Physics AI Training Expert - Theoretical Evaluation

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help researchers and specialists find paid projects, build a lasting portfolio of AI training work, and manage opportunities in one place. Creating an OpenTrain account is free.

Why AI training matters

AI training (also called data labeling or human feedback work) is the human side of building intelligent systems: people prepare, check, and critique examples that modern models learn from. Work in this area is remote, flexible, and a fast-growing way to apply domain expertise directly to how AI behaves.

This role is a chance to shape scientific reasoning in AI by providing rigorous, technical feedback on physics content that models will use as training signals.

The role

OpenTrain is recruiting a Physics AI Training Expert to evaluate physics solutions, mathematical derivations, and theoretical arguments used to train and evaluate AI systems. You will deliver precise, technical reviews and structured feedback so submitted work can be iteratively improved.

This is a contractor, part-time engagement: approximately 10 hours per week for 8–10 weeks. Pay is $80–$150 per hour (USD); estimated project earnings range from $6,400 to $15,000 depending on exact hours and rate within the stated range.

• Contract type: Contractor, Part-time
• Time: ~10 hours/week for 8–10 weeks (less than 20 hours/week)
• Locations: US, CA, GB (not worldwide)
• Language: English

What you'll do

Your day-to-day work is focused on technically precise evaluation and clear written feedback. Tasks are designed to let you apply research experience and computational validation tools to assess scientific claims.

• Critically evaluate physics solutions, derivations, and theoretical arguments produced for AI training.
• Detect errors, unjustified steps, missing assumptions, dimensional inconsistencies, and logical or methodological weaknesses.
• Differentiate substantive scientific issues from stylistic or cosmetic matters and provide technically precise feedback.
• Use LaTeX, SymPy, Python, and Jupyter to independently verify or counter-check calculations and claims.
• Deliver structured feedback aimed at iterative improvement of submitted work.

Requirements

All requirements below come from the project specification and are necessary to perform the work at the expected level of technical rigor.

• PhD in physics with an active record of independent research in a specialized subfield (examples: High Energy Physics, Biophysics, Condensed Matter, AMO/Quantum Optics, Gravitation/Cosmology, Quantum Information).
• Experience as a postdoctoral researcher, research fellow, junior/assistant professor, or senior research scientist.
• Recent (within ~5 years) representative publications in the relevant subfield, with arXiv or DOI links provided.
• Advanced proficiency with LaTeX, SymPy, Python, and Jupyter for theoretical modeling and computational validation.
• Demonstrated experience reviewing others' work (peer review, supervision, dissertation committees, group seminars).
• Exceptional written communication skills able to convey nuanced, constructive feedback with technical rigor.

Who should apply

Researchers who actively publish and supervise work in theoretical or computational physics and who enjoy careful critique and reproducible verification will find this role rewarding. Prior paid AI training experience is not required; the primary qualification is strong real-world subject-matter expertise.

How it works

OpenTrain manages hiring and contracting for this role. Create a free OpenTrain account, complete your profile, and apply through the platform. Be prepared to share links to recent publications (arXiv or DOI) and examples of your reviewing or supervisory work.

Assignments will require using LaTeX and computational notebooks to check derivations and to write structured feedback. You will be paid at the hourly rate agreed within the stated range and billed through OpenTrain per the platform's contractor workflow.