Principal Technical Product Manager

UnitedHealth Group•Eden Prairie, MN

2d•$134,600 - $230,800•Remote

About The Position

Optum AI is UnitedHealth Group’s enterprise AI team, comprised of AI/ML scientists and engineers with deep expertise in AI/ML engineering for healthcare. We develop AI/ML solutions for high-impact opportunities across UnitedHealth Group businesses. As a product leader within the enterprise AI Platforms team in the Chief AI Office, you will own the product strategy and execution for a commercial AI platform. This platform includes a secure, PHI-ready model access layer, a capability gateway for callable intelligence, and an agent harness for agent execution. Your role involves defining the platform's commercial aspects, including packaging, pricing, commitments, and limitations. This is a technical product management role, working closely with an engineering organization to define the platform's commitments, roadmap, interface contracts, evaluation criteria, and service levels. You will collaborate with business and design partners, manage the business case through the capital funding process, align the commercial roadmap with the enterprise multi-platform strategy, and be accountable for measured outcomes such as adoption, reuse, time to value, and unit economics. Expect significant influence over platform direction and a direct impact from design partner conversations to roadmap changes.

Requirements

Bachelor's degree in a technical field or equivalent experience
8+ years of technical product management experience, including platform, cloud or SaaS products
5+ years of experience creating product roadmaps from conception to launch, driving product vision, defining go-to-market strategy, and leading design discussions
3+ years of experience leading teams with diverse backgrounds in a technical organization
Experience owning an API-first or platform product consumed by other engineering teams or external developers, including responsibility for interface contracts, versioning and deprecation
Experience defining or launching products subject to security, privacy or regulatory constraints
Proven working fluency with AI/ML and LLM-based systems - retrieval-augmented generation, tool and function calling, agent orchestration, and evaluation - sufficient to hold a technical design discussion and make trade-off decisions with engineering
Demonstrated ability to explain complex technical results and trade-offs clearly to non-technical stakeholders, senior leaders and customers

Nice To Haves

Experience on a 0-to-1 product or platform, or commercializing an internally built platform for external customers
Experience defining, developing and maintaining platform-level scaled product solutions across a diverse set of stakeholders, with competing priorities between common capabilities and specific partner needs
Familiarity with multi-tenant platform concerns: tenant isolation, quota and entitlement enforcement, and per-tenant cost and usage telemetry
Experience with responsible AI practice, including model risk management, model documentation, or alignment to frameworks such as the NIST AI Risk Management Framework or the EU AI Act
Experience managing partner or vendor relationships with hyperscalers or major data and AI platform providers
Experience building for or selling to payers, providers or health services organizations
Health care industry experience: provider, payer, medical device, pharmaceutical, or other health services
Familiarity with Model Context Protocol (MCP), tool-calling standards, model gateways, or agent frameworks such as LangGraph or equivalent
Proven solid communication, relationship and interpersonal skills, with the ability to influence leaders and multiple stakeholders with and without direct authority

Responsibilities

Own the product strategy, roadmap and measured outcomes for the commercial AI platform: secure, PHI-ready model access with favorable inference economics, the capability gateway that makes licensed intelligence callable from any approved agent or application, and the agent runtime those agents run on
Define the commercial shape of the platform — what ships as a product versus what is delivered as a service, SKU and tier definitions, entitlement models, metering, and pricing mechanics that hold margin as underlying inference costs move
Translate client demand, market signal and internal partner needs into a single prioritized backlog, and write requirements at a technical altitude engineers can build against: API and tool contracts, schemas, versioning and deprecation policy, conformance expectations, and SLOs
Treat developer experience as an explicit product concern, with owned measures for time to first successful call, documentation and reference implementation quality, sandbox and test data access, and support and escalation paths
Build and defend the business case: value story, addressable demand, TCO and margin model, and unit economics expressed per completed unit of work rather than per token — and carry that case through the capital funding process
Define the instrumentation and measures the product is managed by, including adoption, reuse depth, time to value, quality and latency SLOs, cost per task, and retention, and use them to advance or retire roadmap items
Bring build, buy, partner and adopt decisions for platform components to a recommendation jointly with engineering — owning the commercial case, roadmap impact and total cost of ownership while engineering owns the technical assessment — and communicate the trade-offs to senior leaders
Work with security, privacy, legal, compliance and regulatory partners so that PHI handling, tenant isolation, BAA and data processing terms, model risk documentation and residency commitments are part of the product definition
Coordinate go-to-market with sales, client engineering, forward-deployed engineering and operations: positioning, pricing approval, sales enablement, contracting patterns, client onboarding runbook, and the support and service-level model the platform can actually honor
Ensure platform commitments are delivered on time and on budget, identifying and resolving risks and issues, and making scope calls explicitly rather than letting them resolve by default
Partner with enterprise AI platform product leaders so the commercial offering remains an extension of the enterprise platform rather than a fork, and so shared capabilities are built once and consumed in both directions
Use influence rather than authority to build a collaborative culture across engineering, design, architecture and business stakeholders, and represent the platform credibly to senior leadership and directly to clients