occupation
Software Quality Assurance Analysts & Testers
description
Run Jev-native judgments on LLM assistant replies and agent trajectories using shared rubrics. Use when evaluating model output quality, safety, or tool-use trajectories before shipping or merging.
updated