LLM-as-Judge Eval Design for Agentic Task Outcomes
Fixing how LLM judges evaluate multi-step agent trajectories instead of single outputs.
Camille Marchetti
Contributing Editor
Camille Marchetti
Camille Marchetti is a contributing editor at The Agentic Harness covering eval & validation. Based in Mexico City, Camille has written for The Agentic Harness since 2022.
1 story · Mexico City