What the work actually involves
You write and evaluate expert-level materials science content, not lab work. A typical shift mixes several task types: authoring hard problems in your specialization (a phase-purity puzzle from an XRD pattern, a transport-measurement artifact, a thermal budget conflict in a 2.5D packaging flow), writing the reference solution with reasoning that survives scrutiny, and grading model attempts against a rubric. Other days you'll structure existing knowledge — a synthesis procedure, a characterization workflow, process-window data — into clean, model-readable form, with units, conditions, and assumptions made explicit. Rankings and ratings require written justification; "this one is wrong" is not a deliverable, "the model assumed Tc onset where the rubric specifies zero-resistance Tc, so the 12 K figure is inconsistent with the plotted data" is.
What the screen looks for
Mercor's process is AI-led: a recorded conversational interview plus resume parsing, followed by paid or unpaid sample tasks. The interviewer probes depth rather than breadth — expect to be asked what instrument you personally operated, what you did when a growth run drifted, and how you distinguished a real feature from an artifact. Vague answers get follow-ups. Reviewers also weight written clarity heavily, since most of your output is prose: concise technical explanation, correct terminology, and the discipline to flag uncertainty rather than bluff. Claiming a technique you've only read about is the most common way candidates fail the second round.
Fit and logistics
- Remote, US-based, asynchronous — you pick hours inside task deadlines, which usually run a few days.
- 10 hours/week minimum, up to 40; long-term and ongoing rather than a one-off batch.
- $84/hr is the observed rate for this listing, not a guarantee — Mercor sets pay per project and per candidate, and rates can shift with task type or seniority tier.
- Strongest fits hold a PhD or MS in materials science, chemistry, physics, or a related engineering field, with genuine hands-on time at the bench, the beamline, or in the cleanroom.
If you enjoy explaining why a result is wrong, and you can do it in four sentences, this work suits you. If you're looking for open-ended research with no rubric, it won't.