All AI training jobs

Meridial · STEM & research

Pure Mathematics Specialist

Listed on Meridial as “Pure Mathematics Specialist - Freelance AI Trainer Project”

$35-$65/hrRemoteContractPaid in USD
ShareWhatsAppTelegramEmail

How this rate compares

This listing

$35–$65 per hour

Typical in STEM & research

$40–$65 per hour

Within the typical range

Typical: the middle of the lowest and highest rates posted by 193 listings in this category on MantaWork, in US dollars.

What this actually is

You design problems that stump current AI models, evaluate AI reasoning against the correct answer, write rubrics, and provide expert feedback. Often the highest-paid category because the expertise pool is small. The platform title (Pure Mathematics Specialist - Freelance AI Trainer Project) reflects the rate band and the expertise required, not the day-to-day work.

Original posting from Meridial

Work location: Remote, open worldwide

Pay: $35 to $65 an hour, as the posting states it. Meridial sets the rate after reviewing your experience and location.

Are you a theoretical mathematics expert eager to shape the future of AI? Large-scale language models are rapidly advancing, and their ability to handle rigorous proof-based reasoning depends on high-quality mathematical training data. With your advanced expertise, tomorrow’s AI can better understand abstraction, logic, measure theory, homotopy, and the deep structures of modern mathematics. That training data begins with you - your insight will help power the next generation of AI.

We’re looking for specialists with deep foundations in pure mathematics who can engage with advanced topics across probability theory, measure theory, real and functional analysis, optimization, abstract algebra, and category theory. Coupled with this deep foundation, we need each expert to have a normalized fluency in Lean 4, the proof assistant and programming language.

On a typical day, you will construct and evaluate complex proofs, substantiate the mathematical reasoning for correctness and rigor while identifying subtle conceptual or logical failures, and translate this material to successfully build in .lean syntax. You will also collaborate with our team to refine prompts, evaluation frameworks, and mathematical reasoning benchmarks, enhancing the overall quality of our operations.

Candidates fluent in Lean 4 with an in-progress or freshly earned PhD or current PostDoc experience in pure mathematics are ideal. A strong command of core areas is required, and applicants should be comfortable with topics such as Markov’s and Chebyshev’s inequalities, Hoeffding bounds, convergence theorems, the Carathéodory extension theorem, the Riesz representation theorem, Banach-Alaoglu, Lax-Milgram, stochastic approximation, Van Kampen’s theorem, the Yoneda embedding, and representability in cohomology. Clear, metacognitive communication (explicitly articulating reasoning and proof strategies) is essential.

Ready to turn your Lean 4 fluidity and pure math expertise into the reasoning backbone of tomorrow’s AI? Apply today and help teach the model that will support researchers, scientists, and engineers around the world.

We offer a pay range of $35 to $65 per hour, with the exact rate determined after evaluating your experience, expertise, and geographic location. Final offer amounts may vary from the pay range listed above. As a contractor, you’ll supply a secure computer and high-speed internet; company-sponsored benefits such as health insurance and PTO do not apply.

Job title: Pure Mathematics and Lean 4 Specialist - Freelance AI Trainer Project

Employment type: Contract

Workplace type: Remote

Quoted from Meridial's public listing on 2026-11-15. We don't edit platform copy; honest framing is in the title and the “what this actually is” block above.

More on this platform

About Meridial

The marketplace run by Invisible Technologies for AI training projects. The same project is posted once per country, so look for the listing in your country.

All Meridial jobs

See all AI training jobs

Browse by category and compare across all eight platforms we cover.