What if every team at Amazon Fulfillment could build their own AI agent — one that already knows their systems, their data, and their processes — without writing model code?
That's what we're building.
We're the team behind
IRIS AI Studio, a self-service platform where engineering teams across Amazon's fulfillment network create, deploy, and operate custom AI agents.
Teams connect their knowledge and tools, and walk away with a production-grade agent that can investigate incidents, answer domain-specific questions, and take real action — all through a no-code experience.
What makes this role different: You're not building one AI product.
You're building the platform that lets hundreds of teams build their own.
You'll navigate the classic platform challenge: making something flexible enough for advanced use cases while keeping it simple enough for anyone to use.
And unlike most AI platforms, yours operates in a world with physical consequences — when an agent gets it right, a facility avoids downtime;
when it gets it wrong, packages don't move.
You'll work across the full stack — from the studio experience where agents are assembled, to the runtime infrastructure that keeps them reliable, to the integrations that connect agents to real operational systems.
You'll own problems end-to-end, from design through production, and your work will shape how Amazon Fulfillment adopts AI across its global operations.