What the work involves
You will spend most of your time reading AI outputs against real clinical scenarios and judging whether they hold up. That means assessing whether a model-drafted physician query is compliant or leading, whether a suggested diagnosis is supported by the documented clinical indicators, whether a proposed DRG shift correctly applies CC/MCC hierarchies, and whether an HCC capture recommendation would survive audit. Alongside pass/fail judgments, you write structured rationales — the reasoning is the deliverable, not the verdict. Some tasks ask you to rewrite a weak AI query into a compliant one, or to rank several model responses against each other and explain the ordering.
What the screening measures
Mercor's process is an AI-led interview plus resume verification. Expect the screen to probe past the summary line on your CV: which facility types, whether you ran concurrent or retrospective review, what your CC/MCC capture and query response rates actually were, and how you handled a physician who pushed back. Follow-up questions get specific fast — naming MS-DRG methodology is not the same as walking through a sepsis versus SIRS documentation conflict and its DRG consequence. Credential claims (CCDS, CDIP, RN, RHIA, CCS) are checked, and clarity of written English matters because your annotations become training signal.
Logistics
- Fully remote and asynchronous; you pick up task batches rather than holding shift coverage.
- Typical commitments run 10–20 hours per week, though volume fluctuates with project phase and some contributors are offered more.
- Work is contract-based and paid hourly; $84/hr is the observed rate for this listing and is not guaranteed across projects or seniority levels.
- Expect an onboarding calibration round where your judgments are compared against reviewer consensus before you scale up.