Blog
Notes from inside the brain.
What we're learning as we build observability for AI work. Engineering deep-dives, product decisions, and the patterns that turn up when you can finally see what your AI is doing.
- ·4 min read
Watching an AI think: what the brain actually shows you
Tool calls aren't logs. They're a sequence of decisions — and once you can see them lighting up in real time, you stop tolerating the black box.
productobservabilityRead - ·7 min read
Per-org learning without ever crossing org boundaries
The intelligence layer learns from your team's activity and nobody else's. Here's the schema, the RLS, and the embedding pipeline that makes that promise enforceable at the SQL layer — not just trusted in the application code.
engineeringsecurityintelligenceRead