Earth, Ocean And Climate Science Researcher (English)

You will turn computational research you have personally run and debugged into clear tasks for evaluating AI models. The work can cover geophysical inversion, ocean or atmosphere modeling, data assimilation, remote sensing, hydrology, geodynamics, or climate science.

Each task should reflect real scientific reasoning, implementation, and validation. You may also review and calibrate tasks created by other experts in your field.

• Design original evaluation tasks based on your own computational research.
• Build a main problem with connected subproblems.
• Write a Python reference solution and tests that check the result.
• Define useful measures for judging solution quality.
• Revise tasks after structured review feedback.
• Optionally review and calibrate tasks from other subject-matter experts.

This is flexible, task-based independent-contractor work. There is no fixed weekly schedule, and there is no stated cap on the number of tasks.

• Pay: $100 per hour.
• The role description states $200 per approved task, with each task taking about two hours.
• Location: Open to candidates in the listed eligible countries.
• Language: Fluent English.
• Experience: Research-track bachelor's students with research experience, current master's or PhD students, postdoctoral researchers, and researchers in a related field may apply.
• Background: Geophysics, oceanography, atmospheric science, geology, hydrology, or a closely related field.
• Skills: Hands-on experience writing code, preferably Python, to model or analyze Earth, ocean, or atmospheric systems.
• You must be able to turn computational research into a clear task with a reference solution and tests.
• You must work independently, manage your own time, and work as an independent contractor without visa sponsorship.
• Helpful experience includes data assimilation, ocean or atmosphere modeling, seismic inversion, public research code, a thesis, scientific papers, data annotation, data labeling, or AI and machine-learning evaluation.

Apply on OpenTrain with your resume, then complete the application on the hiring site. OpenTrain is free to join and helps people find AI training and data-labeling projects in one place.

AI training is the human work behind systems that learn from examples. In this role, your scientific tasks and reference solutions help assess whether AI models can reason through real computational problems.