DoorDash builds an internal agentic AI platform, evolving from deterministic workflows to dynamic multi-agent AI swarms for complex data analysis.

DoorDash offers insights into food delivery technology, logistics, and customer experience. Developers can delve into DoorDash's engineering blogs, tech talks, and innovation stories to gain a deeper understanding of how technology powers the food delivery industry. By exploring topics such as scaling infrastructure, optimizing delivery routes, and enhancing user experience through innovative technology solutions, developers can use insights to apply to their own projects and industries.

Doordash

DoorDash evolved its internal AI platform from deterministic workflows to sophisticated multi-agent systems. The journey progresses through four stages: workflows for repeatable tasks, single agents using ReAct architecture for dynamic reasoning, deep agents with hierarchical task decomposition, and agent swarms for distributed collaboration. The platform includes hybrid search with vector databases, schema-aware SQL generation with validation, LLM-as-judge evaluation, and integrations with Slack and Cursor. Key lessons emphasize building solid foundations first, implementing multi-layered guardrails for trust, and managing context carefully. The system uses LangGraph for computational graphs and adopts open standards like Model Context Protocol for tool access and Agent-to-Agent protocol for inter-agent communication.

Beyond Single Agents: How DoorDash is building a collaborative AI ecosystem

Evolutionary path of multi-agent collaboration

Taking a high-level look at DoorDash’s agentic AI platform