Competition Mathematics Evaluator at Mercor

Competition Mathematics Evaluator | $55–65/hr | Remote (US Only)

Mercor is partnering with a leading frontier AI research lab on an exciting mathematics and physics research initiative. We're seeking experts with demonstrated Olympiad-level problem-solving expertise to help train large language models to solve competition-level problems at world-class standards.

We invite competition problem writers, selection committee members, coaches, medalists, and participants to push the boundaries of AI reasoning in mathematics and physics.

Key Responsibilities:

• Write and evaluate Olympiad-level mathematics and physics problems with high standards of rigor and difficulty
• Assess AI-generated solutions for correctness, clarity, depth, and logical consistency
• Identify reasoning gaps, incorrect assumptions, incomplete proofs, or flawed derivations
• Provide clear, concise, and accurate written explanations and feedback
• Maintain exceptional standards of precision, structure, and correctness in all evaluations
• Work independently in a remote, asynchronous environment while following project guidelines

Required Qualifications:

Note: Applicants must be highly proficient in writing and formatting documents using LaTeX.

Experience: You are a strong fit if you have experience in Olympiad-level mathematics or physics, including roles such as:

• Competition problem writer
• Selection committee member
• Medalist (gold, silver, or bronze)
• Participant

Competition Background: Preference given to candidates with experience in IMO, IPhO, or RMM, though applications are welcome from a wide range of competitions including:

• Mathematics: IMO, IMO TST, RMM, USAMO, APMO, IMC, EGMO
• Physics: IPhO, IPhO TST, APHO, EuPhO, APhO, national-level Olympiads

Skills & Expertise:

• Deep expertise in Olympiad-level mathematics (algebra, combinatorics, number theory, geometry) and/or physics (mechanics, electromagnetism, thermodynamics, modern physics)
• Strong ability to evaluate multi-step reasoning and solution quality
• Exceptional attention to detail and commitment to correctness
• Clear written communication skills for technical explanations

Project Details:

• Expected commitment: 10 hours/week minimum; up to 20 hours/week available
• Approximate project length: ~1 week, with extension potential

About Mercor: Mercor is a San Francisco-based company specializing in connecting elite professionals with cutting-edge AI initiatives. Our investors include Benchmark, General Catalyst, Adam D'Angelo, Larry Summers, and Jack Dorsey.