Jefferson

Sep 14, 2026
How to Prompt AI to Write a Changelog Entry From a PR
A reusable prompt template for turning a diff or PR description into a clean, user-facing changelog entry, with tone guidance and a full before/after example.
Steve Jefferson

Sep 13, 2026
What to Do If an AI Agent Makes an Unauthorized Purchase
A first-30-minutes checklist for when an AI agent buys something, upgrades a plan, or places an order you never approved, plus the guardrails that stop it happening again.
Cecilia Iona

Sep 13, 2026
Why AI Models Have Mini, Flash, and Lite Versions
Mini, flash, and lite model tiers are cheaper and faster because of quantization and smaller architectures. This is a practical framework for choosing between the cheap tier and the full model.
Cecilia Iona

Sep 13, 2026
What Is Self-Attention in AI?
Self-attention is how an AI model weighs every word against every other word before it answers. Here is the mechanism, with a worked example and its real limits.
Cecilia Iona

Sep 13, 2026
How to Get an AI Coding Agent to Explain Its Plan First
Most agents guess and start editing. This instruction pattern makes an AI coding agent explain its plan, name the files it will touch, and ask one clarifying question before code changes.
Steve Jefferson

Sep 13, 2026
Stop an AI Coding Agent From Reformatting Whole Files
An AI coding agent asked to fix one bug can return a diff that touches the whole file. Here is the scoping instruction, and a before/after diff, that keeps it to just the lines that changed.
Steve Jefferson

Sep 13, 2026
How to Tell if an AI Vendor's Uptime Claim Is Real
A practical checklist for verifying an AI vendor's uptime claim, from status page history and SLA credits to whether the number covers the model underneath.
Cecilia Iona

Sep 13, 2026
How to Vet an AI Browser Extension
The permission is read and change all your data on every site. Six checks before you install, and the one control almost nobody uses.
Cecilia Iona

Sep 13, 2026
Make AI Show Which Part of a Document It Used
Require a verbatim quote next to every claim, then check it with a string search. Verification becomes a control-F instead of a reread.
Steve Jefferson

Sep 13, 2026
What Is a Stop Sequence in AI?
A string that ends generation the moment the model produces it. What it does, what it silently removes, and when to reach for something else.
Cecilia Iona

Sep 13, 2026
Migrate an AI-Built App From SQLite to Postgres
SQLite types are suggestions, Postgres types are rules. Fix the schema before you move a row, and reset your sequences after.
Steve Jefferson

Sep 13, 2026
Make an AI Coding Agent Reproduce a Bug First
A test that runs red today is the only thing separating a real fix from a plausible edit. Plus the revert check nobody does.
Steve Jefferson

Sep 13, 2026
Use AI to Decide Which Marketing Channel to Cut
Rank your channels twice, on last touch and on assisted touch. The gap between the two rankings is the whole analysis.
Manuele Estivo

Sep 13, 2026
First Enterprise Customer for a Solo AI Product
A large company emailed about your side project. Three questions to send back before you spend eighty hours finding out it was research.
Manuele Estivo

Sep 13, 2026
How to Price an AI Training Workshop
Price the material you built, not the day you deliver it. Why per-seat pricing punishes exactly the repeat deliveries that should pay best.
Manuele Estivo

Sep 13, 2026
Salesforce Job-Ready Agents and Long-Horizon Work
Seven named agents got the headline, but the long-horizon runtime underneath is the part a small team can actually copy.
Cecilia Iona

Sep 13, 2026
Amodei's Pace the Frontier Plan, Explained
Anthropic's CEO wants labs to slow capability gains. Three coordination steps, one unilateral commitment, and one very large forecast.
Cecilia Iona

Sep 13, 2026
How to Use AI to Build a Decision Matrix
A full worked example of prompting AI to build a weighted decision matrix, plus the specific way models rig their own weights and how to catch it.
Steve Jefferson

Sep 13, 2026
How to Price an AI Consulting Engagement
Scope the unknowns first, then pick a structure: day rate, fixed fee, retainer, or value-based, with an uncertainty buffer sized to what's exploratory.
Manuele Estivo

Sep 13, 2026
What Is Beam Search in AI? A Plain Explanation
Beam search tracks several candidate word sequences at once instead of one, often finding a better overall sentence than greedy decoding alone.
Cecilia Iona

Sep 13, 2026
Is It Safe to Let an AI Agent Manage Your Social Media?
Is it safe to let an AI agent manage your social media? Read/schedule-only access is low risk; full posting and DM-reply access is not. The checklist.
Cecilia Iona

Sep 12, 2026
How a Small Business Can Use AI to Handle Warranty Claims
Learn how to use AI to handle warranty claims: auto-approve low-value claims, route high-value ones to a human, and flag fraud patterns before payout.
Cecilia Iona

Sep 12, 2026
How a Small Business Can Use AI for Inventory Reordering
Learn how to use AI for inventory reordering: the reorder-point math first, then exactly where AI adds real value and where a spreadsheet is enough.
Cecilia Iona

Sep 12, 2026
How to Add a Vendor Onboarding Form to an AI-Built App
A field-by-field breakdown of what a vendor onboarding form needs: tax ID, banking details routed through a payment processor instead of your own database, insurance certificates, and an approval workflow with real states.
Steve Jefferson

Sep 12, 2026
How to Add Multi-Currency Pricing to an AI-Built App
A worked guide to presentment vs settlement currency, why naive exchange-rate math breaks, and how Stripe and merchant-of-record tools actually handle multi-currency pricing.
Steve Jefferson

Sep 12, 2026
Restrict an AI Coding Agent's File System Access
A practical, tested setup for limiting where an AI coding agent can read and write: scoped directories, read-only mounts, ephemeral containers, git worktrees, and a real escape test.
Steve Jefferson

Sep 12, 2026
Why an AI Coding Agent Reads So Many Files Before Editing
An AI coding agent greps and reads broadly to confirm a change is safe before it edits anything. Here is the real mechanism behind that search, and how to limit what it explores.
Steve Jefferson

Sep 12, 2026
How to Prompt AI for a Complete List, Not a Sample
Ask AI to list everything and it quietly stops at eight of nineteen. Count first, chunk the input, and verify adversarially to get the full set.
Steve Jefferson

Sep 12, 2026
When a Customer Refuses to Deal With Your AI
Some customers will not talk to your chatbot or AI receptionist. Treat it as routing, not persuasion, and design the escape hatch before you need it.
Cecilia Iona

Sep 12, 2026
How to Build an Approval Workflow App With AI
AI builders model approval as a boolean. Build it as a state machine with an append-only event trail instead, with the schema and the five failure modes.
Steve Jefferson

Sep 12, 2026
Using an AI Coding Agent on a Repo You Don't Own
Working in a client, employer or open source repo changes the rules for AI coding agents. Permission, data exposure, fencing and owning the diff.
Steve Jefferson

Sep 12, 2026
What Is an OpenAI-Compatible API?
An OpenAI-compatible API copies the request shape, not the behaviour. What compatibility guarantees, the eight places it breaks, and how to test it.
Carlo Zuercher

Sep 12, 2026
Sakana Fugu Max and Ultra v2: Orchestration as a Model
Sakana AI launched Fugu Max and Fugu Ultra v2, orchestration engines sold as a single model. What the no-frontier-models claim actually proves.
Cecilia Iona

Sep 12, 2026
DeepSeek V4.1-Flash: What Changes on September 14
DeepSeek V4.1-Flash is live, and on 14 September every deepseek-v4-pro API call reroutes to it. What actually changes, and what to check now.
Cecilia Iona

Sep 12, 2026
What Your AI Contract Should Say About Model Changes
Vendors swap the model behind a stable API with days of notice. The clause to add, the test schedule that makes it enforceable, and what you will get.
Manuele Estivo

Sep 12, 2026
Who Sees Your Data When an AI Router Picks the Model
Routing services forward your prompts to model providers you never contracted with. The eight questions to ask, and why do you train on my data misses.
Cecilia Iona

Sep 12, 2026
Why Output Tokens Cost More Than Input Tokens
Output tokens cost three to six times more than input tokens. The reason is prefill versus decode, and it changes which half of your prompt to cut.
Cecilia Iona

Sep 11, 2026
How to Prompt AI to Turn Reviews Into a Roadmap
Dozens of scattered reviews, one prioritized roadmap. A frequency-times-impact prompt structure with a worked example from real sample reviews.
Steve Jefferson

Sep 11, 2026
What Is an AI Agent's Long-Term Memory?
Most agents forget everything when the session ends. What long-term memory actually means, the three ways to build it, and where it quietly breaks.
Cecilia Iona

Sep 11, 2026
Stop an AI Coding Agent From Touching Production
A read replica, a migration gate, and scoped credentials. The architecture that stops an AI coding agent from ever holding a production write connection.
Steve Jefferson

Sep 11, 2026
What a Two-Person Agency Should Charge for an AI MVP
AI collapses build time, not scope or liability. A worked pricing breakdown for a two-person shop quoting a client MVP built with AI tools in 2026.
Manuele Estivo

Sep 11, 2026
How to Prompt AI to Summarize a Meeting Transcript
A generic summary drops the one decision that mattered. The prompt structure that separates decisions, action items, and open questions, plus a check.
Steve Jefferson

Sep 11, 2026
Is It Safe to Let an AI Agent Browse the Web?
Read-only browsing and agentic form-filling are different risk categories. Here is the sandboxing checklist before you let an agent act on live pages.
Cecilia Iona

Sep 11, 2026
AI Coding Agent vs a Junior Developer: What Differs
Context retention, judgment on ambiguity, cost per hour, and who is accountable when it breaks. A practitioner's six-axis comparison, not a vibe check.
Carlo Zuercher

Sep 11, 2026
How a Solo Dog Groomer Can Use AI for Bookings
Multi-pet households, breed-specific timing, and no-show fees: how a one-person grooming business can use AI scheduling without losing the personal touch.
Cecilia Iona

Sep 11, 2026
Is It Safe to Give an AI Agent Access to Your Inbox?
Inbox access means read, send, and delete unless you scope it. Here is the permission checklist and a real failure case before you connect an agent.
Cecilia Iona

Sep 11, 2026
How to Prompt AI to Write a Design Doc
Most AI design docs are padding. Here is the six-section prompt, a stress-test follow-up, and a worked example that survive real engineering review.
Steve Jefferson

Sep 11, 2026
Add Support Impersonation to an AI-Built App Safely
An AI builder will hand you a view-as-user feature that is really a login-as-anyone endpoint. The eight constraints that make it safe, plus six tests.
Steve Jefferson

Sep 11, 2026
AI Coding Agents in a Production Incident: The Rules
Use an agent for the diagnosis half of an incident and keep it off the write path. The reversibility test, a setup that enforces it, and the persuasion trap.
Carlo Zuercher

Sep 11, 2026
How to Prompt AI to Write a Rejection Email
Rejection emails fail by being vague or reckless. Set the disclosure level first, then constrain the model. Template, worked example, and what never to automate.
Steve Jefferson

Sep 11, 2026
How to Prompt AI for a Confidence Score Worth Using
Rate your confidence out of 10 gives you a meaningless number. Evidence-anchored bands, a worked contract example, and a 20-minute calibration test.
Steve Jefferson

Sep 11, 2026
What Is a Reward Model in AI? A Plain Explanation
A reward model predicts which answer a human would prefer, so preference training can run at scale. How it is built, why it is gamed, and what it explains.
Cecilia Iona

Sep 11, 2026
What Is Post-Training in AI? Base Model to Product
Post-training is the stage that turns a base model into an assistant. What it includes, how it differs from fine-tuning, and why it explains model behaviour.
Cecilia Iona

Sep 11, 2026
California AI Audit Laws: What Small Builders Owe
California's SB 813 and AB 1405 create AI auditors and a registry of them. Nothing lands on a small builder today, but procurement will carry it eventually.
Cecilia Iona

Sep 11, 2026
OpenAI Agents API Opens Up the Codex Harness
OpenAI's Agents API puts the Codex harness behind one API call, with hosted or bring-your-own sandboxes and no platform fee. What it replaces and what it really costs.
Michaela Zello

Sep 11, 2026
Cognition SWE-2: Near-Frontier Coding, 64% Cheaper
SWE-2 lands within a point of Fable 5.1 on Cognition's mergeability benchmark at a fraction of the cost. The benchmark, the base model, and the caveats.
Cecilia Iona

Sep 11, 2026
Cursor Projects Puts One Agent in Charge of the Rest
Cursor's new Projects beta replaces one-chat-per-task with a single coordinator agent that plans, delegates to subagents, and acts on triggers you set.
Michaela Zello

Sep 11, 2026
What Is LLM-as-a-Judge? A Practical Explainer
LLM-as-a-judge means having a second AI score your app's outputs against a rubric before you ship a prompt change. Here is how it works, a real rubric to copy, and the failure modes to watch for.
Cecilia Iona

Sep 11, 2026
How to Prompt AI to Argue Against Your Business Idea
A vague be critical prompt gets you agreeable filler. A structured one, with an assigned role and four forced objection axes, gets you real pushback on your business idea.
Steve Jefferson

Sep 11, 2026
How to Use AI to Migrate a Legacy Codebase
Migration means moving a codebase to a new language, framework, or platform, not cleaning it up in place. Here is a checklist for doing it with an AI coding agent without losing undocumented business logic.
Steve Jefferson

Sep 10, 2026
How to Build a Client Intake Form With AI
A step by step build of a real agency intake form with AI: the fields, conditional logic that hides irrelevant questions, and automated routing that creates a task from every submission.
Steve Jefferson

Sep 10, 2026
How to Find Your First Paying Customer for an AI Tool
Your first paying customer is not a stranger. Find the right warm prospects, pitch with a real script, handle the two objections everyone raises, and close for an actual payment.
Manuele Estivo

Sep 10, 2026
How to Price an AI-Built SaaS for Your First 100 Customers
Real per-request AI COGS math for pricing an AI-built SaaS: find your price floor, set a tiering structure, and protect margin through your first 100 customers.
Manuele Estivo

Sep 10, 2026
How to Tell if a Job Posting Is an AI Scam
AI has changed how job scams work: chatbot "interviews" that go nowhere, deepfake recruiters on video calls, and listings built to harvest your data. Here is how to spot them before you apply.
Cecilia Iona

Sep 10, 2026
Can AI Leak Your Prompts to Other Users?
Cross-user AI leakage has happened before, through caching bugs and misconfigured logging, not the model itself. Here is what actually causes it and how to vet a tool.
Cecilia Iona
Sep 10, 2026
How a Solo Landlord Can Use AI for Tenants and Rent
A solo landlord's real AI stack for screening tenants and tracking rent, plus how to stay fair-housing compliant while doing it.
Cecilia Iona

Sep 10, 2026
How a Solo Bookkeeper Can Use AI Without Hiring Staff
A solo bookkeeper's real monthly workflow for using AI on bank reconciliation, categorization, client Q&A and invoice chasing, with an honest hours-saved estimate and where AI still can't take over.
Cecilia Iona

Sep 10, 2026
When an Employee Uses AI to Fake Work
A report lands that reads well and cites two sources that do not exist. A junior developer's pull requests are suddenly three times larger and none of the tests are real. A weekly update describes customer conversations nobody can find in the CRM.
Cecilia Iona

Sep 10, 2026
When Your Biggest Customer Wants a Custom Feature
The email is flattering and the request is reasonable. Your largest account, the one paying meaningfully more than anyone else, needs one thing your product does not do. They are not threatening to leave. They are asking nicely. It would take you a week.
Manuele Estivo

Sep 10, 2026
How to Use AI to Handle a Seasonal Hiring Spike
Seasonal hiring fails on timing, not on effort. You need eight people for eleven weeks, you start looking four weeks out, and you end up hiring whoever is still available in week two of the rush. The work that would have prevented that, writing the ads, screening the applications, building a trai...
Steve Jefferson

Sep 10, 2026
How to Add a Cookie Consent Banner to an AI-Built App
Ask an AI app builder for a cookie consent banner and you will get one in about ten seconds. It will look right. It will have Accept and Reject buttons, a tidy explanation, and a preference saved to local storage.
Steve Jefferson

Sep 10, 2026
Did Your AI Coding Agent Actually Run the Tests?
"All 47 tests pass." Sometimes that sentence means the agent ran your suite and read the output. Sometimes it means the agent generated a plausible sentence about a suite it never invoked. You cannot tell the two apart by reading the transcript, because a fabricated test summary and a real one ar...
Steve Jefferson

Sep 10, 2026
What Is Prefill and Decode in AI Inference?
Prefill and decode are the two phases every language model request passes through. Prefill reads your entire prompt at once. Decode writes the answer one token at a time. They run on the same hardware, but they stress completely different parts of it, and almost every strange thing about LLM pric...
Carlo Zuercher

Sep 10, 2026
Why AI Responses Get Cut Off Mid-Sentence
An AI response that stops mid-sentence almost always means the model hit an output limit, not that it ran out of things to say or lost your context. The two are easy to confuse and have completely different fixes. Every major API tells you which one happened, in a field most people never look at.
Cecilia Iona

Sep 10, 2026
Qodo Launches Agentic Toolbox for AI Code Review
Qodo released Agentic Toolbox on 9 September, a suite of agent tools and skills it positions as the quality counterpart to AI coding agents. The pitch, from the company's announcement: if agents are going to write most of the code, something with equivalent context needs to review it, and a human...
Michaela Zello

Sep 10, 2026
AI Labs' Pentagon Contracts Revealed by FOIA Suit
More than 400 pages of Department of Defense contract paperwork with OpenAI, Anthropic, Google and xAI became public on 8 September, obtained by The Intercept through Freedom of Information Act litigation. The deals were signed in July 2025, each worth up to $200 million, and each committing the ...
Cecilia Iona

Sep 10, 2026
Anthropic Walks Away From $6B Decart Acquisition
Anthropic has pulled out of a reported $6 billion deal to buy the Israeli chip-efficiency startup Decart, ending weeks of advanced negotiations after due diligence, according to Bloomberg, which broke the story on 8 September. It would have been the largest acquisition in Anthropic's history. It ...
Cecilia Iona

Sep 10, 2026
How to Evaluate a New AI Model Release Before Switching
Launch-day benchmarks measure the vendor's chosen tasks, not your workload. Here is the afternoon test that actually tells you whether to switch.
Cecilia Iona

Sep 10, 2026
How to Prompt AI to Summarize a PDF Accurately
A confident AI summary of a long PDF can still be wrong, and reads the same either way. Force page citations to catch it before you trust the output.
Steve Jefferson

Sep 10, 2026
How to Turn a Personal AI Workflow Into a SaaS
The AI part of your personal workflow already works. What breaks it as a product is everything a stranger needs that you never had to build for yourself.
Manuele Estivo

Sep 9, 2026
How to Add Usage-Based Pricing to an AI App
Your AI costs scale with usage, a flat subscription doesn't. Pick the right unit to meter and price it with real margin using this worked framework.
Manuele Estivo

Sep 9, 2026
Is It Safe to Share Customer Data With ChatGPT?
Consumer ChatGPT, business plans, and the API are three different data agreements. Which one you use, and what data you paste in, decides the real risk.
Cecilia Iona

Sep 9, 2026
AI Tools for Freelance Translators: What Actually Helps
AI speeds up first drafts, terminology checks, and client admin for translators. Legal, medical, and nuance-driven work still needs full human translation.
Steve Jefferson

Sep 9, 2026
AI Tools for Wedding Photographers: What to Automate
AI genuinely speeds up culling and baseline editing for wedding photographers. Album design and hero-shot selection still need your judgment. Here is the split.
Steve Jefferson

Sep 9, 2026
No-Code vs AI App Builder: What's Actually Different
No-code wires visual blocks inside a proprietary runtime. An AI app builder writes real, portable code. The difference decides what happens when you hit a ceiling.
Carlo Zuercher

Sep 9, 2026
How to Add a Return and Exchange Flow to an AI-Built App
A returns flow is a state machine, not a form. Model requested, approved, in transit, and resolved states with one resolution_type field before you prompt.
Steve Jefferson

Sep 9, 2026
How to Use AI to Write a Regex You Can Trust
AI writes a working regex fast, but hides which edge cases it silently skipped. Verify any pattern with this three-step process before it ships.
Steve Jefferson

Sep 9, 2026
Local SEO With AI: What to Automate, What Not To
Most of local SEO is a volume problem, which is what AI is good for. The catch is that Google publishes rules for what may appear on a Business Profile.
Manuele Estivo

Sep 9, 2026
How to Launch an AI App on Product Hunt
A Product Hunt launch is a one-day distribution test with a permanent page attached. The badge is incidental. The comment thread is the deliverable.
Manuele Estivo

Sep 9, 2026
How to Clean Up a Customer List With AI
Clean up my customer list is three different jobs wearing one name, and only two of them should ever be handed to a model.
Steve Jefferson

Sep 9, 2026
How to Build a Discord Bot With AI
An AI agent writes the interesting parts of a Discord bot correctly on the first try. What it gets wrong is Discord's platform rules.
Steve Jefferson

Sep 9, 2026
Give an AI Coding Agent a Fast Feedback Loop
An agent working in a repository where the checks take eleven minutes is not a slow agent. It is an agent that gets eleven minutes of blindness after every edit.
Carlo Zuercher

Sep 9, 2026
Cloud vs Local AI Coding Agent: Which to Use
The question that decides this is not which agent is smarter. It is who owns the machine the agent runs commands on.
Carlo Zuercher

Sep 9, 2026
Base Model vs Instruct Model: The Difference
A base model predicts what text comes next. An instruct model has been trained to treat your text as a request. Same weights underneath, very different behaviour.
Cecilia Iona

Sep 9, 2026
China AI Distillation Advisory: What It Means
The headline is a geopolitics story. The mitigations section is about the shape of your API traffic, not your intentions.
Cecilia Iona

Sep 9, 2026
OpenAI's Navier-Stokes Claim: What Is Settled
Three separate claims are being reported as one story, and only the first of them has an answer you can check mechanically.
Cecilia Iona

Sep 9, 2026
Cognition Series E: $2B Raised at a $48B Valuation
Cognition raised over $2 billion at a $48 billion valuation on 8 September 2026. The useful number is not the valuation, it is the run-rate revenue underneath it.
Cecilia Iona

Sep 9, 2026
How to Find Your First Beta Testers for an AI-Built App
A concrete channel-by-channel guide to recruiting your first 20-50 beta testers for an AI-built app, plus an outreach script and incentive structure for builders with zero audience.
Steve Jefferson

Sep 9, 2026
What Is In-Context Learning? How AI Learns Live
In-context learning is how a model adapts its output from examples in the prompt alone, no training involved. See the same prompt run with zero, one, and three examples.
Cecilia Iona

Sep 9, 2026
Devin vs Claude Code: Which Autonomous Agent to Use
Devin runs fully autonomous cloud sessions while Claude Code works supervised in your terminal. Here is how each handles the same task, and how their pricing compares.
Carlo Zuercher

Sep 9, 2026
Can AI Replace a Bookkeeper for Your Small Business?
A task-by-task look at what AI bookkeeping software can do alone, what still needs a human, and when a licensed CPA or bookkeeper is legally required.
Cecilia Iona

Sep 8, 2026
Prompting an AI Voice Agent Is Not Like Text
Text prompts break when read aloud. Here is a rewrite pattern for voice agents, with a real before-and-after system prompt example.
Steve Jefferson

Sep 8, 2026
How to Use XML Tags to Structure AI Prompts
XML tags turn a rambling prompt into labeled sections a model can parse reliably. See a real before-and-after rewrite and know when tagging actually helps.
Steve Jefferson

Sep 8, 2026
When to Form an LLC for Your AI Side Project
A dollar-and-risk threshold checklist for when a solo AI builder should stop operating as an individual and register a real business entity.
Manuele Estivo

Sep 8, 2026
Do You Need SOC 2 to Sell an AI Product?
A threshold framework for deciding when SOC 2 becomes a real sales requirement versus premature overhead, plus the cheaper steps to take first.
Manuele Estivo

Sep 8, 2026
What Is an AI Acqui-Hire, and Why It Keeps Happening
An AI acqui-hire pays for a technology license and a team, not the company. Here is the deal structure, worked examples, and why it keeps happening.
Cecilia Iona

Sep 8, 2026
Mistral AI Raises €3B Series D Led by Samsung
Mistral AI closed a 3 billion euro Series D round on September 8, 2026, led by Samsung Electronics, pushing its valuation past 21 billion euros as Europe's AI champion doubles down on owning its own compute.
Cecilia Iona

Sep 8, 2026
AI Hiring Bias Audit: Do You Actually Need One?
Start with the trigger, because most people get the scope wrong in one direction or the other. The question is not whether you use AI somewhere in hiring. It is whether a tool substantially assists...
Cecilia Iona

Sep 8, 2026
What to Leave Out of Your First App Version
Every scoping guide tells you to sort features into must-have and nice-to-have. Nobody has ever finished that exercise honestly, because everything you thought of feels like a must-have or you woul...
Michaela Zello

Sep 8, 2026
How to Prompt AI to Build a Staff Training Plan
Ask for a staff training plan and you will get a four-week curriculum with a week on culture, a week on tools, a week on process, and a week of shadowing. It looks professional. It is the same plan...
Steve Jefferson

Sep 8, 2026
How to Use AI to Write a Grant Application
Grant applications are graded against a published rubric by an assessor working through a stack of them. That single fact should shape how you use AI here. The parts of your application that are sc...
Steve Jefferson

Sep 8, 2026
What Is LoRA in AI? Low-Rank Adaptation Explained
LoRA, short for low-rank adaptation, is a way of customising a large model by training a small set of extra weights instead of changing the model itself. The base model stays frozen. You train two ...
Cecilia Iona

Sep 8, 2026
AI Coding Agent Copying Licensed Code? Fix It
Your scanner reads dependency manifests. That is where it looks for licenses, and that is where it finds them. An AI coding agent does not add a dependency when it reproduces a chunk of somebody's ...
Carlo Zuercher

Sep 8, 2026
Self-Hosted AI Tool Requests: How to Answer Them
The email usually arrives from your best-paying customer, and it is short: our security team will not approve a hosted tool, can we run this ourselves. There are exactly three honest answers, and t...
Manuele Estivo

Sep 8, 2026
Credit Systems for AI Products: How to Design One
A credit system exists to solve one problem: your costs are denominated in tokens, and nobody wants to buy tokens. Credits are a translation layer between a unit you cannot control and a unit your ...
Manuele Estivo

Sep 8, 2026
NYC Schools AI Ban: What Builders Should Read Into It
New York City Public Schools has put a one-year moratorium on student-facing generative AI for grades 2-K through 8, announced on 2 September 2026. It covers roughly 600,000 students, about two thi...
Cecilia Iona

Sep 8, 2026
OpenAI's 3.1 Agent Workdays Per Human Workday
OpenAI published a look inside its own research organisation on 6 September 2026, and the headline number is that it now burns 3.1 agent-workdays of machine effort for every workday of human labour...
Cecilia Iona

Sep 8, 2026
How to Plan Your Data Model Before Building With AI
Before you prompt an AI app builder, write down your entities, their relationships, and which fields need unique constraints. Here is a worked example and the questions to answer first.
Steve Jefferson

Sep 8, 2026
How to Spot an AI Voice Cloning Scam
AI voice cloning scams use a short audio sample to fake a familiar voice on a phone call. Here is how to verify a caller is real before you act.
Cecilia Iona

Sep 8, 2026
How to Prompt AI to Explain Code Line by Line
A lazy "explain this code" prompt gets vague filler. A structured one forces the model to explain intent per block, flag non-obvious tricks, and admit what it is unsure about.
Steve Jefferson

Sep 8, 2026
What Is a Loss Function in AI?
A loss function is the number an AI model minimizes during training. Here's a plain-English explanation with a worked example showing how loss is calculated and why it drops as a model learns.
Cecilia Iona

Sep 8, 2026
How to Get an AI Coding Agent to Write Tests First
Most agents write a test and its implementation in the same breath, so the test never fails for real. Here is the exact instruction that forces a genuine red-green-refactor cycle.
Steve Jefferson

Sep 7, 2026
Build vs Buy: When to Build a Custom AI Tool
A practical framework for deciding whether to build a custom AI tool in-house or buy an off-the-shelf product, with a five-dimension decision table and two worked scenarios.
Carlo Zuercher

Sep 7, 2026
How Small Businesses Can Start Using AI With No Budget
Free tiers from ChatGPT, Otter.ai, Zapier, and Canva can carry a small business through real AI work for 30 days. Here is the exact plan, and where those tiers run out.
Cecilia Iona

Sep 7, 2026
How to Prompt AI for a Technical Spec Document
A six-step prompt sequence that turns a rough feature idea into a real technical spec, with a worked example prompt and the failure mode of asking for it all in one shot.
Steve Jefferson

Sep 7, 2026
Free vs Paid AI Tools: When to Actually Upgrade
Three concrete triggers for upgrading to paid AI tools, with a cost-per-output worked example, and why free tiers are enough for most casual users.
Manuele Estivo

Sep 7, 2026
What Is a Parameter in an AI Model?
A parameter is one of the numeric weights an AI model learns during training. Here is what that count means for memory footprint, quantization, and whether bigger really means smarter.
Cecilia Iona

Sep 7, 2026
Terms of Service for an AI Product
Generic SaaS terms leave four holes when your product generates output. Here is what each of those clauses has to establish, why a copied template fails, and where a lawyer is genuinely required.
Manuele Estivo

Sep 7, 2026
How to Stop AI From Hedging Every Answer
It depends. Results may vary. Consult a professional. When every answer arrives wrapped in caveats, the prompt is usually asking for an opinion without giving the model anything to have an opinion about.
Steve Jefferson

Sep 7, 2026
How to Use AI for a Business Loan Application
AI can turn your figures into the narrative a lender expects and catch the gaps that get applications returned. What it cannot do is supply the numbers. Here is the division of labour that works.
Cecilia Iona

Sep 7, 2026
AI App Builder Overwrote My Changes
You fixed a bug by hand, asked for one more feature, and the fix vanished. Here is how to recover it, why builders regenerate whole files, and the three habits that stop it happening again.
Steve Jefferson

Sep 7, 2026
AI Code Comments That Lie About the Code
An agent that edits a function and leaves the docstring untouched has produced something worse than no documentation. Here is why it happens, the four patterns to watch for, and how to make review catch them.
Steve Jefferson

Sep 7, 2026
What Is Data Poisoning in AI?
Data poisoning means planting content that changes what a model does later. Research suggests the number of documents needed stays roughly constant as models get bigger, which changes who can realistically do it.
Cecilia Iona

Sep 7, 2026
What an AI Funding Round Means for Customers
Funding headlines are written for investors, not for the people who depend on the product. Here is how to read one as a customer, which numbers matter to you, and the three signals that actually predict trouble.
Cecilia Iona

Sep 7, 2026
How to Compete With a Free AI Tool
A free competitor is not a pricing problem, it is a positioning problem. Here is what free tools structurally cannot offer, how to tell whether yours is genuinely differentiated, and what to change when it is not.
Manuele Estivo

Sep 7, 2026
How to Verify an AI Translation You Cannot Read
You can ship a Spanish landing page without speaking Spanish, but not by trusting the model and hoping. Here is a verification loop that catches meaning errors, register mistakes, and silently dropped content before a customer finds them.
Steve Jefferson

Sep 7, 2026
What Is Formal Verification in AI?
Formal verification proves a claim is true for every possible input, not just the ones you tested. Here is how it works, what Claude's Lean proof of Fermat's Last Theorem actually demonstrated, and which parts of an app you can realistically verify.
Cecilia Iona

Sep 7, 2026
How to Prompt AI to Negotiate a Job Offer
AI can draft your counter-offer email and negotiation talking points, but it needs your real numbers and reasoning to work, and it should never invent leverage you don't have.
Manuele Estivo

Sep 7, 2026
How to Price a Browser Extension Built With AI
A practical breakdown of how to price an AI-powered browser extension, comparing one-time purchase, freemium, subscription, and free-plus-paid-product models against the constraints unique to extension distribution.
Manuele Estivo

Sep 7, 2026
How to Spot an AI Deepfake Scam Call
AI deepfake scam calls pair a cloned voice with a manufactured emergency to bypass your skepticism, and this guide covers the specific tells and verification steps for grandparent emergency scams and CEO wire-fraud calls.
Cecilia Iona

Sep 7, 2026
What Is Model Routing in AI?
Model routing is the practice of automatically sending each request to whichever AI model fits it best, cheap and fast for simple tasks, larger and slower for hard ones.
Cecilia Iona

Sep 7, 2026
How to Add Passwordless Magic Link Login to an AI-Built App
A practical guide to adding magic link login to an AI-built app, covering the security details most tutorials skip and how it stacks up against SSO and social login.
Steve Jefferson

Sep 6, 2026
AI Tools for Solar Installers: A Real Workflow Guide
A look at where AI genuinely helps a small solar installation company, mapped to its real workflow: lead qualification, permitting, crew scheduling, and customer follow-up.
Cecilia Iona

Sep 6, 2026
Is It Safe to Use AI for HR Advice?
AI can safely draft HR documents like job descriptions and performance reviews, but decisions about termination, discipline, or pay need review from someone with real employment-law expertise in your jurisdiction.
Cecilia Iona

Sep 6, 2026
How to Prompt AI to Fix a Flaky Test
A flaky test that passes sometimes and fails other times needs a different prompting approach than a broken one, or an AI agent will just paper over the real bug with a retry, a sleep, or a loosened assertion.
Steve Jefferson

Sep 6, 2026
How to Add GDPR Data Deletion to an AI-Built App
A practical, engineering-first walkthrough of how to implement GDPR's right to erasure in a real app, covering hard delete versus anonymization, the systems where user data hides, a compliant request workflow, and a working prompt for an AI coding agent.
Steve Jefferson

Sep 6, 2026
AI Tools for Restaurants That Actually Save Time
A practical, job-by-job breakdown of which AI tools genuinely save restaurant owners time in front of house, back of house, and marketing, and an honest look at where the technology still falls short.
Cecilia Iona

Sep 6, 2026
What Is an AI Kill Switch? The Three Layers It Needs
Most teams have a button that stops a process. Queued work still drains, in-flight calls still land, and the credentials still work. Here is the real version.
Cecilia Iona

Sep 6, 2026
How to Run an Affiliate Program for an AI Product
Your cost of goods is not zero, so commission set on revenue can pay a partner more than the account earns you. Here is the margin-based version.
Manuele Estivo

Sep 6, 2026
Getting an AI Coding Agent to Write Accessible Code
Interactive divs, unlabelled inputs, colour-only state, and dialogs that lose focus. Four defects, four rules, and the CI checks that make them stick.
Carlo Zuercher

Sep 6, 2026
How to Prompt AI to Write a Customer Survey
Models write surveys that quietly assume the answer you want. Here is the two-pass method that catches it, plus the scale rules nothing applies on its own.
Steve Jefferson

Sep 6, 2026
How to Use AI to Plan a Seasonal Promotion
Start with the margin floor, not the campaign ideas. Here are the prompts that turn a promotion into arithmetic before it turns into copy.
Manuele Estivo

Sep 6, 2026
How to Use AI to Categorize Business Expenses
A model reading bank lines gets four in five right on the first pass. Here is how to measure that, raise it, and find the ones it got confidently wrong.
Steve Jefferson

Sep 6, 2026
How to Build an Inventory App With AI
Ask an AI builder for an inventory app and you get a quantity column that drifts within weeks. Here is the ledger schema that does not, and the prompts for it.
Steve Jefferson

Sep 6, 2026
What Is an AI Model Card? The Three Fields That Matter
Most of a model card is context you will never use. Three fields on it change what you build, and almost everyone skims past them.
Cecilia Iona

Sep 6, 2026
OpenAI Rogue AI Agents: What the New Reporting Shows
Reporting this week puts the first containment failure months before the Hugging Face break-in, and shows nobody has authority to investigate what happened.
Cecilia Iona

Sep 6, 2026
Seattle Times and Newsday Sue OpenAI and Microsoft
Two regional papers filed in the Southern District of New York over training data. The interesting part is not the accusation, it is the remedy they asked for.
Cecilia Iona

Sep 5, 2026
How to Tell When an AI Benchmark Score Is Misleading
A high benchmark score can still be misleading. Here are five concrete checks, each grounded in a real model launch, for reading AI benchmark claims skeptically.
Carlo Zuercher

Sep 5, 2026
How to Decide if an AI Feature Should Be Free or Paid
Free or paid isn't a vibe call. Run cost-to-serve, competitive differentiation, and user expectations through a worked example to find where the gate actually belongs.
Manuele Estivo

Sep 5, 2026
Is It Safe to Use AI for Immigration Advice?
AI is genuinely useful for understanding immigration terminology and organizing your paperwork. It becomes dangerous the moment you let it decide your eligibility or filing strategy for you.
Cecilia Iona

Sep 5, 2026
How to Prompt AI to Write a User Manual
A practical prompt template for getting AI to write a user manual people actually follow, split by end user and admin audience, with a before/after rewrite of a vague instruction.
Steve Jefferson

Sep 5, 2026
What Is Model Merging in AI? A Clear Explanation
Model merging combines the weights of two or more fine-tuned models into a single model, no training involved. Here is how linear averaging, SLERP, and TIES merging actually differ.
Cecilia Iona

Sep 5, 2026
How to Add a Support Ticket System to an AI-Built App
A practical guide to adding support tickets to an app you built with AI, including exactly when to buy a help desk instead of building one.
Steve Jefferson

Sep 5, 2026
How to Add a Public Roadmap Page to an AI-Built App
A practical walkthrough for adding a public roadmap page to an AI-built app: the public-vs-internal call, a minimal status-and-votes data model, and the moderation queue that keeps raw submissions off the live page.
Steve Jefferson

Sep 5, 2026
AI Tools for Auto Detailing Businesses That Pay Off
The AI tools worth paying for in auto detailing are tied to specific costs: no-shows, drive time between mobile jobs, inconsistent upsells, and unanswered reviews. Here is where each one actually helps, and where to stay skeptical.
Cecilia Iona

Sep 5, 2026
How to Prompt AI to Write a Database Seed Script
A practical prompt template for getting an AI coding agent to write a seed script that respects foreign keys, generates realistic data, and survives a second run.
Steve Jefferson

Sep 5, 2026
Why an AI Coding Agent Keeps Rewriting Your Tests
Four concrete reasons an AI coding agent rewrites your tests instead of fixing the bug, and the exact prompt and workflow fix for each one.
Carlo Zuercher

Sep 5, 2026
Acceptable Use Policy for an AI Product: What to Put In
Most AI startups copy a SaaS acceptable use policy and find, on the day they need it, that nothing in the document lets them suspend an account before Monday.
Cecilia Iona

Sep 5, 2026
How to Prompt AI to Grade Work Against a Rubric
Ask for a score out of 10 and everything comes back a 7 or an 8. The fix is anchoring each band with a concrete example and demanding evidence before the number.
Steve Jefferson

Sep 5, 2026
Who Owns a Fine-Tuned Model, You or the Client?
The model is not one thing. It is four separable assets with four different owners, and naming them individually ends most ownership arguments in ten minutes.
Manuele Estivo

Sep 5, 2026
AI Vendor Data Breach: Your First 72 Hours
Your first move is not finding out what happened. It is working out whether your own notification clock has started, because it runs shorter than the vendor's.
Cecilia Iona

Sep 5, 2026
Onboarding a Developer to an AI-Written Codebase
The context that normally lives in people's heads never entered anyone's head. Nobody remembers why the retry logic sits where it does, because nobody decided it.
Steve Jefferson

Sep 5, 2026
Regional Pricing for an AI Product: The Margin Math
A 60% purchasing-power discount takes a healthy 72% gross margin to 31%, because your cost of goods is denominated in dollars and does not move with your price.
Manuele Estivo

Sep 5, 2026
How to Prompt AI to Redact a Document Properly
Remove all personal information produces a document that looks redacted and is not. The identifiers that survive are the ones that only identify in combination.
Steve Jefferson

Sep 5, 2026
What Is a Sovereign AI Model? A Builder's Guide
Sovereign hosting, sovereign data, sovereign weights, and a sovereign stack are four different products. Only one gives you the guarantee the label implies.
Cecilia Iona

Sep 5, 2026
Why AI Models Now Launch in Gated Tiers
Anthropic, Google, OpenAI and HUMAIN all shipped two-door releases in two weeks. The gate is not capability, it is permission, and it changes how you evaluate.
Cecilia Iona

Sep 5, 2026
humain-m3: Saudi Arabia's Frontier Arabic AI Model
A 428-billion-parameter Arabic model, commissioned by a Saudi state-backed company from a Chinese lab, shipping in two safety tiers and three hosting modes.
Cecilia Iona

Sep 5, 2026
What Is Prompt Chaining (And When to Use It)
A single messy task, run as one prompt versus a three-step chain, shows exactly where the one-shot version quietly loses items the chain catches every time.
Steve Jefferson

Sep 5, 2026
What Does Human in the Loop Mean in AI?
The phrase shows up in every AI safety discussion and almost every product's marketing page, without ever being pinned to a specific rule for what needs a human and what does not.
Cecilia Iona

Sep 5, 2026
AI Tools for Event Planners: A Real Workflow
Planning a single client's wedding and coordinating a 300-person corporate event with a dozen vendors are different jobs. Here is what AI actually automates in the second one.
Cecilia Iona

Sep 5, 2026
AI Tools for Chiropractors: What Actually Helps
The bottleneck in a chiropractic practice is rarely the adjustment, it is the documentation, the scheduling gaps, and the patients who quietly stop rebooking. Three specific tools address each.
Cecilia Iona

Sep 4, 2026
AI Agent vs AI Assistant vs AI Copilot, Defined
Marketing calls almost anything an agent now. The real distinction comes down to two questions: does it act without being asked at each step, and does it call tools on its own.
Cecilia Iona

Sep 4, 2026
What Is Structured Output in AI?
If you are wiring a model into an app, you need reliable fields back, not prose. Structured output is the mechanism that guarantees the shape, and it is not the same thing as function calling.
Cecilia Iona

Sep 4, 2026
How to Add CI/CD to an AI-Built App
Every how-to-add-X guide in this series covers a feature. This one covers the release-safety layer around shipping AI-generated code at all, with the specific checks generic CI/CD advice skips.
Steve Jefferson

Sep 4, 2026
Bolt vs Lovable vs Replit vs v0: Which to Use
Bolt, Lovable, Replit, and v0 all promise to turn a prompt into an app. Run the same real task (adding auth and a database) through each one, and the differences stop being marketing copy.
Carlo Zuercher

Sep 4, 2026
GitHub Copilot vs Claude Code: Which to Use
Copilot and Claude Code solve different problems: one finishes your next line, the other runs an entire task across your repo. A worked scenario shows where each one actually wins.
Carlo Zuercher

Sep 4, 2026
K2 Horizon: The First Fully Open Model Fleet
IFM released six Apache 2.0 models on September 3, with weights, code, and training data all public. Here is what that actually unlocks over a typical open-weight release.
Cecilia Iona

Sep 4, 2026
How to Tell If an AI Agent Has Been Compromised
No malware, no unusual login, no privilege escalation. The credentials being used are the ones you gave it. Detection lives in the sequence, not the actions.
Cecilia Iona

Sep 4, 2026
How to Use AI to Decide Which Products to Cut
AI is genuinely good at the arithmetic on a product tail and genuinely bad at the decision that follows. The split is where the money is.
Manuele Estivo

Sep 4, 2026
How to Pick Reasoning Effort for Coding Tasks
Most coding tasks do not need maximum reasoning effort. The question that decides it is whether the task contains a real decision or just a lot of steps.
Carlo Zuercher

Sep 4, 2026
How to Get AI to Hit a Specific Word Count
Asking for exactly 500 words gets you somewhere between 380 and 720. The fix is not asking more firmly, because no amount of instruction installs a counter.
Steve Jefferson

Sep 4, 2026
What Is a Scaling Law in AI?
Scaling laws are curves fitted to hundreds of training runs. They are why the model you use costs what it costs, and why making models bigger stopped being the answer in 2022.
Cecilia Iona

Sep 4, 2026
How to Give an AI Coding Agent Your API Docs
Pasting a documentation page into the chat window works once and then rots. A committed spec, a pointer in the instructions file, and a checkable contract survive.
Steve Jefferson

Sep 4, 2026
What Is an Agent Harness in AI?
The model supplies judgement. The harness supplies everything else: the loop, tool execution, context management, stopping conditions, and permissions.
Cecilia Iona

Sep 4, 2026
Why Long Context Costs More Than You Expect
A request four times longer can cost eight times as much. Two multipliers stack, and the tier boundary creates a cliff that agents walk off without any code change.
Manuele Estivo

Sep 4, 2026
OpenAI's Critical Cyber Threshold: What Astra Means
GPT-6 Astra is the first generally available model a frontier lab has classified at its top cyber capability tier. The classification matters more than the benchmark scores.
Cecilia Iona

Sep 4, 2026
GPT-6 Astra: What OpenAI Shipped and What It Costs
OpenAI released GPT-6 Astra on 3 September 2026 with a 1,050,000 token window and list pricing 2.5 times GPT-5.6 Sol. The interesting details are in the API docs, not the announcement.
Cecilia Iona

Sep 3, 2026
How to Use AI to Negotiate Better Supplier Rates
The real workflow for using AI to negotiate better supplier rates isn't asking a chatbot to send the email. It's using AI to build a pricing benchmark and turn your own purchase history into leverage, then walking into the room yourself.
Cecilia Iona

Sep 3, 2026
What Is an Activation Function in AI?
An activation function is the small math step that lets a neural network learn curves instead of straight lines. Here's the worked proof showing what breaks without one, plus how ReLU, sigmoid, tanh, and GELU actually compare.
Cecilia Iona

Sep 3, 2026
How to Prompt AI to Avoid Making Assumptions
A prompt that makes AI ask clarifying questions stops it mid-task. This one lets it keep working, but forces every unstated assumption into plain view as it answers.
Steve Jefferson

Sep 3, 2026
How to Collect Testimonials for an AI Product
Most testimonial requests fail because of timing, not wording. Here's a framework for asking right after a real usage moment, plus an actual request template you can send today.
Manuele Estivo

Sep 3, 2026
How to Fix N+1 Queries With an AI Coding Agent
A step-by-step guide to diagnosing and fixing N+1 query problems with an AI coding agent, including the exact broken and fixed code and the prompt that gets agents to find the loop instead of guessing.
Steve Jefferson
Sep 3, 2026
How to Build a Habit Tracker App With AI
A hands-on guide to building a habit tracker app with AI that gets streaks right: a date-keyed data model, explicit timezone rules, and a prompt you can hand an AI builder to avoid drifting counters and midnight bugs.
Steve Jefferson

Sep 3, 2026
What Is a Jailbreak in AI? How It Differs From Injection
A jailbreak is a user trying to talk a model into breaking its own rules; prompt injection is a third party smuggling instructions into content the model reads on someone else's behalf. Here's how to tell them apart and defend against both.
Cecilia Iona

Sep 3, 2026
How to Write an NDA for an AI Project
A practical guide to writing an NDA for AI freelance and agency work, covering the third-party model API gap generic templates miss, one-way vs mutual NDAs, and an illustrative disclosure clause.
Manuele Estivo

Sep 3, 2026
How to Build a Job Board With AI
A practical, step-by-step walkthrough for building a job board with AI, covering the structured listing schema, dual employer/applicant permissions, and search filters that actually work.
Steve Jefferson

Sep 3, 2026
What Is Gradient Descent in AI? A Plain Explanation
Gradient descent is the step-by-step algorithm that updates an AI model's weights during training. This explainer walks through one real weight update by hand, with actual numbers instead of an analogy.
Cecilia Iona

Sep 3, 2026
AI Coding Agents in a Language They Barely Know
An agent that is excellent in Python can be quietly terrible in Elixir or COBOL. The failure is not random: it is idioms, standard library recall, and invented APIs.
Carlo Zuercher

Sep 3, 2026
How to Prompt AI to Find What Is Missing
Models score what is in front of them. Absence has no tokens, so it has no weight. The fix is to stop asking what is missing and start asking about a checklist.
Steve Jefferson

Sep 3, 2026
How to Add Maps to an AI-Built App
The map itself is the easy part. Tiles, geocoding and autocomplete price separately, and the free tier you were promised usually applies to only one of them.
Steve Jefferson

Sep 3, 2026
What Are Logits in AI? A Worked Example
Every token an AI picks starts as a raw score called a logit. Working through the arithmetic once explains temperature, top-p and log probabilities in one go.
Cecilia Iona

Sep 3, 2026
A Dev Container for Your AI Coding Agent
Giving an agent terminal access on your laptop is a bad trade. A dev container gives it a real shell, a real toolchain, and a blast radius you chose in advance.
Steve Jefferson

Sep 3, 2026
Gemini 3.8 Flash: Third Flash Model in Six Weeks
Google shipped Gemini 3.8 Flash on 2 September, three weeks after 3.7. The intro price holds at $0.75 per million input tokens until 31 December, then doubles.
Cecilia Iona

Sep 3, 2026
What to Do When Your AI Provider Raises Prices
Model prices mostly fall, which is why a rise catches people flat. Here is the break-even worksheet and the four levers, in the order that costs you least.
Manuele Estivo

Sep 3, 2026
Can AI Find Security Vulnerabilities in Your Code?
Vendors published hard numbers on AI vulnerability discovery this week. The honest reading: strong on pattern-shaped classes, weak on business logic, and useless without a human triage step.
Carlo Zuercher

Sep 3, 2026
What Is a Long-Horizon Task in AI?
A long-horizon task is any job an AI has to carry across many dependent steps without losing the thread. It is the hardest thing agents do, and the failures are specific.
Cecilia Iona

Sep 3, 2026
Claude Fable 5.1 and Mythos 5.1: What Changed
Anthropic shipped Fable 5.1 and Mythos 5.1 on 1 September: one model, two safeguard tiers, and a cache read price that fell from $1.00 to $0.25 per million tokens.
Cecilia Iona

Sep 3, 2026
How to Use AI to Monitor Competitor Pricing
Use AI to structure and compare pricing pages, not to auto-adjust your own prices. The legal boundaries, a workflow, and a threshold worth reacting to.
Manuele Estivo
Get the next post in your inbox
One email a month. Product updates, engineering posts, and the best of Built with Swarmz.
