Mercor Research Fellowship — APEX
Core
Design, implement, and release new AI benchmarks, evaluation techniques, or economic analyses of AI's impact on the labor market using proprietary enterprise data and expert networks.
Role type
Research Fellow (AI Evaluation & Economics)
Builds
New AI benchmarks, evaluation methodologies, and economic studies
Domain
Artificial Intelligence, AI Safety, Labor Economics, Professional Services
Deliverable
production ML models | research
Required skills
evaluation methodology design, empirical analysis, task specification design, rubric creation, statistical analysis, agentic task design
Preferred skills
agentic evaluation, RL environments, domain expertise in law/finance/medicine/science, startup environment adaptability
Technologies
frontier model APIs, GPU compute, enterprise data platforms
Responsibilities
Propose and scope new benchmark evaluation techniques or economic studies, build and validate benchmarks with expert networks, publish results as papers or datasets, calibrate scoring and analyze model failures
Seniority
Mid-level, hands-on IC