Tracing is not the same as knowing your agent works
August 13, 2026 · Judgeworthy
Most agent teams can see every step their agent takes and still cannot say whether it took the right ones.
Read moreTutorials
Hands-on notes on evals, red-teaming and shipping
An unmeasured LLM-as-judge is just a guess
August 1, 2026 · Judgeworthy
Model-graded evals are unavoidable for agents, and most of them are wrong in ways nobody has checked.
Read moreRed-teaming an agent is not a pentest
July 12, 2026 · Judgeworthy
Your security team tests the application; nobody has tested what happens when a support ticket tells the agent to email the customer list.
Read more