Researcher, Benchmark Reviews

Responsibilities

• Assess a new benchmark at least every two weeks
• Evaluate benchmark methodologies and implications for AI capabilities
• Write and maintain public-facing research reports
• Update reports as new model versions and benchmarks are released
• Examine individual tasks within benchmarks using coding agents

Requirements

• Excellent written communication skills for publication-quality reports
• Strong critical thinking regarding research methodologies
• Familiarity with existing AI benchmarks and their methodologies
• Ability to work with and analyze dense datasets
• Professional level English proficiency

Preferred Qualifications

• Experience writing about AI/ML for public or research audiences
• Hands-on experience running or building AI evaluations

Benefits

• Annual salary between $100,000 and $200,000 USD
• Fully remote environment with flexible work hours
• Comprehensive health insurance and life insurance
• Generous paid time off (PTO) with 30 days protected
• 4 months paid parental leave for permanent staff
• Flexible expense policy for equipment and AI tools
• Paid work trips and staff retreats

About the Company

Epoch AI is a research institute that investigates trends in machine learning and the economic consequences of AI. Our mission is to develop a comprehensive, publicly accessible knowledge base on AI that informs policymakers, industry leaders, and society at large.