Design and build advanced LLM-driven agentic systems for healthcare, focusing on clinical workflows, evaluation frameworks, and production-grade deployment. Collaborate with cross-functional teams to ensure reliability, scalability, and clinical trustworthiness of AI-generated outputs. Requires 3+ years of production-grade LLM experience, deep fluency in orchestration, and rigorous evaluation methodologies.
Key Highlights
Build agentic LLM systems for clinical documentation with retrieval, tool-use, and structured outputs
Develop and apply automated/human-in-the-loop evaluation frameworks for accuracy, robustness, and multilingual capabilities
Own end-to-end productionization: deployment, monitoring, observability, and guardrails for low-latency, high-uptime systems
Key Responsibilities
Design and implement agentic LLM workflows for clinical documentation, integrating retrieval systems, tool-use, and structured outputs
Collaborate with ML/infra engineers to scale agentic systems while managing latency, context windows, and model selection
Develop rigorous evaluation frameworks (automated + human-in-the-loop) to assess accuracy, robustness, and multilingual capabilities
Prototype and test frontier LLM capabilities, open-source tools, and novel prompting techniques for clinical use cases
Deploy and monitor LLM workflows in low-latency, high-uptime environments with observability and guardrails
Drive architectural decisions for data flow, caching, and generative output structuring
Conduct A/B tests and analyze clinician feedback to guide model improvements
Discover our full range of relocation jobs with comprehensive support packages to help you relocate and settle in your new location.
Benefits & Perks
Competitive compensation ($255K–$300K annually)
Generous time off (14 paid holidays, flexible PTO)
Comprehensive health plans (medical, dental, vision, HSA contributions)
Paid parental leave and family-forming benefits
401(k) matching, personal device allowance, lifestyle wallet (fitness/professional development)
Sabbatical leave after 5 years
Mental health support (therapy/coaching)
Hybrid work model (3x/week in-office, SF/NYC)
Nice to Have
Experience with async programming, performance profiling, and deployment tooling
Familiarity with vector databases (e.g., semantic/lexical retrieval, efficient kNN)
Knowledge of function calling and tool-use patterns in LLM systems
