Senior Principal AI/ML Product Engineer- Full Stack

UnitedHealth Group•Boston, MA

20h•$184,600 - $316,500•Remote

About The Position

The Senior Principal AI/ML Product Engineer is a hands-on senior individual contributor in Optum Health’s AI & Product Office. The role owns difficult technical decisions and builds, deploys, and operates AI-enabled products supporting the Chief AI & Product Officer’s enterprise AI-enabled product strategy, Agenix solutions portfolio, and product capability agenda. The engineer writes production code and builds end-to-end systems, including services, integrations with electronic health records and care systems, AI agents, and evaluations that demonstrate whether the products work. The role sets engineering standards through direct technical work rather than team management and is accountable for the reliability of products used by patients and care teams, as well as making systems operable by other teams. The engineer works with product, design, clinical, operations, and Optum Health engineering partners. The small, cross-functional team embeds with local Optum Health market teams to identify problems, design and build products, deploy them, and measure their impact. Learnings from initial markets are documented and developed into a repeatable method for subsequent markets. The engineer helps shape the team and its engineering practices, with AI coding and agent tools used as part of software development. The position may be performed remotely from anywhere in the United States. Employees hired in the Minneapolis or Washington, D.C. area must work in the office at least four days per week. Remote employees must follow UnitedHealth Group’s Telecommuter Policy. The salary range is $184,600–$316,500 annually for full-time employment, with pay dependent on factors such as location, education, experience, and certifications. Benefits include a comprehensive benefits package, incentive and recognition programs, an equity stock purchase plan, and a 401(k) contribution, subject to eligibility requirements.

Requirements

At least 10 years of professional software engineering experience shipping and operating production systems.
Experience in healthcare and/or within a large, complex, matrixed organization.
Full-stack experience with backend services, APIs, and data stores, plus sufficient front-end or mobile experience to ship a complete product without waiting on another team.
Experience building and shipping a large language model application or agent to real users, and ability to explain how it was evaluated.
Ability to use AI coding and agent tools in daily work, with experience building and running an agent, automation, or large language model tool at work or as a personal project.
Demonstrated testing and operational discipline, including automated tests, continuous integration, observability, and on-call ownership of shipped systems.
Experience owning cloud infrastructure and deployment for a production system, including infrastructure as code, continuous integration and deployment, and security controls that pass enterprise review.
Ability to work at both enterprise and individual-market levels and determine which level is appropriate for a decision.
Clear writing skills, including documenting design decisions and findings where future engineers can find them.
Working proficiency with GitHub, AI coding and agent tools, cloud platforms, and infrastructure as code.
Experience as a hands-on senior individual contributor who writes production code regularly, owns difficult problems end to end, and works across functions to deliver them.
Ability to serve as the technical owner of a system other engineers build on, including its architecture and difficult-to-reverse decisions.

Nice To Haves

Bachelor’s degree in Computer Science, Engineering, or a related field, or equivalent experience.
Experience with healthcare data and integrations, including FHIR, HL7, SMART on FHIR, or APIs for Epic, athenahealth, or eClinicalWorks.
Experience building on HIPAA-regulated systems that handle protected health information.
Experience with large language model evaluation frameworks, including building evaluation sets with domain experts.
Production experience with voice or SMS agents.
Experience taking an AI capability through an enterprise governance or review process in healthcare.
Experience on a small, early-stage product team where engineers owned the whole stack.

Responsibilities

Design, build, test, and deploy production software end to end, including backend services, data pipelines, and interfaces used by care teams and patients.
Own deployment practices, including cloud infrastructure as code, continuous integration and deployment, environments, and release practices suitable for software involved in patient care.
Build integrations that read from and write to electronic health records and care systems, including Epic, athenahealth, eClinicalWorks, and home care platforms, using FHIR, HL7, and vendor APIs in partnership with system-owning teams.
Build AI agents and large language model features across channels such as SMS and voice, including prompt and tool design, retrieval, and guardrails that keep agents within approved limits.
Prototype against real systems quickly and take successful prototypes into production.
Build shared foundations with Optum Health engineering teams, including integrations, deployment pipelines, and evaluation harnesses.
Build product evaluations from the outset, including automated assessments of model and agent behavior, tests that detect behavior regressions, and post-launch drift monitoring.
Operate shipped products through instrumentation, alerting, incident response, and continuous improvement.
Require evidence that changes work before considering them complete, and report unsuccessful results promptly.
Partner with technology, data, security, legal, privacy, compliance, care teams, and other stakeholders to deliver responsible, secure, enterprise-ready products; protect patient data through data minimization, access control, and audit.
Work with Optum Health engineering teams and provide documentation, tests, and runbooks that enable other teams to operate the work independently.
Set and improve engineering practices, including code review, AI-assisted development, and standards for evaluating AI systems.
Determine which technical decisions should be made for an individual market and which belong to the enterprise platform, and build accordingly.
Document weekly team learnings for review and conversion into decisions.
Help market teams build AI fluency by contributing shared templates, evaluation standards, and supported approaches to shipping AI safely.
Own the team's technical architecture and decisions that are costly to reverse, documenting each decision and what could change it.