AI training jobs for doctors and medical experts
Frontier AI labs pay clinicians, coders and researchers to judge whether a model's medical answer is correct, safe, and defensible against the standards their own field is held to.
- Open roles
- 104
- Median rate
- $98/hr
- Publish an hourly rate
- 92%
listings, updated continuously
from 96 published bands
the sample every median here uses
What each medical specialty pays
Every listing we track, grouped by specialty. Rates are read from the platform's own posting — the median is of published hourly bands, and the range runs from the lowest published floor to the highest published ceiling.
| Specialty | Roles | Median | Published range | Distribution |
|---|---|---|---|---|
| Research, Trials & Biostatistics | 5 | $163n=3 | $60–230 | |
| Adolescent & Mental Health Safety | 9 | $125n=9 | $45–200 | |
| Clinical Practice & Nursing | 23 | $103n=22 | $55–450 | |
| General Medical & AI Evaluation | 25 | $94n=22 | $45–180 | |
| Pharma, Regulatory & Medical Writing | 5 | $80n=5 | $55–210 | |
| Medical Coding, Billing & Payer Operations | 7 | $60n=7 | $20–80 |
30 listings are not yet assigned to a specialty, and are counted in the total at the top of the page but not in this table.
Open medical roles
The listings behind the figures above, richest published ceiling first.
What medical experts do in AI
Review model output for clinical safety
Judge whether an AI's clinical answer is correct — and whether it would be safe in front of a patient.
Audit coding and payer decisions
Review AI-generated ICD-10, HCC and prior-authorisation outputs the way you would audit a coder's or a payer's work.
Author benchmark cases and rubrics
Write the expert-level questions and grading rubrics a model is measured against, each with a defensible correct answer.
Build regulatory and trial documents
Produce the study-report sections, safety narratives and synthetic trial data that AI agents are trained and tested on.
Join a clinical talent pool
Apply once, pass a screening interview, and be matched to lab projects as they open rather than to one fixed role.
Who this work is for
Backgrounds in demand
- Medical coders, billers & RCM leaders7roles
- Physicians & hospitalists23roles
- Pharmacovigilance, regulatory & medical writers5roles
- Clinical researchers & biostatisticians5roles
- Child & adolescent mental-health clinicians9roles
Commonly asked for
- Clinical reasoning
- Clinical risk assessment
- Severity staging
- ICD-10 & CPT coding
- HCC / risk adjustment
- Medical necessity review
- Payer criteria (InterQual, MCG)
- Clinical documentation
- Pharmacovigilance
- Regulatory writing
- Trial data interpretation
- Rubric-based evaluation
Drawn from the listings themselves and not counted — the skills field on a listing is free text, and no ranking of it would mean anything.
Platforms hiring medical experts
- MercorRoles25Median$120Published range$45–450
- EthosRoles6Median$100Published range$80–100
- micro1Roles36Median$98Published range$20–285
- AfterQueryRoles29Median$88Published range$45–160
Figures are calculated over the medical listings each platform currently has open, on the same basis as the rate report.
Also join these expert networks
- PROFILE ENTRYEthos
One profile with Ethos, matched to paid opportunities across fields as they open — not an application to a single listing.
Create profile on Ethos → - PROFILE ENTRYHandshake AI
One profile with Handshake AI, matched to paid opportunities across fields as they open — not an application to a single listing.
Create profile on Handshake AI →
How AI expert work actually works
- 5 min read
What Medical AI Reviewers Are Actually Asked to Judge
Separate factual support, clinical context and safety-sensitive omissions without treating a fluent answer as a clinical assessment.
- 5 min read
What Is AI Evaluation Work?
An output, a set of instructions and a judgment to make. Understand the work, who uses it and where human reviewers fit into AI development.
- 5 min read
How to Pass an AI Evaluation Interview or Assessment
Prepare to demonstrate your reasoning, follow the assessment rules and explain your experience without scripts or invented credentials.
- 6 min read
Mercor vs micro1
Mercor and micro1 side by side — open roles, advertised rates, fields and screening from live Sidequest data — then how to compare a specific opportunity on either.
Questions, answered plainly
Mostly judgement work: reviewing AI-generated clinical answers for correctness and safety, auditing coding and payer decisions, and writing the benchmark cases and rubrics a model is scored against. The clinical judgement is what is being bought — the AI side is the subject, not the skill.
The median and each specialty's median and published range are at the top of this page, every one with the number of listings behind it. Rates are read from the platform's own posting; where a platform publishes nothing, we show nothing.
Usually not. What listings ask for is verifiable depth in your own field — a licence, a board certification, a coding credential, years in a specialty. Where familiarity with AI tools appears in the eligibility criteria it is typically listed as preferred rather than required, and each role page separates the two.
Most listings name a specific credential — board certification, an RN or physician licence, a professional coding certification, or a regulatory or biostatistics background. Arrangements vary by listing: much of this is project-based contract work paid by the hour, some listings are restricted to a single country, and a few are full-time. Each role page states what that listing says; we publish no remote percentage, because many listings say nothing about location and silence is not the same as “anywhere”.
Other fields
New medical roles, by email
One message on the days medical listings open. Nothing on the days they do not.