Mercor is seeking highly skilled Generalists (Bachelor’s degree minimum) to support a fast-moving project with one of the world’s leading AI labs. This initiative focuses on improving AI systems designed for therapeutic-style conversations by evaluating, comparing, and refining model-generated responses.
In this role, you will review pairs of AI-generated replies and assess them against structured criteria (e.g., empathy, emotional insight, communication quality, and appropriateness for supportive dialogue). When needed, you will also produce clear, improved rewrites that better meet the project’s standards. Your work will directly support research teams as they strengthen safe, supportive, and human-aligned AI behavior in mental-health-oriented conversation settings.
Important: This project has an accelerated timeline and requires contributors who can begin immediately following a mandatory onboarding session.
