About OpenTrain
OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help freelancers discover specialized AI training roles, consolidate proof-of-work into a single portfolio, and grow durable freelance careers in the human side of AI.
About AI training and this work
AI training (data labeling, annotation, and human feedback) is how modern models learn from real-world examples and human judgment. Contributors review model outputs, rate accuracy, and provide structured feedback so models improve over time.
This role focuses on evaluating AI outputs used in healthcare payment variance, underpayment detection, and managed-care contract compliance — high-impact work that helps both models and payer/provider operations become more reliable.
The role
OpenTrain is recruiting a Healthcare Payment Variance AI Specialist to evaluate AI-generated underpayment alerts, contract compliance outputs, and payment variance analyses. You will validate AI findings using payer contract interpretation, contract modeling, and payment reconciliation, and produce structured feedback to improve training datasets.
• Employment type: Contractor, part-time.
• Time requirement: 20+ hours/week.
• Location: United States only.
• Pay: $85.00 USD per hour.
• Labeling work: evaluation/rating and RLHF on text outputs; software listed as OTHER.
What you'll do day to day
You will combine domain expertise with careful review of AI outputs to confirm or correct automated findings, document issues, and support recovery actions.
• Review AI-generated underpayment detection alerts and payment variance analyses for accuracy and completeness.
• Interpret payer contract terms, fee schedules, carve-outs, OpenTrain provisions, and payment methodologies (commercial, Medicare Advantage, Medicaid managed care).
• Perform contract modelling and Excel-based payment reconciliation to identify systematic underpayments.
• Prepare underpayment claims and recovery correspondence to payers and track responses.
• Track recovery KPIs including underpayment dollars, recovery rates, and payer response rates.
• Provide structured, annotated feedback on AI outputs to improve model training datasets and downstream accuracy.
Requirements
Candidates must demonstrate hands-on experience with managed care contracting, underpayment recovery, and payment variance analysis. Clear written and verbal English is required.
• 5+ years experience in underpayment recovery, payment variance analysis, or managed care contracting.
• Strong contract interpretation skills across commercial, Medicare Advantage, and Medicaid managed care contracts.
• Familiarity with DRG-based, per diem, percent-of-billed charges, and fee schedule reimbursement methodologies.
• Proven Excel-based financial analysis skills and experience with revenue cycle analytics platforms.
• Comfortable reviewing and providing feedback on AI-generated compliance and detection outputs.
Helpful background (not required)
These credentials and experiences will help you hit the ground running and strengthen your feedback to AI training teams.
• CHFP, CRCR, or managed care contracting certification.
• Experience with contract management or recovery software such as Experian Health or Recondo.
• Background in hospital, health system, or large physician group underpayment recovery operations.
• Experience presenting underpayment analysis to executive leadership.
How the work is structured
This is a contract, part-time assignment managed through OpenTrain. You will review AI text outputs, apply domain knowledge to validate findings, and submit structured annotations and ratings that improve future model behavior.
• Work consists of evaluation/rating and RLHF-style feedback on AI text outputs.
• You will log findings, prepare recovery correspondence as needed, and update KPI tracking.
• Apply through OpenTrain and include relevant underpayment recovery examples on your profile to speed selection.
Who should apply
Experienced managed-care contracting and underpayment recovery professionals who want flexible, part-time AI training work and can commit 20+ hours/week. Ideal for practitioners with strong contract interpretation skills, Excel experience, and comfort evaluating AI outputs.