Role Overview: We are seeking an experienced Mathematics Expert to lead a team of 5–10 technical trainers in developing challenging mathematical and computational tasks for advanced AI training and evaluation. The role includes reviewing tasks for mathematical correctness, reproducibility, and evaluation quality, mentoring trainers, and ensuring high standards of technical rigor.
Responsibilities:
- Lead a pod of 5–10 trainers focused on mathematical and computational task development.
- Review mathematical and computational tasks for accuracy and reproducibility.
- Validate reference solutions and automated graders.
- Provide precise, constructive technical feedback to team members.
- Maintain quality standards and ensure rigorous evaluation processes.
Requirements:
- Ph.D., postdoctoral experience, or equivalent advanced expertise in Mathematics, Applied Mathematics, Statistics, or a related field.
- Strong programming skills in Python, Julia, C/C++, MATLAB/Octave, or computer algebra systems.
- Proficiency in Linux and terminal-based environments.
- Hands-on experience with numerical methods, scientific computing, symbolic computation, optimization, computational statistics, or mathematical modeling.
- Strong ability to evaluate mathematical derivations, algorithms, proofs, numerical accuracy, and computational results.
- Experience reviewing research outputs, code, mathematical work, or scientific workflows.
- Prior experience mentoring or leading small technical or research teams.
- Excellent written communication skills and ability to provide precise, constructive technical feedback.
Preferred Qualifications:
- Familiarity with tools such as NumPy, SciPy, SymPy, SageMath, Mathematica, or mpmath.
- Experience with optimization frameworks like CVXPY, JuMP, or Pyomo, or numerical tools for solving differential equations.
- Knowledge of formal verification and proof assistants such as Lean, Coq, or Isabelle.
- Experience with Docker, Git, continuous integration systems, and automated testing.
- Experience designing mathematical benchmarks, competition problems, or computational evaluations.
- Experience evaluating AI coding, reasoning, or terminal-based agents.
- Mathematics olympiad or competition problem-setting and judging experience.
- Publications or open-source contributions in computational mathematics or numerical software.
- Prior experience in AI training, data generation, or model evaluation programs.
Project Details:
- Full-time commitment, minimum 40 hours per week, with at least 4 hours overlap with Pacific Standard Time.
- Duration of 3 weeks.
- Contractor assignment (no medical or paid leave).
- Remote work location.
