Sr Software Engineer

Blue Cross Blue Shield of Nebraska • Omaha, NE

4d • Hybrid

About The Position

Blue Cross and Blue Shield of Nebraska is hiring a Sr Software Engineer to help build and scale an AI platform that improves members’ healthcare experiences. The platform is live and is expanding toward a multi-agent architecture in which specialized agents collaborate to answer clinical questions, navigate prior authorizations, close care gaps, and explain member benefits. The engineer will shape the orchestration architecture, agent integration patterns, and evaluation systems that support trustworthy operation at scale.

The role focuses primarily on agent design and orchestration. The engineer will develop a LangGraph-based layer that classifies member intent and routes requests to appropriate agents, while preserving state and context and handling failures. They will establish standardized integration contracts for agents, manage context across conversational turns and agent handoffs, and build safety and guardrail mechanisms for clinical boundaries, PHI handling, response consistency, and compliance.

Additional work includes architecting retrieval-augmented generation for benefits documents, formularies, and policy content using Azure AI Search, and creating a versioned prompt registry with structured output contracts and testable, rollback-capable prompts. The target stack includes Python/FastAPI, LangGraph, Azure OpenAI or Claude via Azure, Azure AI Search, Azure Container Apps, LlamaIndex, LangSmith, Azure App Insights, PostgreSQL, and Redis.

The role allows remote flexibility but requires one day per week in the Omaha office. The ideal candidate lives within driving distance of Omaha. Candidates residing in approved states—Florida, Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and Texas—may travel to headquarters. The company is unable to sponsor or take over sponsorship of an employment visa.

Requirements

• Bachelor’s degree.
• 4–7 years of experience in software engineering.
• 2–3 years building and operating production LLM or agentic AI systems.
• Production experience with agentic AI systems involving multiple agents, explicit state management, and real users; experience limited to LLM wrappers, single-agent chatbot demos, or RAG tutorials is not sufficient.
• Ability to describe specific decisions made in a multi-agent system, including state management, handling agent failures mid-conversation, and keeping the system governable as agents are added.
• Production use of LangGraph, AutoGen, or custom graph-based orchestration, with understanding of failure modes and workarounds.
• Experience debugging retrieval quality, iterating on chunking strategies, and addressing stale knowledge bases in a live system.
• Experience building evaluation systems, including golden datasets, automated regression testing, and domain-appropriate accuracy metrics.
• Experience in healthcare, insurance, fintech, or legal domains, with safeguards designed for AI systems where incorrect answers have consequences.
• Strong knowledge of Python and the ability to work in an existing codebase.
• The position requires work eligibility that does not require the company to sponsor or take over sponsorship of an employment visa.
• The role requires one day per week in the Omaha office; the posting identifies the ideal candidate as living within driving distance of Omaha and lists approved states for potential headquarters travel.

Nice To Haves

• Experience working in an Agile/Scrum environment.
• A creative, lateral approach to conceptual and practical problem-solving.
• Strong focus on addressing customer needs.
• Strong communication and interpersonal skills.
• Emotional intelligence.

Responsibilities

• Design and build a LangGraph-based multi-agent orchestration layer that classifies member intent and routes requests to the appropriate agent.
• Create an orchestration system that preserves state and context, tolerates failures, and supports commercial agents and internally built capabilities through a common interface.
• Define standardized agent integration contracts, including input and output schemas, confidence signaling, latency service-level expectations, and fallback behavior.
• Manage context across multiple conversational turns and agent handoffs, retaining relevant member context, prior agent activity, and retrieved information.
• Build guardrails for clinical safety boundaries, PHI handling at the LLM layer, response normalization, consistent tone, and compliance constraints.
• Architect the RAG pipeline for benefits documents, formularies, and policy content using Azure AI Search, including ingestion, chunking, hybrid search, and retrieval quality standards.
• Develop a versioned prompt registry and structured output contracts, and design prompts to be testable, rollback-capable, and understandable by the team.
• Build and operate production LLM or agentic AI systems and evaluation systems, including golden datasets, automated regression testing, and domain-appropriate accuracy metrics.
• Work in an existing codebase and extend it with the AI layer.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior