·1 min
What this blog is
A short note on why I'm posting again, what I'll cover, and how to follow along.
I've been building AI agents for a while now — coding agents, browser agents, QA agents — and I keep noticing the same thing: the interesting parts almost never show up in release notes. They live in the small decisions you make while wiring two systems together at 2am.
This blog is where I'll write those down. Mostly: notes on agent loops, what LLMs do when you push them past their comfort zone, local-first patterns, MCP, and small tools I keep reaching for. Short posts. No takes. Just things I'd want to read.
If that sounds useful, follow me on X or grab the RSS feed. New posts go up whenever I learn something worth keeping.