Agents
Agent evaluations are the new unit tests
In production — Episodes are being produced. Each page previews the question and what the brief will cover before it ships.
What this brief will cover
Choosing the simplest architecture that reliably completes the task, and building evals that catch regressions before your users do.
Related guide
Agent evaluations are the new unit tests
Build this yourself with a practical kit
Agent kit is not finished yet — we do not link a product that does not exist.