Director, AI Application Engineering

Marriott•Bethesda, MD

16h•$119,000 - $215,000•Hybrid

About The Position

Use Case Engineering is the production delivery arm of Marriott's AI Engineering organization. We embed forward deployed AI engineers with teams across the enterprise, carry the company's highest priority AI use cases from intake through production, and turn each delivery into reusable patterns, SDKs, and reference architectures that make the next one faster. As Director, AI Application Engineering you own the engineering of AI applications and agent systems across a portfolio of enterprise use cases. You lead a group of forward deployed engineers, set the architecture, and stay hands on in the code. Your portfolio carries the hardest engagements: multi system integrations, vendor engineering teams delivering alongside us, and decisions that set precedent for the rest of the AI program. You work across build, consult, advise, and monitor engagements at the same time, and you are accountable both for what reaches production and for what happens to it afterward. Reports to the Senior Director, AI Engineering.

Requirements

8+ years in software engineering, including 3+ years building AI or machine learning applications, enterprise integrations, or platform services; or 6+ years with a Master's degree
Production grade Python and/or TypeScript; Rust is a plus
Hands on LLM application development: agent design, tool calling, prompt engineering, retrieval, and evaluation
Working depth in the agent interoperability stack: MCP, A2A, Watson ADK, and LangChain or comparable frameworks
Technical leadership across several concurrent teams or deliveries
Daily use of AI powered coding assistants as working engineering tools
Experience directing vendor engineering teams or systems integrators

Nice To Haves

Two or more enterprise AI platforms (AWS, Azure or Vertex, watsonx, Writer, Copilot Studio, or comparable)
Production retrieval systems: vector stores, embedding models, chunking and context strategy, offline and online evaluation
Multi agent orchestration frameworks or vendor native agent SDKs
Reusable engineering assets, such as SDKs, libraries, or reference architectures, adopted by teams outside your own
Enterprise governance and compliance workflows for AI, including security, privacy, and model risk
REST API integration design in a large enterprise, and architecture documentation that stands up to a review board
Hospitality, travel, or large scale consumer services

Responsibilities

Lead forward deployed engineers delivering AI applications and agents for a business vertical, holding the technical bar from intake through production handoff
Design and build agent systems, tool and API integrations, and retrieval pipelines on the enterprise AI platform stack, using vendor SDKs and agent development kits
Own architecture decisions for your portfolio and carry them through the enterprise governance gates, including agent tier classification, security engagement review, and architecture review
Direct vendor and systems integrator engineering teams delivering alongside your own, setting the design and holding the quality line
Stand up evaluation and monitoring for everything you ship, so agent quality is measured rather than asserted
Contribute working patterns, SDK modules, and reference implementations back to the shared library after every build, and retire the ones that stop earning their place
Build handoff packages, including runbooks, dashboards, and documentation, that let business and platform partners own solutions long term
Mentor engineers across the function and set architectural guardrails others can follow without you in the room