Remote | Research Physicist — $75–$130/hour

We are sharing a specialised part-time consulting opportunity for advanced physics researchers with expertise in solving, verifying, and adjudicating research-level problems across frontier areas of theoretical, computational, and experimental physics.

This role supports the development of a research-level physics benchmark used to evaluate advanced AI systems. Selected researchers will produce fully verified reference solutions, audit submissions from other experts, decompose complex challenges into assessable checkpoints, and develop reproducible computational tools for symbolic and numerical evaluation.

Key Responsibilities

Research-Level Problem Solving

• Solve advanced physics challenges from initial formulation through final verified result
• Produce rigorous derivations supported by calculations, code, and peer-reviewed references
• Apply appropriate analytical, numerical, and computational methods within the relevant subfield
• Validate assumptions, approximations, boundary conditions, and final conclusions
• Document solution methodology clearly enough for independent expert verification

Reference Solution Development

• Author high-quality reference solutions for complex physics benchmark tasks
• Decompose research challenges into standalone checkpoint problems requiring genuine physical reasoning
• Define acceptable symbolic forms, numerical tolerances, and equivalent solution approaches
• Create verification test cases covering expected results and relevant edge conditions
• Ensure reference materials are technically complete, internally consistent, and reproducible

Python & Automated Evaluation

• Develop Python-based answer templates and automated grading functions
• Support symbolic and numerical answer checking using tools such as SymPy
• Build reproducible computational workflows in Jupyter
• Validate grading logic against equivalent expressions, numerical precision requirements, and alternative valid methods
• Identify unintended shortcuts or grading gaps that could weaken benchmark reliability

Solution Audit & Adjudication

• Audit expert-submitted solutions for correctness, scope, and methodological soundness
• Identify derivation errors, unsupported assumptions, missing cases, and verification weaknesses
• Provide detailed, actionable feedback across revision cycles
• Compare parallel solution attempts and determine which should serve as the final reference
• Resolve disagreements involving methodology, interpretation, numerical results, or symbolic equivalence

Ideal Profile

Strong candidates may have:

• A PhD, postdoctoral background, or comparable research experience in a relevant physics subfield
• Active research expertise in high-energy physics, mathematical physics, biophysics, statistical physics, condensed matter, atomic, molecular and optical physics, gravitation, cosmology, astrophysics, quantum information, or related areas
• Hands-on familiarity with at least two established methods used within their specialist field
• A publication record demonstrating recent research expertise
• Strong proficiency in mathematical derivation, computational verification, and scientific literature review
• Working knowledge of LaTeX, Python, Jupyter, and SymPy
• Excellent written English and meticulous attention to technical detail
• Availability to contribute approximately 10 hours per week over an 8–10 week task window

Educational Background

• A PhD in physics or a closely related discipline is highly relevant
• Senior PhD candidates may be considered for solver-level work
• Postdoctoral researchers or junior faculty may be suited to auditing responsibilities
• Senior postdoctoral researchers, professors, or industry research leaders may be considered for adjudication work
• Equivalent senior research experience supported by publications and demonstrable technical expertise may also be considered

Nice to Have

• Three to five representative publications within the relevant physics subfield
• Recent publications supported by arXiv identifiers or DOIs
• Experience peer reviewing research papers, supervising graduate work, or assessing advanced technical solutions
• Expertise spanning multiple physics subdomains or canonical solution methods
• Experience developing computational physics benchmarks or automated grading systems
• Familiarity with symbolic computation and numerical validation
• Background creating research-level scientific problems or reference datasets
• Previous involvement in AI evaluation, benchmark design, or technical training-data projects

Why This Opportunity

• Solve and verify frontier research-level physics problems
• Help establish trusted reference data for evaluating advanced AI reasoning
• Work across rigorous derivations, scientific computing, peer review, and benchmark design
• Contribute within a specialist subfield aligned with your publication and research background
• Participate in asynchronous, flexible work with competitive hourly compensation
• Take on solver, auditor, or adjudicator responsibilities based on experience

Contract Details

• Independent contractor role
• Fully remote with asynchronous scheduling
• Expected commitment of approximately 10 hours per week
• Typical task-pool duration of approximately 8–10 weeks
• Competitive rates between $75–$130 per hour depending on role, research background, and demonstrated expertise
• Weekly payments via Stripe or Wise
• Work may include problem solving, reference-solution development, computational verification, peer auditing, and adjudication
• Projects may be extended, shortened, or adjusted depending on scope and performance
• Work will not involve access to confidential or proprietary information from any employer, client, or institution

About the Platform

This opportunity is available through 24-MAG LLC. We connect experienced professionals with remote consulting opportunities across technical, evaluation, and project-based workstreams.

By submitting this application, you acknowledge that your information may be processed by 24-MAG LLC for recruitment and opportunity matching in accordance with our Privacy Policy: https://www.24-mag.com/privacy-policy.