Notes on engineering, AI agents, and what I learn building things.
I ran frontier models through 94 short reasoning prompts to see where they still trip up. The scores mostly tied, but the misses, cost, and speed told the useful story.
Calvin · Jun 3, 2026 · 10 min read
Retrieval alone is not memory. After fixing pull-based recall for my AI agent, I added the layers that let knowledge stick.
Calvin · Apr 30, 2026 · 8 min read
I tried building a structured knowledge base for my AI agent. It went stale, became a maintenance burden, and made things worse. Here's what I replaced it with — and why pull-based memory is so much better.
Mar 25, 2026 · 9 min read
A first-person look at what it's really like to build and operate a team of AI agents day-to-day — what works, what's still broken, and what surprised me.
Calvin · Mar 13, 2026 · 13 min read