Physics
Physics research, evaluated by physicists
How physical judgment shows outside a department, and the standard the founding cohort will hold it to. Everything below is a draft in public, on purpose.
What counts as evidence of physics skill
Physics has the oldest preprint culture in science — arXiv has been the field's working publication venue since 1991 — so public evidence of skill is unusually rich. A carefully argued preprint, a simulation codebase with honest convergence tests, or an instrument paper that documents its systematics tells an evaluator more than the name of the journal it eventually landed in.
Outside universities, physics work hides inside other job titles. The person building the noise model at a sensor company, validating a Monte Carlo pipeline at a fusion startup, or writing the error budget for a quantum-hardware team is doing physics research whether or not the word appears on their badge. Their artifacts — technical reports, design studies, analysis code — are evaluable even when they are not citable.
The hardest thing to see from a CV is judgment about approximation: knowing which effects to keep, which to bound, and which to ignore, and saying so explicitly. That judgment is exactly what a structured evaluation can surface and a publication list cannot.
Refereeing itself is evidence. Physicists read for dimensional sanity, limiting cases, and whether an effect survives an order-of-magnitude check — habits that transfer directly to evaluation. A record of careful review, journal-club notes that dissect a claimed detection, or a public post-mortem of a result that did not hold all demonstrate the reading skill this community runs on.
The physics evaluation rubric, first draft
Physics evaluation weighs how a claim is bounded as much as whether it is right; the field's culture of error analysis is this rubric's spine.
- Physical reasoning
- Chooses the right level of description for the problem, and states and justifies its approximations instead of inheriting them silently from prior work.
- Uncertainty accounting
- Separates statistical from systematic uncertainty, propagates both, and is honest about which dominates. An analysis that hides its error budget scores low whatever its conclusion.
- Computational reproducibility
- Where simulation or data analysis carries the result: the pipeline can be rerun, convergence and resolution checks are shown, and the code is inspectable.
- Contact with measurement
- States what observation would strengthen or kill the claim, even when that experiment is out of reach today.
Founding physics evaluators will pressure-test this draft first — including against their own past work.
What founding physics evaluators will do
First, take this rubric apart: where it rewards formalism over insight, where it punishes honest exploratory work, where a dimension cannot actually be scored from a real artifact.
Then run calibration rounds on public artifacts — preprints, analysis code, instrument papers — scoring independently and comparing spreads, so the first scores the platform ever publishes come with a known uncertainty rather than false precision.
As the community opens, those calibration records will seed the weighting system: the first physicists whose evaluation history is itself part of the instrument.
Who this is for
The founding cohort is looking for physicists whose judgment is already in daily use, wherever they happen to practice it:
- Physicists in industry — sensing, photonics, semiconductors, quantum, energy — whose research work never reaches a journal.
- Postdocs and staff scientists who referee regularly and want that judgment to count somewhere.
- Computational physicists whose main artifact is a validated code base rather than a paper.
- Experimentalists who can read a systematics section critically and enjoy doing it.
Who this is not for
Self-selection matters more than any filter we could write, so here is the honest version:
- Anyone after a quick credential — the founding stage produces standards, not badges.
- Researchers looking to promote their own work; evaluation here is of other people's artifacts, under a published rubric.
- Anyone uncomfortable having their evaluation accuracy tracked — the calibration record is the point of the design.
- Anyone who needs a live scoring platform today; the mechanics on this page are in design, and the tense is deliberate.
Apply to evaluate physics
Physics is pre-selected on the application. Attach an arXiv listing, ORCID profile, or repository we can read.