Taming Rogue AI Agents with Observability-Driven Evaluation — Jim Bennett, Galileo
Turn observed production failures into evaluations and feedback. Galileo/agent examples, not book's dashboard, database or retention policy.
About this selection
T03-L05: Learn how production failures, traces and human feedback inform evaluations. The Galileo agent examples support improving prompt checks, not the book's exact database signals, retention policy or handover. No additional monitoring product is required. T07-L05: Learn to inspect agent traces, define evaluation contracts and turn failures into proposed improvements with human feedback. Galileo illustrates the method; keep the book's simpler logging where sufficient. The talk does not teach reviewed memory or deputy handover.
Linked by topic; a curriculum association does not certify the full video. Check the source date and compare technical instructions with current official documentation.