A self-contained Streamlit prototype of a production agent evaluation system: golden-set / LLM-judge / human-review scoring, SQLite storage, daily aggregation, and a fleet dashboard.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results