Building DoorDash Assistant: An engineering overview
DoorDash details the runtime architecture behind its conversational shopping assistant: an Orchestrator agent routing to domain agents (restaurant discovery, grocery), shared Managed Agent Services (artifacts, session, memory), and an MCP tool layer grounding every claim in live catalog data, plus the evaluation harness and AI-assisted workflow the team uses to build it.
More from Careersatdoordash
Building a unified consumer memory for personalization at scale
Personalization at DoorDash spans restaurants, groceries, retail, and more - but capturing true consumer intent requires moving beyond sparse engagement patterns.
Inside DoorDash's one-click simulation and evaluation platform for support chatbots
The LLM Simulator acts as a realistic DoorDash customer interacting with the support chatbot, using structured decision-making prompts to drive dynamic, scenario-based behavior across multi-turn conversations.
