Physicist Talent Network

$60–80/hrrate as listed

Mercor maintains a standing network of physicists who are matched to AI training and evaluation projects as labs request them. You apply once, pass an AI screening interview, and get contacted when work matching your subfield opens up — typically 15–30 hours per week at an observed $60–80/hr.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

What the work involves

This is an open application to a talent pool, not a specific assignment. Once verified, you may be invited to projects where frontier labs need physics expertise: writing problems that probe a model's grasp of mechanics, electrodynamics, statistical physics, quantum theory or condensed matter; grading model responses against a rubric and explaining exactly where the reasoning breaks; constructing multi-step derivations with defensible intermediate results; or reviewing model-generated code for numerical simulation, data reduction, and error analysis.

The recurring theme is that a plausible-sounding wrong answer is the hard case. Models often produce dimensionally consistent algebra that quietly drops a factor, misapplies a boundary condition, or invokes an approximation outside its regime of validity. The value you add is catching that and articulating it in a sentence a non-specialist reviewer can audit.

What the platform screens for

Verifiable credentials. A PhD or equivalent research track record in physics or a closely adjacent field, with publications, thesis work, or industry R&D you can describe in specifics.
Depth under follow-up. Mercor's AI interview asks a question, then pushes on your answer. Vague framing collapses quickly; be ready to name methods, instruments, codes, and the assumptions you made.
Evaluation judgment. Whether you can separate "wrong" from "unjustified," apply a rubric consistently, and write feedback that is short and actionable rather than a lecture.
Scientific programming. Most projects assume comfort with Python and the standard numerical stack; some want C++, Julia, Mathematica, or lattice/FEM/Monte Carlo experience.

Logistics

Fully remote and largely asynchronous, with work scheduled around your own calendar. Typical commitments run 15–30 hours per week when you are on a project, but the network model means gaps between engagements — this suits people with flexible academic or consulting schedules better than those needing continuous income. The $60–80/hr band reflects rates observed on physics-tagged Mercor work and is not guaranteed; actual rates are set per project and vary with subfield scarcity and task difficulty. Payment is contract-based, and you supply your own equipment.

What they look for

PhD or equivalent research experience in physics or an adjacent quantitative field
Scientific programming in Python (NumPy/SciPy) or comparable numerical tooling
Theoretical and computational modeling with explicit attention to assumptions and validity regimes
Experimental design, data analysis, and uncertainty quantification
Clear written technical explanation for a mixed technical/non-specialist audience
Self-directed remote work across asynchronous project timelines

Before you apply

Would you pass this screen?

Drawn from what this listing actually requires. Answered in your browser — nothing is sent to us, and we do not store your answers.

Do you hold a bachelor's degree or higher in physics, or can you demonstrate 5+ years of professional scientific research or engineering experience in physics or a closely related quantitative field?
Physics degree or equivalent experience · required
Have you written and debugged scientific code (in languages such as Python, C++, Julia, MATLAB, or similar) for computational modeling, data analysis, or simulation as part of professional or research work?
Scientific programming experience · required
Are you located in a country where Mercor currently operates, and can you work in your local time zone as a contractor?
Work location · required
Can you commit to work in the range of 15–30 hours per week on a contract basis, with flexibility for project-to-project variation?
Availability for 15–30 hours per week · preferred
Have you worked remotely or independently on research or technical projects where you were responsible for managing your own time and communicating progress asynchronously?
Remote work experience · preferred
Are you able to write and speak English clearly enough to document technical work, take direction from AI labs, and participate in interviews conducted in English?
English communication · preferred

0 of 6 answered

Included with this listing

Interview prep — 67 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: Whether the candidate has hands-on depth in theoretical and computational modeling, not just coursework familiarity; whether they understand the gap between model and reality and how to test it.
A strong answer: The answer names a specific physics domain (e.g. fluid dynamics, quantum systems, particle interactions), describes the mathematical or numerical approach they chose or implemented, and explains how they verified correctness—against experiment, analytical solutions, or known benchmarks.
Where people lose it: Vague description of 'running simulations' without explaining the underlying physics, the tool choices, or any validation step. Or claiming expertise in modeling but unable to articulate why their approach was sound.