Principal AI Architect

UnitedHealth Group•Eden Prairie, MN

7h•$134,600 - $230,800•Remote

About The Position

Optum is seeking a visionary Digital AI architect for our Contract Center Transformation. The successful candidate will play a critical role in transforming our voice and digital experiences to conversational experiences, through multiple engagement channels (including web, app, chat, voice, and search), with a particular emphasis on transforming how consumers engage in the US health care system. In this role you will help make UnitedHealth Group's consumer obsessed digital vision come to life by using AI to deliver our next generation Conversational Experiences. Define and lead the strategic vision, architecture, and implementation of AI contact solutions across our consumer-facing digital experiences. Partner closely with business and product teams to identify and prioritize strategic opportunities for leveraging AI within web, mobile, chat messaging, voice messaging, and search channels. Oversee the development of scalable, secure, and innovative AI solutions designed to enhance consumer engagement and deliver personalized, intuitive self-service experiences that transform our current digital experiences into conversational experiences. You’ll enjoy the flexibility to work remotely from anywhere within the U.S. as you take on some tough challenges. For all hires in the Minneapolis or Washington, D.C. area, you will be required to work in the office a minimum of four days per week.

Requirements

Bachelor’s degree in Computer Science, Engineering, or equivalent practical experience
5+ years of experience designing and delivering voice bot, conversational voice AI, IVR, or customer-facing voice automation solutions in production environments
3+ years of hands-on experience designing or implementing conversational voice systems, including one or more of the following: STT/TTS pipelines, V2V systems, multi-turn voice interactions, bot orchestration, session memory/context, or escalation to human agents
3+ years of experience with modern voice integration technologies or protocols such as SIP, WebRTC, WebSocket, telephony integration, or signaling/media handling
3+ years of experience designing conversational/self-service voice experiences, including intent handling, fallback/escalation design, and context transfer between automation and human-assisted channels
3+ years of experience with one or more of the following: voice memory/context handling, multi-bot orchestration, reasoning bots, voice guardrails, voice telemetry/evaluations, downstream voice integration patterns, or protocol tradeoffs
Experience evaluating tradeoffs between deterministic voice flows, reasoning-enabled bots, domain-specific bots, and human escalation paths
Experience with AI in engineering workflows, including standards, guardrails, reusable prompts/patterns, and evaluation/quality checks
Demonstrated ability to produce architecture/design artifacts, lead technical workshops, mentor engineers, and influence cross-functional stakeholders
Proven solid troubleshooting mindset with demonstrated ownership of production support and root cause analysis for customer-facing voice systems

Nice To Haves

Experience designing voice systems where the telephony/routing platform is distinct from the downstream bot, orchestration, or AI layers
Experience with reasoning-enabled voice agents, grounded tool use, and policy-controlled action execution
Experience designing V2V / speech-to-speech systems with low-latency streaming, natural turn-taking, and interruption/barge-in handling
Experience with speech analytics, transcript analysis, redaction/governance, or voice quality evaluation frameworks
Experience with data pipelines and telemetry patterns for voice quality, reporting, evaluation, and conversational analytics
Experience embedding or integrating live agent workflows with contextual handoff from voice automation systems
Experience with Infrastructure-as-Code and CI/CD for cloud platform delivery
Experience operating in regulated or high-compliance environments and designing for privacy-by-default, encryption, auditability, and retention controls
Proven ability to architect cross-channel experiences that unify voice, chat, and human-assisted workflows while minimizing customer effort

Responsibilities

Own the end-to-end technical architecture for conversational voice AI systems, including voice bot orchestration, memory/context handling, backend integrations, agent handoff, and quality feedback loops
Design and implement voice bot architectures for natural language self-service, including multi-turn conversations, fulfillment, fallback, escalation, and smooth transitions to human-assisted experiences
Define architecture patterns for STT/TTS, V2V (voice-to-voice), and hybrid voice interaction models, including tradeoffs in latency, controllability, grounding, observability, interruption handling, and user experience
Design and implement reasoning-enabled voice bot architectures, including policy-controlled tool use, grounding/validation, fallback behavior, and escalation patterns
Define and implement multi-bot / multi-agent orchestration patterns for voice, including dispatcher, orchestrator, domain-bot, and escalation models
Define approaches for session memory, context propagation, authentication context, and conversation continuity across bots, backend systems, and live agents
Establish voice guardrail patterns including policy enforcement, safe fallback behavior, hallucination/grounding controls, redaction/governance, escalation rules, and compliance-aware protections
Define telemetry, observability, and evaluation patterns for conversational voice systems, including latency, recognition quality, interruption/barge-in behavior, drop-offs, failure rates, containment, transcript quality, and task completion
Design integration patterns between voice bot systems and enterprise platforms such as identity, CRM, customer profile, case management, knowledge systems, and downstream APIs/services
Design voice architectures that cleanly separate conversation intelligence from telephony/platform concerns while still supporting reliable context transfer and operational continuity
Apply technologies such as SIP, WebRTC, WebSocket, APIs, and event-driven services where appropriate to support real-time voice interactions and downstream voice bot integrations
Lead design reviews, publish reusable patterns and standards, mentor engineers, and influence technical direction across voice initiatives
Partner closely with Product, Engineering, Security, Operations, and telephony/platform teams to translate business outcomes into scalable technical designs