What the work involves

You are handed model outputs on materials problems and asked to say, with reasons, whether they are correct. That might mean checking a lever-rule calculation against a binary phase diagram, catching a model that has confused a diffraction peak assignment, judging whether a proposed heat treatment would actually produce the described microstructure, or deciding if a stated Young's modulus for a fibre-reinforced laminate follows from the rule of mixtures the model claims to be using. Alongside grading, most contributors write original prompts and reference answers: problems hard enough that a strong model fails them, with a worked solution that a second reviewer can verify independently.

The grading is rubric-driven. You will be scoring on dimensions like factual accuracy, unit and dimensional consistency, whether the reasoning chain actually supports the conclusion, and whether the answer is calibrated about uncertainty rather than confidently inventing a value. Written justification matters more than the score itself — a rating with a vague comment attached is close to useless to the team training the model.

What micro1 screens for

The screen is AI-led and conversational, and it pushes on depth rather than breadth. Expect to be asked to work a problem aloud, then to be challenged on a step — the follow-up is where the assessment happens. Interviewers look for a graduate-level or industrial background in metallurgy, ceramics, polymers, semiconductors, composites or characterisation, and for the ability to state what you do not cover. Claiming the whole of materials science is a red flag; naming your two subfields precisely is not. A short written or recorded task usually follows the conversation.

Logistics

  • Fully remote, asynchronous, no fixed shifts; work is claimed from a task queue.
  • Contract / 1099-style engagement, paid per hour of logged reviewed work.
  • Volume varies by project — many contributors treat this as 5–15 hrs/week alongside other work; some ramp higher when a project is live.
  • Rates in the $80–130/hr band have been observed on this platform and are set per project and per assessed skill level, not guaranteed.
  • Expect to sign an NDA and to keep task content confidential.