Blog

Writing on building with language models. Ideas from building LLM features for real users, not from demos.

LLMs are functions, not brains.
A case for treating LLMs as typed functions, not autonomous agents. With live demonstrations.
May 2026
Agentic-First: Designing Systems for AI From Day One
Most teams hit a ceiling with AI and assume the model is the problem. It isn't. It's the system around it.
April 2026