Optum is a global health organization dedicated to improving lives through technology. Join a collaborative team focused on advancing health optimization worldwide. This role is instrumental in connecting individuals with the care, benefits, data, and resources essential for well-being. You'll find a supportive environment that champions inclusion, professional growth, and comprehensive benefits.

Develop and deploy production LLM agents for automated issue detection, triage, and remediation. Design and refine RAG pipelines using runbooks, incident history, logs, code, and internal documentation. Implement agent orchestration including tool calling, multi-step planning, and memory management. Integrate agents with observability and security platforms. Build supporting agent infrastructure like tool wrappers and human-in-the-loop approvals. Own model and prompt engineering, including optimization, cost, and latency tuning. Create evaluation pipelines and quality metrics for agent behavior, iterating based on results. Ship agent capabilities into the EVA platform for developer use. Collaborate with SRE, InfoSec, and platform teams to define automation scope. Contribute reusable patterns and documentation for agent development.

A Bachelor's degree in Computer Science, Data Science, Machine Learning, or a related field, or equivalent experience, is required. Candidates should possess 4+ years in software or machine learning engineering, with at least 2 years hands-on LLM experience including prompt engineering and agent orchestration. Proven experience designing and deploying RAG systems, utilizing vector databases, and performing retrieval evaluation is essential. Proficiency in Python and experience with agent frameworks like LangGraph or LangChain is necessary. Experience with cloud platforms, containers, CI/CD, and building LLM evaluation workflows is a must. The ability to collaborate across teams and translate operational needs into technical requirements is highly valued.