Data Scientist-FHIR, HL7, ICD/CPT, X12 EDI formats.-Woodland Hills, CA (Hybrid)-C2C- Ch

Role: Data Scientist

Location: Woodland Hills, CA (Hybrid)

Duration: 12+ Months

Must have:

• AI agent architectures, LLMs, NLP developing A2A Protocols and Model Context Protocols (MCP)
• LLMs and NLP models (e.g., medical BERT, BioGPT)
• Retrieval-augmented generation (RAG)
• Coding experience in Python, with proficiency in ML/NLP libraries
• Healthcare data standards like FHIR, HL7, ICD/CPT, X12 EDI formats.
• AWS, Azure, or Google Cloud Platform including Kubernetes, Docker, and CI/CD

We are seeking a Senior Data Scientist with deep expertise in LLMs, NLP, and agent architectures to lead the development of interoperable, self-improving AI agents in the healthcare domain. This role focuses on designing advanced multi-agent systems that interact intelligently across clinical, administrative, and benefits platforms using Agent-to-Agent (A2A) protocols and Model Context Protocols (MCP).

Responsibilities:

• Design and implement A2A protocols for autonomous task delegation and collaboration among specialized AI agents (e.g., ClaimsAgent, ProviderMatchAgent).
• Develop MCP pipelines to enable persistent memory and context continuity across multi-turn healthcare interactions.
• Architect and deploy LLM-orchestrated agent systems for use cases like prior authorizations, benefit optimization, and clinical summarization.
• Fine-tune domain-specific LLMs and NLP models (e.g., medical BERT, BioGPT) for intent classification and personalized recommendations.
• Build retrieval-augmented generation (RAG) systems with structured/unstructured healthcare data (e.g., FHIR, ICD-10, EHR).
• Collaborate on building scalable, secure, and compliant ML pipelines (HIPAA, CMS, NCQA).
• Lead research in memory-based agents, RLHF, and context-aware planning.
• Contribute to end-to-end MLOps pipelines for deployment, monitoring, and iteration.

Required Qualifications:

• Master s/Ph.D. in CS, ML, NLP, or related field.
• 7+ years in applied AI, particularly with LLMs, transformers, or agent systems in healthcare.
• Proficiency with tools like LangGraph, AutoGen, CrewAI, and hands-on A2A protocol development.
• Proven experience with Model Context Protocols, LLM pipelines, and healthcare NLP.
• Strong Python skills with libraries such as Hugging Face, LangChain, spaCy, and PyTorch.
• Understanding of healthcare systems (e.g., claims, eligibility, plan design).
• Experience with healthcare data standards: FHIR, ICD/CPT, HL7, X12 EDI.
• Cloud-native development: AWS/Google Cloud Platform/Azure, Docker/Kubernetes, CI/CD.

Preferred Qualifications:

• Expertise in MCP + VectorDB for agent memory and dynamic context retrieval.
• Experience building production-grade LLM agents in healthcare.
• Background in voice AI, AI navigation, or triage systems.
• Published work or patents in LLM-based agent systems or contextual AI.