
Physician, Medical AI Evaluation
Posted 3 days ago

Posted 3 days ago
This is a fully remote position, open to applicants in Minnesota.
• Assess clinical EHR vignettes alongside corresponding questions, suggested answers, and distractor trap answers related to diagnosis and treatment tasks.
• Evaluate the quality of clinical reasoning, accuracy and completeness of vignettes, correctness and gradeability of answers, and quality of traps.
• Determine if reasoning can arrive at the answer based solely on the facts presented in the vignette.
• Correct reasoning pathways and compose concise rationales.
• Engage in three independent blind reviews followed by physician adjudication.
• Contribute to a project-based medical AI evaluation initiative aimed at enhancing AI tools.
• Must be a US-based, actively practicing physician holding an MD or DO.
• Possess an active, unrestricted US medical license, confirmed by the issuing medical board.
• Board certification or completion of residency training is required.
• Recent experience in direct patient care is preferred, ideally at the attending level.
• Extensive multi-system diagnostic and treatment knowledge in internal, family, emergency, or hospital medicine.
• Strong clinical reasoning skills, including differential diagnosis, next-step management, and adherence to evidence-based guidelines.
• Previous experience in medical AI evaluation, clinical content or exam-question review, or medical annotation/QA is advantageous.
• Proficient written English skills (C1+ level).
• Not suitable for medical students, unlicensed medical graduates, non-physician clinicians, or individuals with non-clinical healthcare titles.
• Tasks must be completed by specified deadlines and fulfill the stated acceptance criteria.
• Participation is project-based.
• Offers part-time, non-permanent project opportunities.
• Flexible workload estimated at approximately 10–20 hours per week during active project phases.
Mindrift
Mindrift
ClickHouse
1mind
Get handpicked remote jobs straight to your inbox weekly.