Lead GenAI Engineer

Overview Client•South San Francisco, CA

10h•Hybrid

About The Position

The client is seeking a hands-on senior GenAI engineering leader to design and build a stateful, headless GenAI application that enables commercial marketers to create grounded, personalized content. The application will orchestrate multiple agents and enterprise services, maintain conversational and project state, and generate structured JSON or HTML content blueprints for downstream platforms. The role combines architectural leadership with hands-on implementation. The selected consultant will propose the application architecture, make key technical decisions, establish engineering patterns, review technical work, and build critical production components. The role also includes developing streaming GenAI experiences, integrating enterprise platforms and content repositories, and implementing observability, evaluation, grounding, error handling, and production monitoring capabilities. Bay Area candidates are preferred and should be able to work from the office 1–2 days per week. Candidates outside the Bay Area must be able to provide 4–6 hours of daily overlap with Pacific Time.

Requirements

Significant hands-on experience architecting and building production GenAI applications.
Strong software engineering experience using Python or TypeScript, including microservices architecture.
Experience designing and implementing stateful, multi-turn, and multi-agent orchestration systems.
Strong experience with API-first architecture and structured-contract design.
Experience building headless services that produce structured outputs for downstream applications.
Experience integrating GenAI applications with enterprise platforms, data services, and content repositories.
Experience with RAG, grounding, application state, memory management, evaluations, and production observability.
Experience building streaming, multi-turn GenAI interfaces with conversation history, clarification flows, and robust error handling.
Experience rendering structured responses, including tables, charts, citations, and reusable results.
Ability to propose architecture, make technical decisions, establish coding patterns, review engineering work, and develop production code.
Strong experience with AWS, Amazon Bedrock, and AgentCore.

Nice To Haves

Experience with content generation, content-as-code, personalization platforms, or marketing technology ecosystems.
Experience working with Workfront, Veeva Vault PromoMats, AEM Assets, or comparable enterprise content management systems.
Experience in commercial pharmaceutical environments or other regulated content domains.

Responsibilities

Design and build a stateful, multi-agent GenAI application for personalized and grounded content generation.
Architect and implement conversational and project state management, multi-turn interactions, memory, clarification workflows, and agent orchestration.
Develop headless, API-first services that generate structured JSON or HTML outputs for downstream content assembly platforms.
Integrate GenAI applications with enterprise platforms, data services, and content repositories.
Establish technical architecture, coding standards, structured contracts, and key engineering patterns.
Build production-grade components and conduct technical reviews of other engineers’ work.
Implement observability, evaluation frameworks, grounding strategies, error handling, and production monitoring capabilities.
Develop streaming GenAI experiences that support conversation history, structured responses, and reusable outputs.