Remote Applied Math Benchmark Architect

Synthires; United Kingdom; Remote; GBP 62,000 - 79,000; Full time

Job summary

Synthires seeks an Applied Mathematics Benchmark Specialist to author and verify rigorous mathematics content for AI benchmarking. This remote role focuses on deep concepts, formal reasoning, and precise problem formulation.

Ideal candidates hold a PhD and have strong proof-writing skills, with the ability to work asynchronously for 10+ hours per week. Compensation is hourly, with remote collaboration across time zones.

Qualifications

• PhD or doctoral-level study in Mathematics, Applied Mathematics, Statistics, or a closely related field.
• Strong command of graduate-level mathematical concepts and formal proof writing.
• Deep expertise in at least one of the covered mathematics domains.
• Strong ability to formulate, solve, and critically evaluate advanced mathematical problems.

Responsibilities

• Author original, challenging MC questions that assess deep conceptual understanding.
• Review and verify pre-written mathematics questions for accuracy and rigor.
• Ensure each question is unambiguous, self-contained, and precisely defined.
• Assign difficulty ratings: Medium, Hard, or Expert.
• Provide one correct answer and nine plausible distractors.
• Develop clear, step-by-step mathematical solutions in Markdown format.
• Provide 1–5 reputable academic references for each question.
• Identify issues in clarity, precision, completeness, or solvability and document edits.
• Apply rigorous mathematical standards to support AI benchmarking.

Skills

Mathematical proof writing

Graduate-level math concepts

Education

PhD in Mathematics/Applied Mathematics/Statistics

Job description

Synthires seeks an Applied Mathematics Benchmark Specialist to author and verify rigorous mathematics content for AI benchmarking. This remote role focuses on deep concepts, formal reasoning, and precise problem formulation.

Ideal candidates hold a PhD and have strong proof-writing skills, with the ability to work asynchronously for 10+ hours per week. Compensation is hourly, with remote collaboration across time zones.