NEWS · COMPANIES · #1472
DoorDash details architecture and lessons from building an internal GenAI platform
At QCon AI, DoorDash engineers Swaroop Chitlur and Siddharth Kodwani presented the company’s journey building an internal GenAI platform for ~5,000 internal users, covering architectural bets, the shift from vendor-first to open-weights models, LLM and agent gateway design, and trade-offs between accuracy, latency, and cost. They emphasized API-/SDK-first design, focusing on product engineers as customers, embedding best practices, and prioritizing business impact over prototypical chatbots.
KEY POINTS
- At QCon AI, DoorDash engineers Swaroop Chitlur and Siddharth Kodwani presented the company’s journey building an internal GenAI platform for ~5,000 internal users, covering architectural bets, the shift from vendor-first to open-weights models, LLM and agent gateway design, and trade-offs between accuracy, latency, and cost.
- They emphasized API-/SDK-first design, focusing on product engineers as customers, embedding best practices, and prioritizing business impact over prototypical chatbots.
- Provides a practitioner-level playbook from a large product company on how to operate and scale an internal GenAI platform, including vendor-to-open-model transitions and agent gateway design.
WHY IT MATTERS
Provides a practitioner-level playbook from a large product company on how to operate and scale an internal GenAI platform, including vendor-to-open-model transitions and agent gateway design.