What the work involves

You are handed model outputs — differential diagnoses, treatment plans, patient-facing explanations, exam-style question answers, summaries of clinical literature — and asked to judge them the way an attending judges a presentation. Typical tasks include rating two candidate responses against each other and writing the rationale, flagging hallucinated citations or invented drug interactions, rewriting an unsafe answer into a correct one, or authoring original clinical prompts that expose where a model's reasoning breaks down. Much of the value is in the written justification: a score without a defensible clinical reason is close to useless to the team training the model.

Safety judgment carries disproportionate weight. A response can be factually accurate and still fail — because it omits a red-flag symptom, because it gives dosing without the contraindication, or because it answers a question that should have been redirected to in-person care. Reviewers who can articulate that distinction, consistently and against a rubric they did not write, are the ones who get continued work.

What the platform screens for

micro1 runs an AI interviewer that asks about your training stage, specialty exposure, and licensure, then pushes into clinical detail with follow-ups. Expect it to probe whether your stated experience holds up: naming the guideline you would cite, walking through a workup, explaining why one management pathway beats another. It also tests calibration — whether you can apply someone else's grading criteria rather than substituting your own preferences — and whether your availability claims are realistic. Students and residents are explicitly in scope; the interview is trying to place you accurately, not screen you out for being early in training.

Logistics

  • Fully remote and asynchronous; work is claimed from a queue rather than scheduled
  • Volume is project-driven and uneven — busy stretches followed by quiet ones
  • Most contributors commit 5–20 hours per week around clinical duties
  • Paid hourly at observed rates of $40–90/hr, varying by specialty, training stage, and task difficulty; rates are not guaranteed and are set per project
  • Requires a reliable computer, stable connection, and a quiet environment for the recorded screening interview