Responsibilities
- Map complex health plan constraints (NCQA, HEDIS, Risk Adjustment) and verticalized business workflows (PBM formulary rules, pharmacy dispensing logic, specialty drug protocols) into logical, safe, conversational AI agent scripts.
- Work from existing agent inventory to prioritize and sequence the build based on business impact and reuse potential across clients.
- Define the clinical guardrails, edge cases, and escalation paths for each agent—accounting for the distinct regulatory and operational nuances across payor, PBM, and retail pharmacy contexts.
- Write and refine dialogue flows, including intent recognition, empathetic response patterns, and multi-turn conversation logic.
- Leverage LLMs (e.g., Claude) as an accelerant—using AI to draft and iterate on agent scripts while applying domain expertise to QA and tune the output.
- Work closely with Engineering on prompt design, tool calling, and integration with standard healthcare data formats.
- Rigorously test each agent through internal QA before handoff—running simulated conversations, stress-testing edge cases, and validating clinical safety and compliance across scenarios.
- Define success criteria and acceptance standards for each agent (e.g., conversation completion rates, correct escalation behavior, clinical accuracy) so the internal team can measure post-deployment performance.
- Iterate and refine agents based on internal testing feedback, peer review, and engineering integration testing prior to production deployment.
- Document agent logic, guardrails, and known limitations so the internal team has what they need to monitor, maintain, and improve agents after the contract ends.
- Inform the prioritization of the agent catalog by identifying which use cases have the highest reuse potential across payors, PBMs, and retail pharmacy clients.
Team
Structure: You will sit at the intersection of clinical operations, payor strategy, and conversational AI. Work closely with Engineering on prompt design, tool calling, and integration with standard healthcare data formats.
Additional Information
- Contract duration: 2–3 months
- Recruitment scams warning: All recruiting communication will come from @hippocraticai.com email addresses. The company will never request payment or sensitive personal information during the hiring process.
