CritPt is a public benchmark of research-level physics problems, and this panel needs physicists who have published on torsionful gravity, Nieh-Yan or Chern-Simons inflation, or photohadronic blazar emission to create, solve, review or audit those problems. Assignments are agreed after a narrow subfield match, run about 10 hours a week for 8–10 weeks, and every deliverable must be complete enough for another specialist to verify independently.

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

The work

CritPt tests whether frontier models can carry out genuine physics research reasoning rather than textbook problem solving. Depending on where your publication record sits, you may be asked to author research-level problems in your own subfield, produce reference solutions, review someone else's completed work, or audit a finished item end to end. The assignment is fixed with you once you are matched to an area, not before.

This is research-grade output, not volume output. A problem you write has to be unambiguous and solvable by a specialist; a solution you write has to carry its derivation, its conventions and its numerical setup clearly enough that another physicist in the same subfield can reproduce it without asking you a question. Written reasoning is part of every assignment, in LaTeX, usually with Python/SymPy and Jupyter behind it.

What the screen actually checks

The panel covers four narrow areas: Einstein-Cartan/Nieh-Yan axion inflation (perturbations and power spectrum), the same in FRW background dynamics, dynamical Chern-Simons gravity in Palatini formalism, and high-energy astrophysics of blazar jets with photohadronic emission and multimessenger constraints. Matching is deliberately narrow.

Papers, not adjacency. Work in modified gravity does not qualify if your papers are not on torsion, Nieh-Yan or the Pontryagin coupling. This is stated as the most common reason strong physicists are declined.
Verifiable record. Three to five representative papers with arXiv IDs or DOIs, ideally within five years, first author strongly preferred. Each selected area needs your own paper attached as proof — a selection without proof is not scored. Authorship is checked against the public record.
Convention discipline. Tetrad and spin-connection sign conventions are a live technical filter here, not a formality.
Honest gaps. LaTeX, Python, SymPy and Jupyter are expected; agentic coding in VS Code is a nice-to-have. Declaring a gap is explicitly acceptable.

Logistics

Remote and asynchronous, no fixed hours, roughly 10 hours per week sustained across an 8–10 week window starting immediately. English at B2 or above, including written argument. After the form, papers are verified; then either direct onboarding or a short live alignment call to agree the area. A 30–45 minute assessment may be added in some cases, but most applicants will not see one. Observed pay is $80–110/hr, set by depth of subdomain match.

Gravitation, Cosmology and Astrophysics Physicist (PhD)

The work

What the screen actually checks

Logistics

Interview prep — 10 questions