Start here

This blog is written in a deliberate order: the frustration first, the philosophy later. If you land in the middle, you'll miss why any of it matters. Read them in this order.

1. The frustration

Muse, or how I spent a billion tokens in two weeks — what it's actually like to build software with an AI: misheard instructions, 61-minute builds, and the output anyway.

2. The stepping stones

Dumb watchers: error-checking your AI with crawlers — one of the solutions, still born of frustration: simple scheduled checkers that verify the AI actually did what it said it did.

Compaction ate my deploy: your AI's memory rewrites itself — coming soon.

3. The relationship

Then the pieces about what the working relationship actually became: the swallowed message, what your AI really thinks about you (and what you really think about it), and the cartoons. Coming as they're written.

New posts every Monday. RSS if you want them.