Mathematician for AI Model Evaluation

SaidGig; New York (NY); Remote; USD 55,000 - 110,000; Part time

Job summary

SaidGig is seeking a remote contract mathematician to design and evaluate advanced AI-driven mathematics tasks across undergraduate to Ph.D.-level topics. You will craft challenging problems, solve them with rigorous justification, and provide precise feedback on model solutions.

The role emphasizes independent work, strong mathematical training, and clear written explanations, with 20–40 hours per week and overlap with Pacific Time. Remote contract, no medical coverage provided.

Qualifications

• Master’s/Ph.D./postdoc in Mathematics or related field required or strongly encouraged.
• Strong mathematical foundation and research, analytical and creative thinking.
• Excellent structured communication and collaboration for remote work.
• Self-motivated, able to work independently and remotely.
• Proficient in Python and capable of formal reasoning tools such as Lean.

Responsibilities

• Design original mathematics problems that test large language model reasoning in multi-step, abstract, and proof-based settings.
• Solve problems independently and produce detailed, logically structured solutions with clear justifications.
• Review model-generated solutions, identify mathematical errors and provide precise feedback, annotations and corrections.
• Help define mathematics evaluation benchmarks spanning undergraduate through Ph.D.-level curricula.
• Design precise computational prompts, write reliable Python solutions, validate numerical answers with explanations.
• Complete theorem-prover work in Lean, translate problems and proofs into formal language and verify proofs compile.
• Break down complex mathematical concepts into clear explanations using simple language, visuals and examples.

Education

Master’s/Ph.D./postdoc in Mathematics or related field

Tools

Lean

Python

Job description

Shape and evaluate advanced AI systems through rigorous mathematical reasoning, problem-solving, computational work, and clear written explanations. This remote contract role focuses on creating and assessing challenging mathematics tasks across undergraduate through Ph.D.-level subject areas.

Key Responsibilities

• Design original, challenging mathematics problems that test large language model reasoning in multi-step, abstract, and proof-based settings.
• Solve problems independently and produce detailed, logically structured solutions with clear justifications.
• Review model-generated solutions, identify mathematical errors and missing arguments, and provide precise feedback, annotations, and corrections.
• Help define mathematics evaluation benchmarks spanning early undergraduate through Ph.D.-level curricula.
• Design precise, closed-ended computational prompts, write reliable Python solutions, validate numerical answers, and provide clear rationales using approved scientific libraries.
• Complete theorem-prover work in Lean, including translating mathematical problems and proofs into formal language and verifying that formal proofs compile correctly.
• Break down complex mathematical concepts into clear explanations using simple language, visuals, and examples.

Qualifications

• Strong mathematical foundation at engineering entrance-exam and graduate-program levels.
• Research, analytical, creative, and lateral-thinking skills.
• Ability to solve complex mathematics problems with a structured, logical approach.
• Ability to provide constructive feedback and detailed annotations.
• Excellent structured communication and collaboration skills for a remote environment.
• Self-motivated, able to work independently, and able to work efficiently.
• Desktop or laptop with a reliable internet connection.

Eligibility

Candidates pursuing a Master’s, Ph.D., or postdoctoral degree in Mathematics, Applied Mathematics, Statistics, or a related field are eligible and encouraged to apply.

Work Terms

• Fully remote contract, independent-contractor assignment.
• Choose a commitment of 20, 30, or 40 hours per week.
• Commit at least 4 hours per day, with a minimum of 20 hours per week.
• Maintain 4 hours of overlap with Pacific Time.
• This contractor engagement does not include medical coverage or paid leave.