AI Trainer - Mathematics Expert (Remote, Freelance)

[About the Role]
We are looking for experienced mathematics professionals to help train and evaluate advanced AI models so they can reason like real-world math experts. In this role, you’ll use your mathematical expertise to design challenging problems, review AI-generated solutions, and define clear evaluation standards across a range of mathematical domains.

This is a flexible, remote freelance role that can be done alongside your existing research, teaching, or professional commitments.

[What You’ll Do]
As a Mathematics Expert – AI Trainer, you will:
• Evaluate AI-generated answers to math questions for correctness, logical rigor, and clarity of reasoning.
• Design realistic math problems, prompts, and multi-step exercises (e.g., algebra, calculus, probability, discrete math, and applied modeling).
• Create scoring rubrics and structured evaluation guidelines to assess model performance in your area of expertise.
• Verify and, when necessary, correct step-by-step reasoning, notation, and final answers.
• Provide high-quality written feedback to help improve the problem-solving strategies and mathematical communication of AI models.
• Collaborate with AI teams to refine tasks, datasets, and evaluation methods over time.

[Requirements]
Bachelor’s, Master’s, or PhD in Mathematics, Applied Mathematics, Statistics, or a closely related quantitative field.

At least 2–3 years of research, teaching, or professional experience in one or more of the following areas:
• Algebra and Geometry (including pre-algebra, algebra, intermediate algebra, Euclidean/analytic geometry)
• Calculus and Analysis (single-variable, multivariable, real/complex analysis)
• Probability and Statistics
• Discrete Mathematics (combinatorics, graph theory, number theory)
• Applied Mathematics or Mathematical Modeling

Strong grasp of core mathematical skills such as:
• Rigorous proof-writing and logical reasoning
• Step-by-step problem solving and error checking
• Translating real-world scenarios into mathematical models
• Using precise notation and clear structure in solutions

• Native or professional proficiency in Traditional Chinese (TW).
• Excellent written English, with the ability to explain complex mathematical ideas clearly and precisely to a non-expert audience when needed.

[Preferred Requirments]
• Master’s or PhD in Mathematics, Applied Mathematics, or a closely related field.
• Experience designing or reviewing exams, olympiad-style problems, problem sets, or curriculum materials.
• Prior experience using or evaluating AI tools (e.g., ChatGPT, Claude, Gemini) in mathematical or educational workflows.
• Python skills with scientific libraries (e.g., NumPy, SciPy, SymPy) or experience in computational/ numerical mathematics.
• Experience with data labeling, RLHF, or QA for LLMs in math-heavy contexts is a strong plus.

Once we review your CV and determine that you may be a good fit for the role, we will contact you by email and provide further details about the interview process.