Building Machine Learning Infrastructure for Autonomous Vehicles - AMZ304

Explore how Zoox revolutionizes autonomous vehicle development through AWS Machine Learning infrastructure in this 55-minute conference talk from AWS re:Invent 2025. Discover the comprehensive approach to processing thousands of multi-modal sequences daily using Amazon SageMaker HyperPod, SLURM, FSx, S3, and EKS within a streamlined machine learning environment designed for both rapid experimentation and production-scale training. Learn about the integrated monitoring solution that achieved an 85% reduction in manual analysis while scaling to support thousands of vehicle analysis use cases. Gain insights into the technical architecture and best practices for building robust ML infrastructure that can handle the complex demands of autonomous vehicle development, including data processing workflows, scalable training environments, and monitoring systems that ensure reliable performance at scale.