Skip to content
AI.info

jobs

ML Infra Engineer, Compute

Who We Are Physical Intelligence is bringing general-purpose AI into the physical world. We are a group of engineers, scientists, roboticists, and company builders developing foundation models and…

Company
Physical Intelligence
Location
San Francisco
Status
Open
Posted
2026-11-16 21:20 UTC

Physical Intelligence is hiring an ML Infrastructure engineer to own scheduling and compute placement for large-scale training across a mixed GPU and TPU fleet spanning hardware generations, cloud providers and cluster topologies. The person will build multi-tenant scheduling with quotas, priority tiers and preemption, a control plane for job lifecycle, migration and failover, fault detection for long-running multi-node jobs, plus observability tooling, and extend orchestration to inference workloads on robots. The posting wants strong software engineering fundamentals, experience operating scheduling or resource-management systems on large clusters, and familiarity with SLURM, Kubernetes or internal equivalents. It is distinguished by treating intelligent placement as a systems problem rather than cloud DevOps.

Original job posting