AgentsThe story, in brief
Presentation: Building GenAI Platform at DoorDash
5,000 internal users. DoorDash's GenAI platform shows how to navigate vendor vs. open-weight tradeoffs at scale.
The KeyNews take
Why it matters
DoorDash's public architecture walkthrough provides practitioner details on building internal agent platforms: how they transitioned from vendor models to open weights, designed LLM and agent gateways, and balanced accuracy/latency/cost across a large user base. Concrete for teams evaluating similar builds.
The key facts
5 to know5,000+ internal users on DoorDash GenAI platform
Architectural transition from vendor-first to open-weight models
LLM and agent gateways implemented as core design layer
Balancing accuracy, latency, and cost optimization discussed
Multi-speaker presentation format (Chitlur, Kodwani) with architectural detail
Go to the source
InfoQ AI/MLinfoq.com
Publisher excerpt: Swaroop Chitlur and Sidd Kodwani share DoorDash’s journey building an internal GenAI platform. They discuss core architectural bets, transitioning from vendor-first setups to open-weights models, navigating LLM and agent gateways, and balancing accuracy, latency, and cost for over 5,000 internal…