ML Infra Engineer, Data Systems

Physical Intelligence is seeking an ML Infra Engineer to build and operate the data infrastructure powering large-scale robot learning. This role involves designing and implementing high-throughput data ingestion, processing, and storage systems, optimizing performance for training, and ensuring operational correctness. You will collaborate with researchers and roboticists to translate data needs into robust systems.

• Design and build high-throughput data ingestion and processing pipelines
• Operate large-scale batch and streaming workflows
• Design object storage layouts and metadata systems
• Optimize dataloaders, sharding, and caching for training performance
• Implement observability and validation for data correctness
• Collaborate with researchers and roboticists on data needs