AI for Developers

Most Recent

AI for Developers

How to Move an AI App from a Frontier Model to Open Weights

The frontier model’s previous outputs were a useful starting point, but I treated them as suggestions rather than truth.

AI for Developers

I wanted to change Claude Code’s wall of text answers. I found output styles, markdown files that control how the agent talks to you.

AI for Developers

AI-generated code can compile, pass tests, and still solve the wrong problem. Context, client communication, and critical review matter.

AI for Developers

Atomic discusses MCP Servers: are they solving a real problem, or reinventing one? Four perspectives on whether they live up to the hype.

AI for Developers

New agent commands and skills are created every day. My emails are a deluge of new agent workflows promising to supercharge my productivity.

AI for Developers

Developers have been using AI as a tool for some time. Something that has slowly become my favorite use of AI: building developer tools.

AI for Developers

Here's a follow-up on our homegrown AI code review workflow, including making it block merges and getting consistent findings run to run.

AI for Developers

About three years into AI adoption, I've learned what burns me out, what speeds me up, and how I stay engaged in agentic development work.

AI for Developers

In learning how to work with AI agents, my team discovered how much of our team's experience had never been written down before.

AI for Developers

Let's look back on 2 months of agentic development on a HIPAA app: what we let an AI agent ship, and where humans stayed in the loop.

AI for Developers

Using spec-driven development, we cut up a large feature into vertical slices. It made our code delivery efficient & easy to test.

AI for Developers

While I haven't figured out how to replace my entire job with "loops" yet, I started with an MVP for my daily SDLC "loop."