STEM PhD AI Fellowship (Remote, Europe, UK)

What You'll Do7

• Draft prompts that test how well an LLM reasons through problems in your scientific field.
• Review model answers and score them against the standards of your discipline.
• Write evaluation criteria that other reviewers can follow when judging output.
• Spend part of your hours researching topics tied to your own interests, supported by AI tools.
• Flag errors, gaps, and weak reasoning that a model produces on technical questions.
• Turn dense research knowledge into clear task instructions that non-specialists can apply.
• Track your hours and complete project milestones while working on your own schedule.

Requirements6

• Enrolled doctoral student, recent PhD graduate, or postdoctoral researcher in Mathematics, Chemistry, Biology, Physics, or Engineering.
• Resident in Europe or the United Kingdom.
• Strong written English, with a record of explaining complex research in plain, structured form.
• Confidence that your subject knowledge outmatches what current AI systems can do in your field.
• Ability to commit around 10 to 20 hours per week to project work.
• Comfort working on your own without fixed hours or direct supervision.

Who Should Apply

Researchers in the middle of a PhD, recent graduates, and postdocs in math, chemistry, biology, physics, or engineering fit this work well, because the tasks demand real depth in a scientific field. You should also write well enough to break a research concept into steps another person can follow. Candidates based outside Europe and the UK cannot take part, so confirm your location before applying. Applications also lose ground when the field of study sits outside the five listed disciplines, or when a writing sample stays vague instead of showing how you would explain a technical idea. Hesitating about your own expertise hurts too, since the work asks you to judge whether a model's answer holds up in your field.

Salary Insight