Senior AI/ML Engineer

Optum is a global organization that delivers care, aided by technology to help millions of people live healthier lives. The work you do with our team will directly improve health outcomes by connecting people with the care, pharmacy benefits, data and resources they need to feel their best. Here, you will find a culture guided by inclusion, talented peers, comprehensive benefits and career development opportunities. Come make an impact on the communities we serve as you help us advance health optimization on a global scale. Join us to start Caring. Connecting. Growing together.

Primary Responsibilities

• Build and deploy production LLM agents that automate detection, triage, and remediation of platform and application issues
• Design and tune RAG pipelines over runbooks, incident history, logs, code, and internal documentation, including chunking strategy, embedding selection, vector store design, hybrid and semantic search, and reranking
• Implement agent orchestration: tool/function calling, multi-step planning, multi-agent handoffs, memory, and state management
• Integrate agents with observability and security platforms (alerting, logging, metrics, vulnerability and incident data) through APIs and MCP-style tool interfaces
• Build the supporting layer agents depend on: tool wrappers, guardrails, human-in-the-loop approval paths, and fallback behavior
• Own model and prompt engineering, including model selection, prompt and context optimization, output structuring, and cost/latency tuning; fine-tune or apply parameter-efficient techniques where warranted
• Build evaluation pipelines and define quality metrics for agent and RAG behavior (retrieval precision and recall, groundedness, hallucination and false-positive rate, escalation rate, MTTR impact) and iterate based on results
• Ship agent capabilities into EVA so developers consume them inside their existing workflow
• Partner with SRE, InfoSec, and platform teams to scope which classes of issues are safe to automate and which require human review
• Contribute reusable patterns, components, and documentation for agent development across the Optum AI DLC
• Comply with all applicable Company policies, procedures, and business directives, changes including those relating to work location, team assignments, work schedules, and flexible work arrangements

Required Qualifications

Bachelor's degree in Computer Science, Data Science, Machine Learning, or a related field, or equivalent experience

4+ years of software or machine learning engineering experience

2+ years hands-on building with LLMs, including prompt engineering, tool/function calling, and agent or multi-step orchestration

Demonstrated experience designing and shipping RAG systems in production, including embeddings, vector databases (for example Azure AI Search, Pinecone, Milvus, pgvector, or FAISS), and retrieval evaluation

Experience integrating third-party APIs and building backend services that run in production

Experience with cloud platforms (Azure, AWS, or GCP), containers, and CI/CD pipelines

Experience building LLM evaluation and testing workflows rather than relying on manual spot checks

Proficiency in Python, with working knowledge of agent frameworks (for example LangGraph, LangChain, LlamaIndex, Semantic Kernel, or CrewAI)

Proven ability to work across teams and translate operational pain points into technical requirements

At UnitedHealth Group, our mission is to help people live healthier lives and make the health system work better for everyone. We believe everyone-of every race, gender, sexuality, age, location and income-deserves the opportunity to live their healthiest life. Today, however, there are still far too many barriers to good health which are disproportionately experienced by people of color, historically marginalized groups and those with lower incomes. We are committed to mitigating our impact on the environment and enabling and delivering equitable care that addresses health disparities and improves health outcomes - an enterprise priority reflected in our mission.