Supercharging Self-Driving Algorithm Development with Ray: Scaling Simulation Workloads and Democratizing Autotuning

Explore a 33-minute conference talk on leveraging Ray to enhance autonomous driving algorithm development at Zoox. Dive into the company's innovative autotuning platform that accelerates development by utilizing large-scale, distributed simulation and metrics evaluation. Learn how Ray's scalability and fault tolerance capabilities are harnessed to scale simulation and metrics workloads. Gain valuable insights into the autotuning process that enables developers to improve autonomous driving without code changes. Discover key lessons learned during the platform's development and deployment, and get a glimpse into the future of metrics-driven algorithm development. Access the accompanying slide deck for visual references and additional information on this cutting-edge approach to self-driving technology advancement.

Supercharging self-driving algor dev w/ Ray: scaling sim workloads and democratizing autotuning@Zoox