ML Infra Engineer, Modeling

Physical Intelligence is seeking an ML Infra Engineer to scale and optimize AI training systems and core model code. You will own critical infrastructure for large-scale training, manage compute resources, and build efficient JAX training pipelines, collaborating closely with researchers and model engineers.

• Own training/inference infrastructure design and maintenance
• Scale JAX-based distributed training on TPU/GPU clusters
• Optimize performance for memory, device utilization, and throughput
• Build abstractions for rapid experiment iteration and debugging
• Partner with researchers to translate needs into infra capabilities
• Contribute to core JAX model and training code evolution