Infrastructure Engineer, GPU Clusters
Core
Own and evolve multi-cluster GPU infrastructure to support training tabular foundation models at scale, managing cluster architecture, scheduling, and cost optimization.
Role type
Senior Infrastructure Engineer (GPU Clusters)
Builds
Multi-provider GPU cluster orchestration, CI/CD pipelines, and internal tooling for research iteration
Domain
AI Infrastructure / High-Performance Computing / Cloud
Deliverable
infrastructure
Required skills
Cluster orchestration (Slurm, Kubernetes), Cloud infrastructure (GCP, AWS, Azure), CI/CD automation, Python, Linux systems engineering, Distributed systems, Cost optimization, Infrastructure as code
Preferred skills
GPU infrastructure management, PyTorch, GPU profiling, Multi-cloud HPC experience, Experiment tracking tooling
Technologies
Slurm, SkyPilot, GCP, AWS, Azure, Docker, Terraform, GitHub Actions, Python, uv, wandb
Responsibilities
Drive architecture and automated provisioning for multi-cluster GPU infrastructure; Build developer productivity layer including CI/CD and reproducible environments; Improve reliability through monitoring, observability, and automation; Guide compute spending decisions based on cost per FLOP analysis
Seniority
Senior, hands-on IC