Mathematics Assessment Specialist (AI Evaluation)
About the Role
A high-impact AI research initiative is building a structured pipeline of rigorous mathematical assessment content to enhance model reasoning and evaluation capabilities. The project focuses on developing benchmark-quality questions across advanced mathematical domains to support next-generation AI systems.
This opportunity is ideal for individuals with deep academic training in mathematics who can design and evaluate complex problems with precision. It suits candidates comfortable working independently, applying formal reasoning, and maintaining high standards of academic rigor.
The work involves authoring and validating challenging multiple-choice questions, constructing detailed solutions, and ensuring conceptual depth and clarity. Success in this role depends on accuracy, structured thinking, and the ability to translate advanced mathematical ideas into well-defined assessment formats.
What You'll Do
- Author original multiple-choice questions across advanced mathematics domains
- Design problems that assess conceptual understanding rather than recall
- Ensure all questions are precise, self-contained, and unambiguous
- Develop one correct answer and multiple plausible distractors
- Write clear, step-by-step solutions using structured mathematical reasoning
- Assign and justify difficulty ratings (undergraduate to postgraduate levels)
- Review and validate existing questions for correctness and rigor
- Identify and correct issues related to clarity, completeness, or solvability
- Document edits and provide justification for revisions
- Reference credible academic sources to support problem design
Requirements
- PhD or doctoral candidacy in Mathematics, Applied Mathematics, Statistics, or a related field
- Strong command of graduate-level mathematical concepts and formal proofs
- Expertise in at least one domain such as algebra, analysis, probability, discrete mathematics, or optimization
- Ability to construct and evaluate rigorous academic assessment content
- Proficiency in writing clear, structured mathematical explanations in English
- Experience with mathematical problem design or competition-level questions preferred
- Ability to work independently in a remote, asynchronous environment
- Availability for consistent weekly contribution (10+ hours preferred)
- Familiarity with markdown formatting for technical content
- Preferred: experience contributing to academic publications or assessment systems