Remote Math AI Benchmark Designer (PhD)

turing; San Francisco (CA); Hybrid; USD 138,000 - 207,000; Full time

Benefits offered by this job

Fully remote

Flexible weekly hours

Job summary

Turing, based in San Francisco, California, seeks PhD candidates in Mathematics or Statistics for remote contract work to help fine-tune large language models. Earn $100+ per hour with fully remote, flexible weekly hours and no AI experience required.

You will design math problems, create benchmarks, and evaluate AI outputs, collaborating with researchers across advanced topics from undergraduate to PhD levels.

Qualifications

• PhD (pursuing or completed) in Mathematics, Applied Math, Statistics, or related field.
• Strong mathematical reasoning and problem-solving skills across advanced domains.
• Ability to communicate complex ideas clearly in writing and provide structured feedback.
• No AI experience required.

Responsibilities

• Design advanced math problems to test AI performance (e.g., multi-step reasoning, abstraction, symbolic manipulation).
• Develop clear, step-by-step solutions with rigorous logic.
• Evaluate AI outputs for accuracy and quality of reasoning.
• Collaborate with researchers to refine benchmarks across undergraduate to PhD-level math topics.

Skills

Mathematics

Analytical thinking

Academic writing

Benchmark design

Education

PhD in Mathematics

PhD in Statistics

Job description

Turing, based in San Francisco, California, seeks PhD candidates in Mathematics or Statistics for remote contract work to help fine-tune large language models. Earn $100+ per hour with fully remote, flexible weekly hours and no AI experience required.

You will design math problems, create benchmarks, and evaluate AI outputs, collaborating with researchers across advanced topics from undergraduate to PhD levels.