Method

Agent evaluation

Evaluating agents on traces, multi-turn scenarios, stubbed tools and behavioral evals.

13 cases from 10 companies apply Agent evaluation. 4 report results in production.

Typical tasks: Evaluate models, Answer questions over data, Build a data platform, Optimize decisions. Companies: DoorDash, Ramp, Pinterest, Airbnb, Grab, Avito and 4 more. Often combined with: LLM-as-a-judge, Golden set / benchmark, Human in the loop, Evaluation regression gate, LLM observability and 19 more.