Senior Software Engineer - Applied AI Remote

At Blue Cross and Blue Shield of Nebraska, we are a mission-driven organization dedicated to championing the health and well-being of our members and the communities we serve.

Our team is the power behind that promise. And, as the industry rapidly evolves and we seek ways to optimize business processes and customer experiences, there’s no greater time for forward-thinking professionals like you to join us in delivering on it! As a member of Team Blue, you'll find purpose, opportunities and the support you need to build a meaningful career and make a powerful impact in our community.

Health insurance is one of the most confusing, high-stakes experiences people navigate. Members don’t understand their coverage, can’t predict costs, and struggle to get answers when it matters most. We’re building the AI platform that changes that – a member experience that feels like having an expert in your corner. It’s already live. It’s already helping people. And it’s just getting started.

The next chapter is multi-agent: specialized AI agents collaborating to answer clinical questions, navigate prior authorizations, close care gaps, and guide members through their benefits. The orchestration architecture, the agent integration patterns, the evaluation systems that keep it all trustworthy at scale – that’s what we’re building together. You’ll shape how it’s designed and set the standard for how it grows.

The ideal candidate will live within driving distance of the Omaha, Nebraska office. This position allows remote flexibility but will have 1 day per week in the office.

If living in one of our approved states (Florida, Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and Texas) – this person may travel to our headquarters.

Your primary focus is agent design and orchestration.

• Multi-agent orchestration layer. Design and build the Lang Graph-based system that classifies member intent and routes to the right agent. Stateful, context-preserving, failure-tolerant. Handles commercial agents and internally built capabilities through the same interface.

• Agent integration contract. The standardized pattern that makes any agent pluggable: input schema, output schema, confidence signaling, latency SLA, fallback behavior. When the team integrates a new agent, they follow a pattern you designed.

• Context management across turns. Multi-turn conversations that maintain full member context across agent handoffs. The member asks a follow-up two turns later; the platform knows what was said, which agent handled it, and what was retrieved.

• Guardrail and safety layer. Clinical safety boundaries, PHI handling at the LLM layer, and response normalization ensuring every agent response arrives in a consistent tone and within compliance constraints regardless of which agent generated it.

• RAG pipeline architecture. Benefits documents, formularies, and policy content retrieval via Azure AI Search. You define the ingestion strategy, chunking approach, hybrid search configuration, and retrieval quality standards.

• Prompt architecture. Versioned prompt registry, structured output contracts, few-shot pattern design. Every prompt is testable, rollback-able, and understandable by the team.

• Bachelor’s degree and 4-7 years of experience in software engineering.

Experience with agentic AI in production – not LLM wrappers, not single-agent chatbot demos, not RAG tutorials. Systems with multiple agents, explicit state management, and real users depending on the answers being correct.

• 2â3 years building and operating production LLM or agentic AI systems
• You can describe specific decisions you made in a multi-agent system: state management approach, how you handled agent failures mid-conversation, how you kept the system governable as agents were added
• Lang Graph, Auto Gen, or custom graph-based orchestration used in production, not just evaluated. You know where it breaks down and how you worked around it
• Debugged retrieval quality problems, iterated on chunking strategies, dealt with stale knowledge bases in a live system with real users
• Built evaluation systems from scratch: golden…