Scientific Computing Research Expert

Researchers who write Python for their own scientific work can earn per task by turning that work into evaluation problems for AI models. It suits PhD students, postdocs, faculty and research-active Master's holders across the life, physical, earth, mathematical and engineering sciences. The role is with an AI training company that builds pools of vetted experts, which project teams then draw from.

What you'll do

• Build original, research-grade evaluation problems from computational work you have personally carried out, such as simulations, model fitting, data reduction or numerical solvers
• Define how a strong solution should be scored, while keeping the exact pass thresholds private
• Rework submissions after structured reviewer feedback; one or two revision rounds is typical
• If you choose, assess and calibrate other experts' submissions in your field

What you'll need

• A PhD in progress or completed, with ongoing hands-on research, or a research-active Master's (thesis enrollment, an RA or lab role, a publication, or a clear path toward a PhD)
• Enough Python to execute and troubleshoot your own computational code
• The self-direction to manage your time without a set weekly schedule
• Scientific computing, simulation or AI evaluation experience is a plus

Good to know

• Independent contractor work paid at $300 per approved task, each taking about four hours
• No fixed hours and no limit on the number of tasks you can complete
• Accepted experts go through background and identity checks