Blog
Long-form posts. Free to read.
How Much Access Should Your Agent Have?New
Coding agents run with all of your access. Here are four ways to limit what they can reach, from approval prompts to a separate machine.
When Should You Use a Cheaper AI Model?New
Which tasks need the smartest model and which can go to something cheaper like Haiku 5.5, comparing cost, speed and capability.
You Don't Need an LLM for EverythingNew
These days, it's tempting to reach for an LLM whenever your code needs to make a decision about some text, like which team should handle a support ticket. Jev, from TypeSafe AI, is a new kind of model built just for these decisions, and it works very differently to an LLM.
You're Not Being Replaced. You're Being Promoted to Editor
There's a lot of worry at the moment about whether AI agents are coming for developer jobs. I get why, and I'm not going to pretend nothing is changing. But from how I now work every day, the job isn't disappearing, it's changing shape.
Build an MCP Server for Your Team (Not for the World)
There's an MCP server for everything now. GitHub, Stripe, Postgres, Figma, you name it. But almost everything written about MCP is about consuming these big public servers, and I think the real win is much closer to home: a small private server that wraps your own team's…
Stacked PRs: Everything You Need to Know
Stacked PRs are a way of breaking one big change into a chain of small pull requests that each get reviewed on their own. They've been standard at bigger companies for years, and GitHub has just opened its native version up to everyone.
The 8 Most Common Mistakes AI Agents Make When Writing Code
Discover the eight most common mistakes AI coding agents make when writing code. Learn how to spot recurring errors, review AI-generated code effectively, and prevent costly issues before they reach production.
Context Engineering: Everything You Need to Know
If you've spent any time building with LLMs (or just following along), you'll have come across prompt engineering. Context engineering is the term that's been replacing it, and once you understand what it covers, you'll see why. Let's dive in.
The Most Powerful Model Is the Wrong Default
Choosing the most powerful AI model isn't always the best strategy. Learn how to match coding tasks to the right model, reduce costs, improve efficiency, and know when a frontier model is truly worth using.
How to Build a Skill With Your AI Agent
Learn how to create reusable AI coding agent skills with SKILL.md files. Discover how skills help your agent apply the right instructions automatically, reduce repetitive prompts, and improve consistency across development tasks.
From Prompts to Specs with OpenSpec
OpenSpec helps you and your AI coding assistant stay aligned by defining requirements before development begins. Learn how a lightweight specification layer reduces misunderstandings, improves code quality, and keeps complex features on track.
The AI Code Review Problem Nobody Talks About
AI-generated code can look clean, structured, and correct—but that’s exactly what makes it risky. This article explores the hidden dangers of trusting AI-written code and why developers need to rethink how they review it before it reaches production.
Top Claude Code Tips for Productivity
Unlock the full potential of Claude Code by learning how top engineers use it as a development environment, with proven practices from Anthropic teams and power users.
5 Misconceptions to being an AI Driven Dev
The rise of AIpowered development tools has transformed how we write, debug, and deploy code. From GitHub Copilot, Cursor, Claude Code to ChatGPT, these tools have become integral to many…
The AI Revolution in Development Is Here: Introducing the AI-Driven Development (AIDD) Course
While you've been using AI casually—maybe for quick autocomplete in Cursor or the occasional ChatGPT query— 30% of GitHub's and Microsoft's code is now written by…
Course 1: The Future Proof Dev: An Intro to AI for Web Developers
You feel it. Change is in there air and it’s been hanging around for a while now but you can’t quite find your place in it. Yes, I’m talking about the impact of AI on the web development world.
Throw Your IDE Away? Why CLI Coding Agents Are Changing Everything
The title sounds dramatic, I know. Throw your IDE away? Really? To be clear: I’m not literally telling you to uninstall VS Code and go live in a cave with…
The Case for Strategic Agents: Moving Beyond AI Autocomplete
Insights from my recent talk at AIDD Development Day As developers, we've grown accustomed to AI tools that complete our code and suggest quick fixes. But what if we're thinking too small?
Top AI tools to look out for in 2026
AI tools for developers hit a turning point somewhere in 2025. What started as autocomplete on steroids evolved into systems that could read entire codebases, write tests, review pull requests, and…
Barebones AI Guide: Your First Steps for Faster Development
AI is moving at such a rapid pace that if you’ve not yet started to integrate it into your daily workflow, it’s difficult to know where to start.
From Google to AI: Debugging Errors with Context
A clear shift has happened from searching Google (which, let’s be honest, almost always led us to Stack Overflow) to debugging with AI.
Improving Your AI Agent, One Rule at a Time
tl;dr. Use an agent markdown file, review it regularly, don’t let it get out of control, and share it with your team. AI agents thrive on context.
From Skepticism to Buy-In: Helping Developers Adopt AI
There’s no question about the impact that AI is having on the software development industry. We’re way past the hype phase, and it’s clear that AI can make developers and teams more productive with…
AI Prompts for Every Development Workflow Step (With Copy/Paste Examples)
Effective prompting is harder than it seems. You’re no longer expressing logical steps directly in code.