The posting
About Patronus AI
Patronus AI is a frontier lab developing simulation research and infrastructure to accelerate progress toward human-aligned AGI. We are on a mission to simulate all of the world’s intelligence.
We are the team behind some of the earliest and most influential research in AI evaluation like FinanceBench, Lynx, SimpleSafetyTests, CopyrightCatcher, Humanity’s Last Exam, and more. We are formerly AI researchers and engineers from companies like Meta AI, Amazon AGI, and Google. Our customers include foundation model labs and Fortune 500 enterprises like Adobe. We are backed by top-tier investors like Lightspeed Venture Partners, Notable Capital, Stanford University, Noam Brown, Gokul Rajaram, and more.
At a glance
- Engagement: Independent contractor, fully remote, asynchronous
- Pay: $50 to $150 per hour. Some projects pay per task instead: you are paid when the full task is submitted, with a bonus for each accepted task. Strong performers move up within that range or onto productivity-based pay. The trial task is paid.
- Commitment: 10 or more hours per week
- Start: Rolling
- Location: Remote, US only
The role
You'll work with our research team on two things. The first is helping us understand how work in your area is actually done. The second is designing realistic healthcare tasks that a qualified professional could solve with the right knowledge and tools, and that today's AI models get wrong.
Each task starts from a scenario we provide. You create the records someone in your role would work from, write the request, solve it yourself and define how the answer is graded. Your work becomes evaluation and training data for frontier models.
You don't need an AI background. You do need to be the kind of clinician or healthcare professional who spots the detail in a chart that changes the plan.
Areas we're hiring for
Each area is defined by your license or certification, so you should fit exactly one. If you work across two, apply under the one where you spend most of your time.
- Clinical medicine. Physicians, nurse practitioners and physician assistants who assess, diagnose, order and plan treatment. They work from history, exam, labs and imaging through differential diagnosis, orders and follow-up. Peer-to-peer and medical director reviews belong here. Typical titles: Physician (MD or DO), Hospitalist, Primary Care Physician, Nurse Practitioner, Physician Assistant, Medical Director.
- Nursing and care coordination. Registered nurses, including utilization review and case management nurses, who assess patients, administer medications, carry out orders and plan care. They decide level of care against payer criteria and coordinate discharges and transitions of care. Typical titles: Registered Nurse, Charge Nurse, Utilization Review Nurse, RN Case Manager, Discharge Planner, Clinical Nurse Specialist.
- Pharmacy and medication management. Clinical and community pharmacists who verify orders, reconcile medications and check interactions and dosing. They manage formularies, therapeutic interchange and drug prior authorizations. Typical titles: Clinical Pharmacist, Hospital Pharmacist, Community Pharmacist, Pharmacy Manager, Formulary or PBM Pharmacist.
- Medical coding, billing and payer operations. Coders, billers, denials specialists and non-clinical authorization staff who assign ICD-10 and CPT codes, submit and defend claims, obtain authorizations and work denials and appeals. Clinical review of those cases belongs to the licensed areas above. Typical titles: Medical Coder (CPC or CCS), Coding Auditor, Medical Biller, Denials Specialist, AR Follow-up Specialist, Prior Authorization Specialist, Claims Examiner.
What you'll do
- Build realistic, fully fictional patient and operational records from a provided scenario, such as clinical notes, lab and imaging reports, medication lists and orders, nursing assessments and care plans, coding worksheets, prior authorization requests, claims and denial letters.
- Write task requests that state the deliverable and say how uncertainty and justified assumptions should be handled.
- Solve each task end to end and document a reference solution with the clinical or operational reasoning behind every step.
- Write scoring criteria precise enough that two independent graders reach the same verdict, including acceptable alternative answers and anything that would make a response unsafe.
- Check your own and your peers' tasks for clinical accuracy and ambiguity.
- Work in reviewer feedback within agreed turnaround times, and flag guideline gaps or edge cases you run into.
- Explain to our research team how work in your area is actually done: the steps, the documents, who does what, and where things go wrong.
- Review the scenarios and documents our team builds from that research, and flag anything a practitioner would find unrealistic.
Who we're looking for
Required
- 3+ years of hands-on experience in at least one of the areas above. That could be as a physician (MD or DO), nurse practitioner, physician assistant, registered nurse or pharmacist, or in medical coding, billing, denials or authorization work. You did the work yourself; supervising it is not enough.
- Current or past licensure or certification in your field, for example MD or DO, NP, PA, RN, PharmD, CCM, CPC, CCS or CRCR.
- Working knowledge of the guidelines, criteria and documentation standards in your area. Examples: clinical practice guidelines, nursing standards of care, formulary and dosing references, ICD-10-CM/PCS and CPT coding rules, or payer medical policies and review criteria.
- Careful reading and the habit of showing the evidence behind every conclusion.
- Clear, precise writing in professional English.
- Ability to work independently and asynchronously against deadlines.
Nice to have
- Prior experience writing or reviewing AI evaluation tasks, exam questions, case studies or training material.
- Familiarity with how AI models fail.
Our standards
- All tasks, documents and solutions are human-authored. LLM-generated content is not permitted.
- Tasks and solutions must be original. Nothing may be copied from published or easily searchable sources.
- Never use real patient data or PHI, even de-identified. All patients and scenarios are fictional.
- Quality is judged on accuracy, rubric validity, difficulty and reviewer agreement, not volume.
How hiring works
- Short AI interview and resume review
- One paid trial task, scored against a fixed quality checklist
- Onboarding and your first project
Patronus AI is an equal opportunity employer. We celebrate diversity in our workplace, and all qualified applicants will receive consideration for employment without regard to age, ancestry, color, family or medical care leave, gender identity or expression, genetic information, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran status, race, religion, sex (including pregnancy), sexual orientation, or other legally protected characteristics.
By clicking ‘Apply’, you agree to both of the following:
1) Greenhouse's Terms of Service and Privacy Policy
2) Patronus AI, Inc. Privacy Policy.
