Senior Staff/Principal Engineer, Foundation Model Compute Infra
Core
Architect, build, and operate large-scale infrastructure powering foundation model training, fine-tuning, evaluation, and inference.
Role type
Senior Staff/Principal Engineer (Foundation Model Compute Infra)
Builds
Distributed scheduling systems, accelerator fleets, sandbox infrastructure for AI agents, and scalable inference services.
Domain
AI Infrastructure / Distributed Systems / Cloud Computing
Deliverable
production ML models
Required skills
Distributed systems design, Kubernetes, Python/Go/C++, Accelerator infrastructure, Fault tolerance, Cluster orchestration, Autoscaling, Technical leadership
Preferred skills
TPU/GPU fleet operations, Large-scale distributed training, JAX/XLA, LLM-based agents, Multi-tenant quota management
Technologies
Kubernetes, TPU, GPU, Python, Go, C++, JAX, XLA
Responsibilities
Design distributed scheduling and capacity management systems; Improve fault tolerance for thousands of accelerators; Architect Kubernetes-based infrastructure; Integrate new TPU generations; Build secure sandbox infrastructure for AI agents; Develop self-service developer tools and observability systems; Partner with researchers to translate requirements into production infrastructure.
Seniority
Principal, hands-on technical leadership
