Remote AI Math Referee & Problem Designer

DataAnnotation; Delaware; Remote; USD 138,000 - 172,000; Full time

Benefits offered by this job

Remote work

Job summary

DataAnnotation is seeking a Mathematician to evaluate and improve mathematical reasoning across contest problems to graduate-level proofs. You’ll judge validity, identify incorrect steps, and set training standards used to teach AI models.

You’ll craft problems, review solutions in detail, and author gold-standard explanations when models fall short, emphasizing clear English and rigorous proofs with a remote-friendly setup.

Qualifications

• Strong background in pure or applied mathematics through graduate study, competition math, or equivalent experience.
• Fluency writing precise, well-structured proofs (LaTeX comfort helps).
• Clear written English: your explanations are the training signal.
• No degree required. We care about what you can do, not where you learned it.

Responsibilities

• Craft problems that expose weak mathematical reasoning, from olympiad-style puzzles to graduate-level proofs.
• Review AI solutions step by step: catch invalid derivations, flag unjustified leaps, and rate the quality of the reasoning.
• Write the gold-standard solution when the model falls short, with the rigor of a referee report.

Skills

Proof-writing

LaTeX

Mathematics

English communication

Education

Graduate study in mathematics

Competition math

Tools

LaTeX typesetting

Job description

DataAnnotation is seeking a Mathematician to evaluate and improve mathematical reasoning across contest problems to graduate-level proofs. You’ll judge validity, identify incorrect steps, and set training standards used to teach AI models.

You’ll craft problems, review solutions in detail, and author gold-standard explanations when models fall short, emphasizing clear English and rigorous proofs with a remote-friendly setup.