Medicine Expert - Freelance AI Trainer
About the role
Mindrift connects specialists with project‑based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project‑based, not permanent employment. We are seeking US‑based practicing physicians (and other licensed clinicians) to contribute to health‑focused AI training projects at scale, with a primary focus on general medicine and openness to specialists (cardiology, radiology, etc.) and adjacent roles such as nurses, psychologists, and dietitians.
Responsibilities
- Annotate and evaluate AI‑generated health content: review responses to clinical questions, assess medical accuracy and safety, and label outputs against clear quality criteria.
- Create complex, realistic tasks that push frontier AI agents to their limits, incorporating scattered patient data, conditional clinical procedures, and genuine diagnostic reasoning.
- Build a detailed version with objective scoring, then write an ambiguous version intended to train the agent to succeed with less hand‑holding, using real clinical complexity.
Requirements
- Degree in Medicine, or a degree in a related field (nursing, psychology, dietetics, etc.) with clinical training that includes diagnosing and managing medical conditions.
- Current, valid US license to practice (medical license for physicians; appropriate state license for adjacent roles such as RN, licensed psychologist, or RDN).
- Active clinical practice with 3+ years of hands‑on clinical experience (board certification is a plus).
- Strong written English proficiency at C1 level or higher.
- Openness to part‑time, non‑permanent, project‑based work.
Pay
Compensation is paid per accepted task. Your rate depends on the qualification tier you reach and how efficiently you complete tasks — up to the equivalent of $55 per hour. Faster task completion raises your effective hourly rate.
Schedule
Tasks are estimated to require around 10‑20 hours per week during active project phases. This is an estimate, not a guaranteed workload, and applies only while the project is active. Tasks must be submitted by the deadline and meet the listed acceptance criteria to be accepted.