AI Evaluation Data Scientist

Mercor is seeking experienced data scientists to define what excellent work looks like and to score AI-generated and human work samples accordingly. You will design task-specific grading criteria, provide rigorous written justifications for every score, and ensure our scoring remains reproducible and defensible.

Ideal candidates bring 5+ years in industry data science, strong experiment design and A/B testing skills, and the ability to clearly communicate findings to executive stakeholders.