Physics Expert (Statistical Physics / Quantum Information / Condensed Matter)

$80–160/hrrate as listed

micro1 is contracting physics PhDs to write, critique, and adjudicate technical problems in statistical physics, quantum information, and condensed matter theory for an AI training project. The work centers on disordered spin models, duality arguments, and toric-code error thresholds, with observed rates of $80–160/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

You'll be handed problems that sit at the boundary of analytic statistical mechanics and numerical benchmarking: replicated random-bond Ising and Ashkin-Teller models, the Nishimori line, square-lattice self-duality, and the toric-code threshold. Depending on the assignment and your subfield, you'll work in one of three seats — Solver (produce a fully worked, defensible solution), Auditor (find the error in someone else's derivation, or confirm there isn't one), or Adjudicator (rule between conflicting submissions and write the reasoning that settles it). Deliverables are written artifacts: derivations with stated assumptions, critiques that point to a specific line rather than a vibe, and occasionally simulation code or numerical results for 4-state Potts and domain-wall free energy calculations.

The underlying purpose is AI training data. Model outputs on this material are frequently confident and wrong in ways that require someone who has actually done a quenched disorder average to catch — a misapplied replica limit, a duality mapping that quietly drops a boundary term, a noncontractible loop defect handled as if the topology were trivial. No prior AI or ML experience is required or expected.

What the screen looks for

Real specialization, not adjacency. Expect follow-ups that push past the first correct-sounding answer: where the Nishimori line comes from, why self-duality fixes the critical point on the square lattice, what breaks in the replica argument.
Written precision. Ambiguous notation and unstated assumptions are the main failure mode; the screen samples how you write, not just what you know.
Evaluation judgment. Given a plausible but flawed derivation, can you localize the error and articulate why it matters, rather than rewriting from scratch?
Numerical honesty. If you've run Monte Carlo on Potts models, you'll be asked about finite-size effects, equilibration, and what your error bars actually mean.

Logistics

Contractor engagement, fully remote and asynchronous. Work arrives in batches; most contributors take on a defined number of tasks per week rather than fixed hours, and volume fluctuates with the customer project's needs. Rates in the $80–160/hr band have been observed for this listing, with placement typically reflecting subfield fit and whether you're solving, auditing, or adjudicating — treat these as observed, not guaranteed. Expect an onboarding calibration round before paid work begins.

What they look for

Physics
Kramers-Wannier duality
quenched disorder averaging
square-lattice self-duality
domain-wall free energy
noncontractible loop defects
4-state Potts model numerics

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 PhD in physics, or can you document at least five years of equivalent full-time research experience in statistical physics, condensed matter, or quantum information?
PhD in Physics or equivalent research experience · required
Have you personally written or run code to simulate the 4-state Potts model, and analyzed numerical output such as free energy, order parameter, or correlation length?
Hands-on experience with 4-state Potts simulations · required
Do you have at least one published paper, thesis chapter, or technical report in which you contributed substantively to work in statistical physics, condensed matter physics, or quantum information?
Published or documented research in target subfields · required
Have you applied Kramers-Wannier duality, quenched disorder averaging, or square-lattice self-duality in a research context—not just read about it—and can you explain your application?
Familiarity with Kramers-Wannier duality or quenched disorder · preferred

0 of 4 answered

Included with this listing

Interview prep — 11 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 you have hands-on, publishable-level experience with the exact mathematical tools this role demands, not just textbook familiarity.
A strong answer: Name a concrete problem (a published paper, thesis chapter, or completed project), describe which technique you used and why, and give one specific result or insight that came from applying it—showing you've debugged and interpreted the method in practice.
Where people lose it: Describing duality or disorder averaging in general terms without anchoring to a specific calculation you did, or naming a paper you read rather than one you contributed to.