HPC/ GPU Cluster Architect
Core
Architecting, deploying, and operating large-scale HPC and GPU clusters globally for customers, ensuring reliability and performance.
Role type
Senior IC HPC/GPU Cluster Architect
Builds
High-performance GPU clusters and datacenter infrastructure
Domain
Datacenter infrastructure + High-Performance Computing (HPC)
Deliverable
production ML models
Required skills
HPC/GPU cluster architecture, server hardware fundamentals (GPUs, NICs, PCIe, memory, thermals, power), full-stack debugging (hardware, OS, drivers, networking), infrastructure-as-code, operational documentation, mentoring
Preferred skills
Data center operations (power, cooling, colo), Linux systems administration (kernel drivers, RDMA), schedulers/orchestration (Slurm, Kubernetes), virtualization (KVM, QEMU), telemetry pipelines, high-speed fabrics (InfiniBand, RoCEv2)
Technologies
Slurm, Kubernetes, InfiniBand, RoCEv2, KVM, QEMU, libvirt, RDMA
Responsibilities
Architect and deploy new clusters worldwide, maintain on-call rotation for cluster operations, deploy new environments, debug performance and reliability issues, automate fleet operations, mentor junior engineers
Seniority
Senior, hands-on IC
