AI Trainer (Physical Sciences)

Physical Sciences Trainer, Part-Time AI Evaluation - Remote (52 countries)

• $60–$120 per hour equivalent, paid per accepted task
• 100% remote
• Part-time independent contractor
• Under 20 hours a week, you choose when

Bring physical-science expertise to realistic AI tasks and assess whether generated calculations, code, simulations, and technical analyses are accurate and reproducible.
OpenTrain AI is hiring for this role.

AI training work needs scientists who can define sound methods and identify invalid results. You will create grounded task instructions, reference solutions, evaluation criteria, and clear judgments about the physical and computational quality of model outputs.

What you'll do
• Design computational physical-science challenges that evaluate calculations, code, simulations, models, and technical analysis
• Prepare repeatable workspaces, answer examples, scoring rules, rubrics, and graders.
• Examine model-written Python, simulations, calculations, models, and related technical work.
• Flag bad premises, numerical mistakes, impossible physical outcomes, unsound approaches, and failures to reproduce results.

Why this is worth your time
• Earn $60–$120 an hour equivalent while building real AI skills
• Apply physical-science judgment to structured technical evaluation and reproducible research tasks
• Build practical experience explaining why model methods and results are scientifically sound or flawed
• Fully remote and flexible, no fixed shifts, no commute

Who this is for
• Have expert-level experience assessing or training AI systems on scientific or technical material.
• Strong practical proficiency in Python
• Bring deep knowledge in at least one physical-science field, such as scientific simulation, computational chemistry, computational physics, chemistry, materials science, astronomy and cosmology, or physics.
• Have practical experience using modeling, simulation, numerical or data analysis, or reproducible research methods.
• Point to public work such as experiments, datasets, models, benchmarks, open-source contributions, research profiles, analysis or simulation code, preprints, computational publications, or reproducible notebooks.
• Experience developing AI tasks, answer keys, rubrics, graders, or benchmark problems is useful.
• Familiarity with scientific libraries such as JAX, PyTorch, Astropy, NumPy, SciPy, or SymPy is valuable.
• Professional fluency in English

Pay and details
• $60–$120 per hour equivalent, paid per accepted task, as an independent contractor
• Part-time, flexible schedule; fully remote; eligible in AE, AR, AT, BD, BE, BH, BR, CA, CH, CL, CO, DE, DK, DZ, EG, ES, FR, GB, GH, GR, ID, IN, IT, JO, JP, KE, KG, KW, KZ, LB, LK, MA, MX, MY, NG, NL, OM, PE, PH, PK, PS, PT, QA, SA, SG, TH, TN, TW, US, UZ, VN, and ZA.

How to apply
1. Apply on OpenTrain (about two minutes; upload your resume): https://www.opentrain.ai/jobs/computational-physical-sciences-ai-trainer--cmw9ij2ci00e207lfj2wegvxk/?ch=qajobsboard-jobboardly&sub=p_qajobsboard-jobboardly-cmw9ij2-20261116
2. Complete the application on the hiring platform.
3. If shortlisted, take a conversational AI interview with technical questions and a coding challenge, whenever it suits you.

This is that work, done remotely around your schedule.

About OpenTrain AI
OpenTrain AI connects experienced professionals with paid, part-time AI training work: creating scientific examples, reviewing model outputs, and improving the reliability of AI systems.