What the work involves

You review outputs from an AI system built for self-pay revenue recovery and patient financial engagement — draft collection letters and text/email outreach, payment plan recommendations, financial assistance screening logic, and proposed resolution paths for balances at various stages of the cycle. For each item you decide whether the output is factually correct, compliant, and appropriate for the patient's situation, then write structured feedback explaining exactly where it fails and what a competent PFS leader would have done instead. The judgment calls are the point: a payment plan that is technically permissible but sets a patient up to default, a letter that is accurate but crosses into FDCPA territory, an outreach cadence that ignores charity care eligibility.

Typical tasks include side-by-side comparison of two model responses, rubric scoring on dimensions like compliance, accuracy, and patient-facing tone, and free-text rationales. Some projects ask for original prompts or gold-standard rewrites drawn from real early-out, bad debt, or point-of-service collection scenarios you have actually managed.

What the platform screens for

  • Verifiable operational depth — Mercor's AI-led interview probes specifics: what you collected pre-service versus post-service, how your early-out vendor was scored, what your bad debt write-off rate was and why it moved.
  • Compliance fluency under follow-up — FDCPA, HIPAA, state collections statutes, and 501(r) financial assistance obligations, applied to concrete fact patterns rather than recited.
  • Written reasoning — feedback that another reviewer can act on. Vague scores get calibration flags.
  • Consistency — your ratings are compared against other reviewers and against expert consensus sets.

Logistics

Fully remote, asynchronous, contractor engagement. Most reviewers commit 10–20 hours per week against a task queue with turnaround windows rather than fixed shifts; some projects offer higher volume during ramp periods and then taper. Pay is stated as observed at $92/hr for this listing — actual rates vary by project, assessed depth, and throughput, and are not guaranteed. Expect an onboarding calibration round before paid volume opens up.