AI Deployment Lead

Booz Allen Hamilton•Washington, DC

1d•$86,800 - $198,000•Remote

About The Position

GRACE is ARPA-H's AI assistant, and its value depends on how well ARPA-Humans put it to work. We are looking for a startup-minded AI Deployment Lead with hands-on experience designing agent and multi-agent workflows on top of existing LLM platforms, and a track record of training people to use new tools well. You will own the relationship between GRACE and the program teams across ARPA-H, from first training session through ongoing use, and you'll be the primary channel bringing what you learn back to the GRACE product, engineering, and design team. This role is ideal for someone who has worked directly with customers or internal teams in high-velocity environments including startups or big tech, listens more than they talk, and wants to apply practical AI across an enterprise. You'll do this while helping teams understand GRACE's capabilities and limits within the constraints of our mission and the broader federal environment including privacy, security, safety, and responsible use by default.

Requirements

5+ years of experience in a customer-facing technical role such as product management or product operations, technical account management, or solutions engineering
Experience with solutions engineering, technical account management, sales engineering, or forward deployed work
Experience in high-velocity environments including tech startups or large-scale industry roles where you owned a customer or stakeholder relationship end to end
Experience building with modern LLM tools, including prompt engineering and agent or workflow construction
Experience with product management or product operations, with an instinct for prioritizing feedback and translating it into requirements
Experience driving adoption of a new tool or process across a busy organization
Ability to train non-technical audiences on technical tools
Ability to navigate ambiguity and figure out the next right step without a lot of direction
Ability to prioritize feedback and translate it into requirements
Bachelor's degree

Nice To Haves

Experience with agent frameworks or workflow builders including LangChain, CrewAI, or low-code tools such as Zapier or Make
Experience with Model Context Protocol (MCP)
Possession of excellent active listening skills to collaborate
Possession of excellent customer service and people reading skills to anticipate their needs

Responsibilities

Work directly with program teams to identify high-value workflows, and build working prompt structures, agent chains, and templates that solve real problems.
Build a reusable library of prompt patterns, agent templates, and playbooks other teams can pick up directly.
Run onboarding sessions, office hours, and team-specific training tailored to how different groups across ARPA-H work.
Grow a network of power users and champions across program teams.
Sit in on real workflows, gather what's working and what's broken, and turn it into clear, prioritized input for the GRACE team.
Translate ambiguous requests into specific product requirements.
Track usage, training completion, and workflow adoption across the agency.
Build a clear view of where adoption is working and where it needs attention and bring it to leadership regularly.
Establish and improve training materials, playbooks, and best practices as GRACE's capabilities evolve.