The work
Create and review physics problems used to improve AI reasoning. You will assess whether AI responses are scientifically accurate and logically sound, then give clear feedback to improve the questions and evaluation materials.
- Design well-structured problems from undergraduate to advanced difficulty.
- Write questions that test reasoning and conceptual understanding, not just calculation.
- Check solution accuracy and refine the clarity and difficulty of physics prompts.
- Review AI-generated answers for scientific rigor and give detailed feedback.
- Apply current scientific principles to complex physics challenges.
What it pays and takes
- Pay: $65 per hour.
- Time: 20+ hours per week.
- Work arrangement: Remote, flexible, asynchronous contractor role.
- Location: Open to candidates in the countries listed for this role.
- Language: Fluency in English.
- Education: A completed master's or PhD in physics or a related field, such as astronomy, astrophysics, or engineering physics.
- Experience: At least two years in academic research, teaching, or applied physics.
- Knowledge: Undergraduate mechanics, electromagnetism, thermodynamics, and quantum mechanics; strong analytical skills and clear scientific communication.
- Helpful: Scientific publication, curriculum development, technical report writing, interdisciplinary projects, or exposure to AI or machine learning.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training is the human work of creating and reviewing examples that help AI systems learn. Physics experts help make those examples and evaluations scientifically accurate and rigorous.
Listed by OpenTrain AI. Details can change, so check the original posting before applying. View original posting