Manager Conversational AI Design

Blue Cross Blue Shield of Michigan•Detroit, MI

About The Position

This role is responsible for designing, optimizing, and governing the conversational content and prompt architecture that powers AI-enabled member self-service. It translates complex healthcare, insurance, benefits, and operational terminology into clear member language and influences model behavior through grounded prompt design, conversational examples, testing, feedback, monitoring, and continuous refinement. The role partners with Product Management, Product Owners, design, engineering, data and AI teams, member experience, operations, legal, compliance, privacy, clinical, and business subject-matter experts. It ensures conversations are understandable, accurate, trusted, accessible, and aligned with product outcomes and organizational standards. The position leads conversational content and prompt strategy, establishes plain-language and health-literacy standards, maps member intents to appropriate health plan concepts and trusted information sources, and designs and refines prompts, response instructions, clarification patterns, and guardrails. It establishes response boundaries and human handoffs, evaluates conversational quality, and turns requirements into implementation-ready content, test cases, evaluation criteria, and backlog priorities. It also maintains governance for conversational content and prompts, including documentation, version control, review standards, approved terminology, grounding requirements, risk controls, monitoring, and change processes.

Requirements

Bachelor’s degree in Communications, Content Strategy, English, Journalism, Marketing, Human-Computer Interaction, User Experience, Information Science, Business, Health Administration, Computer Science, or a related field.
At least seven years of progressively responsible experience in conversational design, content strategy, user experience writing, digital content, prompt design, applied AI, or a related role.
At least one year of hands-on experience working with large language models, including prompt development, testing, evaluation, and refinement.
Demonstrated experience creating consumer-facing conversational content for complex digital products, self-service experiences, chatbots, virtual assistants, or AI-enabled solutions.
Ability to translate complex domain, policy, operational, or technical concepts into clear, accurate language that reflects how consumers naturally communicate and meets approved plain-language, accessibility, inclusive-language, and digital voice standards.
Strong knowledge of conversational design, content hierarchy, progressive disclosure, intent recognition, clarification strategies, response architecture, tone, and consistent voice across supported digital self-service channels.
Working knowledge of large language models, generative AI, model limitations, grounding, retrieval-augmented generation, evaluation, monitoring, human oversight, and responsible AI practices.
Ability to identify unsupported responses, ambiguous member intent, high-risk topics, information dependencies, downstream impacts, and situations requiring clarification, escalation, disclaimers, or human intervention.
Strong stakeholder leadership and influence skills, including the ability to align product, design, engineering, data and AI, operations, legal, compliance, privacy, security, clinical, accessibility, language-access, and business experts.
Strong knowledge of privacy, consumer communication, accessibility, language-access, and regulatory considerations relevant to consumer-facing AI experiences.
Excellent writing, editing, facilitation, presentation, and storytelling skills, with exceptional attention to clarity, accuracy, consistency, tone, and detail.

Nice To Haves

Master’s degree or equivalent advanced education.
Experience in healthcare communications, health insurance, benefits, clinical content, health plan operations, or another regulated consumer environment.
Experience designing and evaluating structured prompts, templates, conversational examples, model instructions, test scenarios, quality criteria, grounding requirements, and guardrails.
Experience analyzing model outputs and refining patterns based on evidence.
Experience using research, analytics, experimentation, quality reviews, and user feedback to improve conversational content and model behavior.
Equivalent or relevant BCBSM experience may be considered to satisfy the stated requirements.
Demonstrated expertise in digital content, AI-enabled solutions, strategic thinking, and business impact may be considered in lieu of certain experience requirements.

Responsibilities

Lead conversational content and prompt strategy for AI-enabled member self-service.
Define member-friendly wording, conversational tone, response length and structure, progressive disclosure, follow-up patterns, and a consistent voice across supported self-service channels.
Establish plain-language and health-literacy standards, including target reading levels where appropriate, aligned with member needs, product objectives, brand expectations, accessibility, and trust.
Translate complex healthcare, insurance, benefits, clinical, policy, and operational terminology into clear language that reflects how members naturally ask questions.
Map conversational intent to appropriate health plan concepts, processes, actions, escalation paths, and approved information sources, including evidence of coverage and other trusted enterprise content.
Design, test, and refine grounded system prompts, prompt templates, conversational examples, response instructions, clarification prompts, and guardrails that influence model behavior.
Determine what information should be presented immediately, progressively disclosed, offered through follow-up prompts, or escalated to another channel.
Establish response boundaries and human handoffs to reduce, detect, and appropriately manage unsupported or inaccurate responses, and prevent medical advice, guaranteed coverage, or communication beyond approved benefit and policy information.
Lead evaluation of conversational quality using member research, analytics, test scenarios, model outputs, failure patterns, feedback, and operational insights.
Define quality measures and acceptance criteria for clarity, accuracy, relevance, tone, health literacy, completeness, grounding, safety, accessibility, consistency, appropriate escalation, and multilingual or language-access needs where applicable.
Partner with product, experience design, engineering, data and AI teams, operations, and subject-matter experts to translate conversational requirements into implementation-ready content, prompt assets, test cases, model-evaluation criteria, and backlog priorities.
Establish and maintain conversational content and prompt governance, including version control, documentation, reusable patterns, review standards, approved terminology, grounding requirements, risk controls, monitoring, and change processes.
Partner with legal, compliance, privacy, security, clinical, accessibility, language-access, and governance teams to ensure sensitive or regulated content meets applicable organizational and regulatory requirements.