$120–150/hr · Mercor · Hourly
You evaluate whether AI systems generate clinically sound assessments, treatment plans, and patient education aligned with evidence-based practice and current guidelines.
What you would do
- Critique AI-generated clinical assessments and treatment plans for diagnostic accuracy and reasoning
- Compare AI responses side-by-side, evaluating which better reflects evidence and clinical appropriateness
- Review patient education materials for clarity, accuracy, and suitability to patient literacy levels
- Assess scenario-based responses involving diagnosis, medication decisions, and acute or chronic illness management
- Provide written explanations of clinical errors in AI outputs for training data
Who they want
- 4+ years of clinical practice as NP, APRN, FNP, ACNP, PNP, or related advanced practice role
- Active unrestricted RN license with specialized graduate education
- Demonstrated experience diagnosing and managing acute illness, episodic conditions, and chronic disease
- Strong written English communication to articulate clinical reasoning gaps clearly
- Commitment to 20+ hours weekly work during the project period
Main skills
What the interview asks about
1.Spotting diagnostic reasoning errors
AI models can produce plausible-sounding clinical narratives that obscure logical flaws; you must recognize when a conclusion is reached despite faulty reasoning.
For example: “An AI recommended amoxicillin for a 3-year-old with presumed otitis media without documenting otoscopy findings or ruling out URI symptoms. What makes this problematic?”
2.Evaluating guideline and protocol adherence
Clinical care requires alignment with current evidence and professional standards; a model output may be technically plausible but contradict your specialty's current best practices.
For example: “A model suggested starting a metformin + statin regimen for newly diagnosed type 2 diabetes in a patient with eGFR 35. Review your discipline's current recommendations: is this appropriate, and what should change?”
3.Assessing medication safety and selection logic
Pharmacology involves interdependent factors-renal and hepatic function, drug interactions, contraindications. Missing one component can create patient harm.
For example: “A model recommended sertraline 50 mg daily for an older adult patient taking fluoxetine. What's the safety issue, and how would you grade this output?”
4.Recognizing scope limitations and referral triggers
Your specialty has defined boundaries; discerning practitioners know when a case exceeds their scope and requires specialist involvement or emergent referral.
For example: “An AI generated a 2-week outpatient management plan for a patient with acute chest pain, ST-segment changes on EKG, and troponin elevation. What's missing from this advice, and how should the response have been framed?”
A task you may get
Write a brief scenario you've encountered in practice: patient presentation, your clinical reasoning, the decision you made, and why an alternative approach would have been unsafe or clinically inferior.
How to prepare
- Compile 2-3 cases from your practice where guidelines clearly dictated treatment but complexity made correct management subtle
- Review your specialty's most recent guideline updates on common conditions you manage
- Prepare examples of medication errors-either narrowly avoided or learned from-where drug interactions or renal/hepatic factors were critical
- Recall situations where you recognized a case exceeded your scope and required referral
The facts
- Pay
- $120–150/hr
- Hours
- Hourly
- Where
- Remote
- Open to
- USA
- Field
- Medicine
- Posted
- 8/21/2026
- Places left
- 15
We wrote this page from the public Mercor listing. It may be out of date, so read the full posting before you apply.