Expert Contributor - Mathematician (Task-Based)
About the Role
Snorkel AI is a data development company spun out of the Stanford AI Lab, focused on improving the truthfulness and reasoning abilities of advanced AI systems. This ongoing project-based opportunity allows you to help shape the next generation of AI models by designing challenging tasks, stress-testing AI agents, and providing expert feedback that steers model behavior toward ideal solutions. No prior AI experience is required.
Responsibilities
- Design and solve challenging, real-world reasoning tasks that test the limits of advanced AI systems.
- Create high-quality physics problems and develop rigorous solutions.
- Identify where models fail in scientific reasoning.
- Produce graduate-level, multi-step analytical problems with precise, fully worked solutions.
- Review and evaluate AI-generated scientific reasoning, identifying subtle errors and suggesting more rigorous approaches.
Qualifications
- Deep fluency in Mathematics, Applied Mathematics, Statistics, or adjacent quantitative disciplines.
- Strong review instincts—able to identify subtle scientific errors and evaluate competing solution paths.
- Master’s or PhD candidate, or graduate, in Mathematics or a related discipline.
- Background in one or more of: academic research, university-level teaching, advanced problem writing, laboratory science, or technical scientific communication.
- Bonus: Familiarity with grading rubrics, olympiad- or exam-style question design, or evaluating AI-generated scientific reasoning.
Schedule
Work independently as a remote 1099 contractor with no fixed hours commitment. Must be eligible for 1099 contractor status and cannot reside in California.
Pay
Compensation is per accepted task rather than an hourly rate. Payments for accepted tasks range from $63 to $410, with an average handling time of 60 to 390 minutes. Experts earn at least $15 per hour but have the opportunity to earn significantly more based on productivity and quality.