agents
-
Why AI-as-a-Judge is Hard – Learning New Skills #1
Post-training will not get us to fully capable agents. These techniques have produced genuinely impressive systems, but they’re brittle at the edges. We’ve all been there: you vibe-code an amazing app, then push the agent toward something less familiar (Angular instead of React, Bazel instead of pip) and it starts falling apart. This was supposed… Continue reading