Physicians
Listing checked October 23, 2026 · pay as published by Handshake AI
Overview
Handshake AI hires physicians to shape how large models handle clinical reasoning. You bring real practice experience in acute care, surgery, or a related specialty and turn it into expert tasks that current models fail, then judge what those models produce. Work covers the arc of a real case: diagnosis, differential reasoning, imaging and test interpretation, treatment planning, and procedural calls. Assignments run on a project basis with no fixed schedule, so you set your own hours alongside a current practice. The program stays open year-round and new projects open as specialty needs come up.
What You'll Do7
- 1Design expert-level clinical tasks that top AI models cannot yet solve, drawn from the workflows you handle in practice.
- 2Build cases around diagnosis, differential reasoning, imaging and test interpretation, treatment planning, and procedural decisions.
- 3Check AI-generated answers for clinical accuracy, patient safety, and consistency with current guidelines and accepted practice.
- 4Write structured feedback that names the specific error and explains the reasoning a practicing physician would use instead.
- 5Flag responses that could lead to patient harm if a clinician acted on them.
- 6Tailor tasks to your specialty depth, whether that is critical care, emergency medicine, surgery, or another listed area.
- 7Complete tasks on your own schedule from a remote setup, with no minimum hours and no set shift.
Requirements9
- 1Completed MD, DO, or an equivalent medical degree.
- 2Finished or current residency or fellowship, or board certification or board eligibility, in one of the listed specialties.
- 3Working clinical knowledge in at least one area such as critical care, emergency medicine, stroke and vascular neurology, interventional cardiology, radiation oncology, infectious disease, OB/GYN and maternal-fetal medicine, or ophthalmology (retina or glaucoma).
- 4Surgical candidates qualify through general surgery, orthopedic trauma and adult arthroplasty, or oral and maxillofacial and craniofacial trauma surgery.
- 5Active US work authorization.
- 6Clear writing that explains complex medical concepts to a non-clinician reader.
- 7Careful attention to detail when reviewing clinical content line by line.
- 8Comfort working independently in a remote, asynchronous setup.
- 9No sharing of protected health information or any identifiable patient data.
Who Should Apply
Board-certified or board-eligible physicians who still practice and can explain their reasoning step by step. Strong fits include intensivists, emergency physicians with POCUS experience, cardiologists reading echoes, and surgeons handling trauma or arthroplasty. Academic clinicians with publications or EHR and clinical decision support exposure stand out.
Salary Insight
$180 per hour applies to this part-time contract work. That sits well above typical clinical review and medical writing rates, which reflects the licensed expertise and specialty depth the projects demand.
Pay and demand for AI Training roles
AggregatedTypical pay
$70/hour
This role
$180/hr
Most AI Training roles pay $42–$98 per hour. This role's pay sits above that range.
Based on 1,073 similar roles that publish pay · 302 publish only a top rate; those count at the rate they gave
Rates shown per hour. Yearly and monthly pay converted; one-time fees and non-USD pay are not included.
- Live similar roles
- 1,214
- Listed in last 30 days
- 654
- Remote
- 98%
Hiring most right now: micro1 (360) · SME Careers (253) · Mercor (161)
Most requested skills · share of roles
- ai evaluation13%
- ai training13%
- llm evaluation12%
- data annotation10%
Figures from AI Training roles live on NearSkill when this page loaded. A role can close before you apply, so check the listing itself.
Compare your resume against these rolesLocation
Compensation
$180/hr
Required Skills
Application Tip
Name your specialty and board status in the first line, then list the concrete clinical activities you do (for example, reading echocardiograms, contouring for radiation oncology, or running POCUS in the ED). Mentioning EHR or clinical decision support experience and any publications gives reviewers extra signals to score.
How NearSkill helps you apply
1. Check your job match
We compare your resume with the job requirements to help you understand how well your experience fits.
2. Continue to the employer’s website
We’ll direct you to the employer’s application portal, where you can complete the application.
3. Get updates by email
If NearSkill is handling follow-up for this role, we’ll email you when there’s an update on your screening.
Want to check your fit first?
See how your resume matches this role before applying.
Check my job matchSimilar open positions
Explore active roles that match your skills and interests.
Handshake AI
VerifiedPhysician AI Medical Content Evaluator
Handshake AI needs licensed physicians to shape how future AI models handle complex medical questions. You will write challenging clinical questions, supply evidence-based reference answers, and judge model outputs for medical accuracy, reasoning, and safety. The role runs as a remote, asynchronous fellowship, so you set your own hours. Most contributors log 5 to 20 hours each week, with no minimum requirement. You join as a 1099 independent contractor and receive payment through Stripe for each completed task.
Mercor
VerifiedMedical Expert (M.D./D.O.) - Remote Part-Time
Mercor keeps this listing open for physicians who want to shape how frontier AI models handle clinical reasoning. The work fits around your practice: you might write a clinical vignette from a real case, complete with presentation, labs, and imaging, then supply the attending-level answer used as a reference. Other projects ask you to grade model output on diagnostic reasoning, checking whether a differential follows a sensible order, whether the model anchors on a first impression, and whether the proposed workup matches guidelines. You also flag recommendations that read well in a textbook but fail the patient in front of you because of comorbidities, contraindications, or access limits. Projects start on short notice, so Mercor draws from this pool when client demand matches your specialty.
Mercor
VerifiedGeneral Physician
Mercor pairs experienced physicians with a leading AI lab to check how well health AI models handle diagnosis, clinical reasoning, and care workflows. You will review model outputs, score how each answer aligns with accepted medical practice, and write feedback that helps researchers see where a model fails. The work is part-time at 20 or more hours each week and runs for six weeks, with an immediate start. Candidates need an MD or DO, an active unrestricted US license, completed residency, and a current patient-facing practice.
AfterQuery
VerifiedMedical MD Expert - Remote Contract Role
AfterQuery, a YC-backed startup, brings practicing physicians onto its team to write and grade the clinical scenarios behind its AI training work. You turn cases from your own experience into clinical case studies that test whether a model can reason through diagnosis and treatment, then score each answer against current medical guidelines. The arrangement runs remote and asynchronous, which leaves room for a residency, a fellowship, or a full patient panel. Physicians who write well and catch a weak differential get the most out of the work.
Handshake AI
VerifiedRadiologists
Handshake AI enlists practicing radiologists to shape how AI models handle real clinical reasoning. You bring your diagnostic imaging background in X-ray, CT, MRI, ultrasound, and nuclear medicine, then use it to judge what a model produces and where it falls short. Tasks mirror actual radiology work: image interpretation, differential diagnosis, protocol selection, and report writing. The program runs all year, with projects opening as different specialty areas come up, so you can take work alongside a current clinical post. Everything happens remotely and on a schedule you set.
Mercor
VerifiedCritical Care Physician
Mercor places board-certified critical care physicians on a project with a leading AI lab, where your clinical judgment becomes the check on machine-generated answers. You read de-identified patient charts, then score AI responses to critical care questions against the chart and against current standards of care. Every rationale you write on clinical rigor, completeness, and clarity feeds directly into how these health models get better. Hours are flexible and part-time, the work happens remotely, and the start date is immediate.