What the work involves
You author original research problems in materials science that run as code. Each task starts from source material you choose — a published paper, an open dataset, a repository, or a scenario you construct — and becomes a prompt with a deterministic, executable answer. You then write the grading criteria: the checks that decide whether a model's output is scientifically correct, not merely plausible. Before a task ships, you calibrate it against frontier models. If strong models solve it reliably, it does not ship, and you revise or discard it. In practice this means a lot of iteration: the first draft of a problem is usually too easy, and the interesting work is finding the failure mode that is real physics rather than a trick of phrasing.
Expect tasks touching semiconductor materials (band structure, defect energetics, transport, device-relevant thermodynamics) and molecular modeling (MD, DFT workflows, force fields, structure generation and analysis). The listing asks for demonstrated depth in both, with a coding focus — this is not a writing role with a Python garnish.
What the screen looks for
- Verifiable domain depth: named methods, codes you have actually run (VASP, Quantum ESPRESSO, LAMMPS, ASE, pymatgen, GROMACS), and what you did with them.
- Evaluation judgment: can you tell a hard problem from an ambiguous one, and can you write a grader that fails a wrong answer for the right reason.
- Engineering hygiene: Git branching, pull requests, running code in Docker, reproducing an environment someone else will execute.
- Honest availability against a 20+ hour weekly commitment with an immediate start.
Logistics
Fully remote and asynchronous. Six-week engagement, part-time at 20+ hours per week, immediate start. Authoring goes through pull requests with automated quality checks, so your throughput depends partly on how cleanly your submissions pass review. The observed rate for this listing is $70/hour; Mercor sets rates per contributor and per project, and it is not guaranteed. Screening is a 25-minute conversational interview covering background, experience and motivation, followed by next steps within a few days.