Computational Physics & Astrophysics Researcher, Task Author

You will turn computational research you have personally implemented, run, debugged, and validated into research-grade tasks for evaluating AI models. Tasks may involve simulations, numerical solvers, data-reduction pipelines, model fitting, and other scientific-computing problems.

Each task should be grounded in real physics, have results that can be checked reproducibly, and include enough detail for strong solutions to be judged without withheld information.

• Design an original main problem with connected subproblems.
• Write a Python reference solution and tests that check the results.
• Define clear metrics and standards for evaluating submitted solutions.
• Revise tasks in response to structured review feedback.
• Optionally review and calibrate tasks submitted by other experts in your field.

This role is suited to someone with research experience in physics, astrophysics, applied physics, or a closely related field. You should be able to turn an authentic computational research problem into a clear task with a reference solution and tests.

• Background: Research-track Bachelor's study or research experience, current Master's or PhD study, postdoctoral experience, or candidate experience in a relevant field.
• Coding: Hands-on experience writing your own code, preferably in Python, to simulate or analyze physical systems.
• Work style: Ability to work independently and manage your own time as a flexible contractor.
• Location: Remote, with eligibility limited to the countries supported for this role.
• Contract terms: You must be able to work as an independent contractor without visa sponsorship.
• Eligibility: F-1 students, including OPT and CPT participants, are not eligible.
• Useful experience: Condensed-matter or statistical physics, atomic, molecular, and optical physics, photonics, electromagnetics, astrophysics, cosmology, gravitation, nuclear or particle physics, plasma, fluid dynamics, nonlinear physics, or continuum mechanics.
• Helpful evidence: Public code, a thesis, or papers that demonstrate computational work.

This is remote, flexible independent-contractor work. You choose when to work, with no fixed weekly schedule and no stated cap on the number of tasks you may take on.

• Schedule: Flexible, with no fixed weekly hours.
• Work arrangement: Remote independent-contractor work.
• Task volume: No stated cap on the number of tasks.

Apply on OpenTrain with your resume, then complete the application on the hiring site. OpenTrain is where people find AI training and data-labeling roles and begin their application.

AI training is the human work behind systems that learn from examples, including writing and evaluating model outputs, checking code, and testing whether results are accurate. OpenTrain helps people find these roles, build a profile, and apply to projects across the industry.