Accelerating LLM-Driven Developer Productivity at Zoox

Over the past year, Zoox has invested in integrating Large Language Models (LLMs) into internal developer workflows through a company-wide initiative called Zoox Intelligence (ZI). This talk shares how we approached this transformation — from identifying high-impact use cases (like code authoring, code review, search, onboarding, and context-aware chatbots) to designing infrastructure, choosing the right LLMs, and embedding them into real developer tools (CLI, IDE, Slack, web apps).

Beyond software development tasks, we’re also tailoring our platform to support workflows unique to autonomous vehicle development, such as triaging events observed by our autonomy fleet, and supporting our eventual customers. Our vision is to create a flexible, intelligent platform that supports both technical and non-technical teams — from engineers to program managers, and beyond.

Whether you're just beginning to explore LLMs in your engineering org or scaling your internal AI capabilities, this session will offer tangible ideas, design patterns, and organizational strategies.