Updates to the Evaluator Playground — Evaluate LLM-as-a-Judge and Other Evaluators

[mmabrouk]

Overview

We are significantly improving the evaluator playground to make it easier to build, test, and validate evaluators — including LLM-as-a-Judge.

This includes a richer editing experience, inline test runs, and the ability to evaluate evaluators against a labeled test set to measure how well they agree with ground truth.

👇 Share your use cases below or upvote if this is useful to you.