Evals Engineer

Role overview

Build and maintain the evaluation process for an AI assistant. The role focuses on measuring how well it performs, detecting regressions early, and making evaluation results part of release decisions.

Responsibilities

- Own the evaluation suite used to measure assistant performance. - Create or maintain checks that reveal quality changes over time. - Identify regressions before they reach customers. - Make evaluation results a required gate in the release process.

Requirements

- The source emphasizes AI evaluation, performance measurement, regression detection, and release gating. It does not specify required experience, education, or particular technologies.