Principal Research Scientist, Part-Time AI Research Evaluation - Remote (United States)

Role overview Evaluate AI-generated research in computer science and computational physics through rigorous technical review. In this flexible, part-time remote contract role, you will set demanding research challenges, verify model-produced results with proofs and reproducible computations, and report on progress, errors, and recurring limitations.

Responsibilities - Define challenging open problems, theoretical limits, and simulation tasks for AI models to attempt. - Lead research activities involving proofs, algorithms, numerical methods, and high-performance computing. - Verify proposed results by checking proofs, rerunning computations, and applying formal methods. - Compare model outputs with current research literature. - Write concise technical memos describing partial progress and repeated failure patterns.

Requirements - At least 7–10 years of active research experience in computer science or mathematical or computational physics. - Strong scholarly contributions, including recent publications in leading venues. - Practical experience in simulation, numerical analysis, algorithm implementation, or high-performance computing. - Sound research judgment and the ability to assess work beyond your own specialty. - Fluent English.

Nice to have - A record of resolving recognized open problems or introducing concepts adopted by peers. - Experience with research leadership, such as fellowships, invited or plenary talks, awards, program committees, or editorial boards. - Familiarity with Lean, Coq, SAT or SMT solvers, or public research involving advanced AI systems. - Previous AI training work is not required.

Benefits and work setup - Remote, part-time work available to candidates in the United States. - Flexible schedule without fixed shifts. - Independent contractor compensation of $100–$400 per hour.