AI Agent Engineer — Compliance

Role overview Build AI agents that help healthcare practices review documentation and follow payer, regulatory, and operational requirements. You’ll develop systems that identify gaps or inconsistencies, explain findings with supporting evidence, and guide users toward resolution while preserving auditability and escalating ambiguous or high-risk decisions for human review.

Responsibilities - Build agents for documentation review, policy checks, workflow compliance, and quality assurance. - Translate complex requirements into testable workflows, tools, policies, and decision boundaries. - Create retrieval systems grounded in approved policies, contracts, payer rules, and internal guidance. - Present missing information, inconsistencies, exceptions, and next steps with citations and evidence trails. - Develop human-review processes, audit logs, and evaluations for grounding, policy adherence, consistency, and safe escalation. - Monitor deployed agents, investigate failures, manage changes to policies and evaluation materials, and apply privacy-by-design and least-privilege principles.

Requirements - Demonstrated experience building AI agents or AI systems that need to be accurate, explainable, and reliable. - Strong engineering judgment for high-stakes workflows, including clear limits on automated decisions. - Ability to turn nuanced requirements into practical system behavior in collaboration with domain experts. - Experience evaluating system quality and handling exceptions, human approval, or auditability. - No prescribed years of experience; relevant evidence may include professional, open-source, independent, or research work translated into working systems.

Nice to have - Experience in healthcare, insurance, finance, legal technology, or another regulated field. - Familiarity with healthcare privacy, payer requirements, documentation, or audits. - Experience with policy engines, case management, document intelligence, semantic retrieval, knowledge graphs, or provenance. - Knowledge of responsible AI, model risk, adversarial testing, access controls, data retention, or audit logging.