Remote Medical QA Physician — Clinical AI Evaluator

Mindrift.ai: Be the “I” in AI; United States; Remote; USD 5,000 - 9,000; Full time

Benefits offered by this job

Fixed per-task payment

$50 onboarding reward after first task

Quality incentives

Job summary

Mindrift.ai seeks physicians to evaluate clinical EHR vignettes and posed questions. You will score items on accuracy, reasoning, and distractor quality, ensuring reasoning aligns with provided facts.

You will review reasoning traces and write concise rationales, then complete adjudication following blind reviews and platform guidelines. Fully remote role with no patient contact; onboarding includes a $50 reward after first approved task and additional quality incentives for high-quality work.

Qualifications

• MD or DO with active, unrestricted US medical license.
• Based in an eligible US state; licensure verified on onboarding.
• Board certification or residency with direct patient care; attending level preferred.
• Broad diagnostic and treatment experience across internal, family, emergency or hospital medicine.
• C1+ written English proficiency.
• Experience in medical AI evaluation or QA preferred.
• Must be a licensed physician; medical students or non-physician clinicians not eligible.

Responsibilities

• Evaluate written clinical EHR vignettes, questions, proposed answers, and distractor answers across diagnosis and treatment tasks.
• Score benchmark items for vignette accuracy, completeness, answer correctness, gradeability, and distractor quality.
• Assess whether clinical reasoning reaches the answer using the facts provided in each vignette.
• Review and correct reasoning traces.
• Write concise rationales for evaluations.
• Complete physician adjudication following independent blind reviews, as assigned.
• Submit completed tasks through the platform in accordance with project acceptance criteria.

Skills

Clinical reasoning

Medical annotation

English proficiency

Education

MD or DO degree

US medical license active

Job description

Mindrift.ai seeks physicians to evaluate clinical EHR vignettes and posed questions. You will score items on accuracy, reasoning, and distractor quality, ensuring reasoning aligns with provided facts.

You will review reasoning traces and write concise rationales, then complete adjudication following blind reviews and platform guidelines. Fully remote role with no patient contact; onboarding includes a $50 reward after first approved task and additional quality incentives for high-quality work.