Principal Solutions Architect for Conversational AI
Minnesota Jobs, Saint Paul, MN, United States
Principal Solutions Architect
Become a part of our caring community and help us put health first. The Principal Solutions Architect works with user groups to solve business problems with available technology including hardware, software, databases, and peripherals. The Principal Solutions Architect provides strategic advice and guidance to functional team(s). Highly skilled with broad, advanced technical experience. Role Summary
The Principal Architect for Conversational AI owns the end-to-end technical architecture for enterprise-grade voice and chat experiencesfrom NLU/LLM orchestration and dialog policy to integrations, observability, security, and production reliability. You will define standards, reference architectures, and platform patterns that enable multiple product teams to build consistent, safe, scalable conversational journeys across channels. This role blends hands-on systems design with strategic technical leadership: you'll shape platform direction, guide engineering teams, and ensure solutions meet performance, compliance, and customer experience goals. Key Responsibilities
Define target-state architecture for conversational AI across voice + chat + messaging. Establish reference patterns for: intent/routing, dialog management, LLM tools/function calling, RAG, and human handoff. Design hybrid approaches using NLU (intents/entities) where deterministic behavior is needed and LLMs where flexibility adds value. Build patterns for prompt engineering, tool use, safety rails, grounding, and response evaluation. Partner with Conversation Designers to translate journeys into scalable dialog systems. Formalize conversation state models, slot-filling strategies, fallbacks, barge-in, and recovery behavior. Architect integrations to CRM, scheduling, claims, EHR/EMR, identity, billing, knowledge bases, ticketing, and contact center platforms. Standardize APIs, event-driven flows, caching, and data contracts. Own architectural decisions for: ASR/TTS pipelines (voice) channel adapters session/state persistence policy engines personalization and context experimentation / A/B testing Define guardrails for PHI/PII handling, data retention, encryption, key management, audit logging. Ensure compliance alignment (HIPAA, SOC2, GDPRdepending on industry). Set SLOs, error budgets, DR patterns, and scalability targets. Implement conversation analytics standards: containment, task completion, deflection, CSAT proxies, latency, hallucination rates, escalation reasons. Establish evaluation frameworks (offline + online), regression testing, and golden conversation sets. Serve as the principal technical advisor across teams. Lead architecture reviews, mentor senior engineers, influence roadmap and investment priorities. Identify build vs buy opportunities and vendor selection criteria. What You'll Build / Own (Deliverables)
Conversational AI reference architecture (voice + chat) Platform standards: tool calling, state, memory/context, grounding, safety policy, fallback patterns A reusable conversation runtime framework (or governance model if using vendor platforms) Integration blueprint for core enterprise services Observability & evaluation standard operating model (dashboards, QA pipelines, red-team tests) Technical Skills (Typical Stack)
Languages: Python, Java, TypeScript/Node, Go (any 12 strong) LLM layer: OpenAI / Azure OpenAI / Anthropic / Bedrock; tool/function calling; RAG patterns NLU/dialog: Dialogflow, Lex, Rasa, Nuance, custom NLU Voice: ASR/TTS services, SIP/telephony concepts, IVR/contact center integrations Infra: Kubernetes, serverless, Terraform, CI/CD, secrets management Data: Redis, Postgres, vector DBs (optional), message bus (Kafka), observability tooling Security: IAM, encryption, audit logging, data classification, tokenization/masking Use Your Skills to Make an Impact
Required Qualifications 10+ years software engineering experience with 3+ years in AI/ML systems or conversational platforms. Proven architecture ownership for distributed systems in production (high availability, scalable, observable). Deep knowledge of at least one: Contact center / conversational platforms (Genesys, NICE, Twilio, Amazon Connect, Dialogflow, Nuance, etc.) LLM orchestration frameworks/patterns (tool use, RAG, retrieval, safety layers, prompt/versioning) Strong experience with: API design (REST/gRPC), event-driven systems (Kafka/PubSub), and integration patterns Cloud architecture (AWS/Azure/GCP), IaC, CI/CD Data security practices for sensitive data (PII/PHI) and governance controls Ability to communicate architecture clearly to executives and engineers (diagrams, decision records, tradeoffs). Preferred Qualifications Experience deploying voice systems: ASR/TTS tuning, barge-in, turn-taking, latency optimization. Experience with evaluation & testing for LLM/agent systems: automated conversation testing, hallucination detection, rubric scoring, red teaming Experience in regulated environments (healthcare, finance, insurance, public sector). Familiarity with knowledge graphs, ontology-driven reasoning, or deterministic policy engines. Track record of leading multi-team platforms (enablement, adoption, governance). Additional Information
Success Metrics (How We'll Know You're Winning) Reduced time-to-launch for new conversational experiences via reusable patterns Higher containment and task completion with stable CSAT/NPS Lower latency and fewer failure loops / escalations Measurable improvement in reliability (SLO attainment), observability, and incident reduction Consistent governance of prompts/tools/models with safe rollout & rollback practices Leadership Competencies Architectural judgment: picks the simplest thing that will work (and scales) Pragmatic AI mindset: knows where LLMs shine and where deterministic logic must own the flow Influences without authority; builds alignment across product, design, security, and ops Mentors and raises the bar across engineering This role is not a "lead developer on one team." It is a cross-domain architecture owner responsible for platform standards and enterprise adoption. Work-At-Home Requirements
WAH requirements: Must have the ability to provide a high speed DSL or cable modem for a home office. Associates or contractors who live and work from home in the state of California will be provided payment for their internet expense. A minimum standard speed for optimal performance of 25x10 (25mpbs download x 10mpbs upload) is required. Satellite and Wireless Internet service is NOT allowed for this role. A dedicated space lacking ongoing interruptions to protect member PHI / HIPAA information Travel: While this is a remote position, occasional travel to Humana's offices for training or meetings may be required. Scheduled Weekly Hours 40 Pay Range $172,200 - $236,900 per year This job is eligible for a bonus incentive plan. This incentive opportunity is based upon company and/or individual performance. Description of Benefits Humana, Inc. and its affiliated subsidiaries (collectively, "Humana")
Become a part of our caring community and help us put health first. The Principal Solutions Architect works with user groups to solve business problems with available technology including hardware, software, databases, and peripherals. The Principal Solutions Architect provides strategic advice and guidance to functional team(s). Highly skilled with broad, advanced technical experience. Role Summary
The Principal Architect for Conversational AI owns the end-to-end technical architecture for enterprise-grade voice and chat experiencesfrom NLU/LLM orchestration and dialog policy to integrations, observability, security, and production reliability. You will define standards, reference architectures, and platform patterns that enable multiple product teams to build consistent, safe, scalable conversational journeys across channels. This role blends hands-on systems design with strategic technical leadership: you'll shape platform direction, guide engineering teams, and ensure solutions meet performance, compliance, and customer experience goals. Key Responsibilities
Define target-state architecture for conversational AI across voice + chat + messaging. Establish reference patterns for: intent/routing, dialog management, LLM tools/function calling, RAG, and human handoff. Design hybrid approaches using NLU (intents/entities) where deterministic behavior is needed and LLMs where flexibility adds value. Build patterns for prompt engineering, tool use, safety rails, grounding, and response evaluation. Partner with Conversation Designers to translate journeys into scalable dialog systems. Formalize conversation state models, slot-filling strategies, fallbacks, barge-in, and recovery behavior. Architect integrations to CRM, scheduling, claims, EHR/EMR, identity, billing, knowledge bases, ticketing, and contact center platforms. Standardize APIs, event-driven flows, caching, and data contracts. Own architectural decisions for: ASR/TTS pipelines (voice) channel adapters session/state persistence policy engines personalization and context experimentation / A/B testing Define guardrails for PHI/PII handling, data retention, encryption, key management, audit logging. Ensure compliance alignment (HIPAA, SOC2, GDPRdepending on industry). Set SLOs, error budgets, DR patterns, and scalability targets. Implement conversation analytics standards: containment, task completion, deflection, CSAT proxies, latency, hallucination rates, escalation reasons. Establish evaluation frameworks (offline + online), regression testing, and golden conversation sets. Serve as the principal technical advisor across teams. Lead architecture reviews, mentor senior engineers, influence roadmap and investment priorities. Identify build vs buy opportunities and vendor selection criteria. What You'll Build / Own (Deliverables)
Conversational AI reference architecture (voice + chat) Platform standards: tool calling, state, memory/context, grounding, safety policy, fallback patterns A reusable conversation runtime framework (or governance model if using vendor platforms) Integration blueprint for core enterprise services Observability & evaluation standard operating model (dashboards, QA pipelines, red-team tests) Technical Skills (Typical Stack)
Languages: Python, Java, TypeScript/Node, Go (any 12 strong) LLM layer: OpenAI / Azure OpenAI / Anthropic / Bedrock; tool/function calling; RAG patterns NLU/dialog: Dialogflow, Lex, Rasa, Nuance, custom NLU Voice: ASR/TTS services, SIP/telephony concepts, IVR/contact center integrations Infra: Kubernetes, serverless, Terraform, CI/CD, secrets management Data: Redis, Postgres, vector DBs (optional), message bus (Kafka), observability tooling Security: IAM, encryption, audit logging, data classification, tokenization/masking Use Your Skills to Make an Impact
Required Qualifications 10+ years software engineering experience with 3+ years in AI/ML systems or conversational platforms. Proven architecture ownership for distributed systems in production (high availability, scalable, observable). Deep knowledge of at least one: Contact center / conversational platforms (Genesys, NICE, Twilio, Amazon Connect, Dialogflow, Nuance, etc.) LLM orchestration frameworks/patterns (tool use, RAG, retrieval, safety layers, prompt/versioning) Strong experience with: API design (REST/gRPC), event-driven systems (Kafka/PubSub), and integration patterns Cloud architecture (AWS/Azure/GCP), IaC, CI/CD Data security practices for sensitive data (PII/PHI) and governance controls Ability to communicate architecture clearly to executives and engineers (diagrams, decision records, tradeoffs). Preferred Qualifications Experience deploying voice systems: ASR/TTS tuning, barge-in, turn-taking, latency optimization. Experience with evaluation & testing for LLM/agent systems: automated conversation testing, hallucination detection, rubric scoring, red teaming Experience in regulated environments (healthcare, finance, insurance, public sector). Familiarity with knowledge graphs, ontology-driven reasoning, or deterministic policy engines. Track record of leading multi-team platforms (enablement, adoption, governance). Additional Information
Success Metrics (How We'll Know You're Winning) Reduced time-to-launch for new conversational experiences via reusable patterns Higher containment and task completion with stable CSAT/NPS Lower latency and fewer failure loops / escalations Measurable improvement in reliability (SLO attainment), observability, and incident reduction Consistent governance of prompts/tools/models with safe rollout & rollback practices Leadership Competencies Architectural judgment: picks the simplest thing that will work (and scales) Pragmatic AI mindset: knows where LLMs shine and where deterministic logic must own the flow Influences without authority; builds alignment across product, design, security, and ops Mentors and raises the bar across engineering This role is not a "lead developer on one team." It is a cross-domain architecture owner responsible for platform standards and enterprise adoption. Work-At-Home Requirements
WAH requirements: Must have the ability to provide a high speed DSL or cable modem for a home office. Associates or contractors who live and work from home in the state of California will be provided payment for their internet expense. A minimum standard speed for optimal performance of 25x10 (25mpbs download x 10mpbs upload) is required. Satellite and Wireless Internet service is NOT allowed for this role. A dedicated space lacking ongoing interruptions to protect member PHI / HIPAA information Travel: While this is a remote position, occasional travel to Humana's offices for training or meetings may be required. Scheduled Weekly Hours 40 Pay Range $172,200 - $236,900 per year This job is eligible for a bonus incentive plan. This incentive opportunity is based upon company and/or individual performance. Description of Benefits Humana, Inc. and its affiliated subsidiaries (collectively, "Humana")