Senior Software Engineer - Cluster Networking
Core
Lead the network architecture for NVIDIA's GPU superclusters, enabling frontier model training across tens of thousands of nodes.
Role type
Senior IC cluster networking engineer
Builds
Kubernetes networking architecture, overlay mesh, L7 gateways, and validation suites for massive-scale GPU clusters
Domain
High-performance computing (HPC) and AI infrastructure
Deliverable
production ML models
Required skills
Kubernetes networking, CNI standards, Linux networking fundamentals, Go/Python/C, distributed system debugging, cloud provider partnership
Preferred skills
Calico, Tailscale/WireGuard, InfiniBand/RoCE/RDMA, Slurm, upstream contributions to networking projects
Technologies
Kubernetes, Calico, Envoy, Cloudflare, Tailscale, WireGuard, Slurm, InfiniBand, RoCE, RDMA
Responsibilities
Own and evolve Kubernetes networking architecture for multi-thousand-node GPU clusters; Design and scale overlay networks (CNI, mesh, VPN); Design and operate L7 gateways/load balancers; Eliminate scale ceilings (packet loss, control-plane saturation); Build scale-test environments and validation suites; Diagnose hard distributed network problems; Partner with cloud providers on network topology
Seniority
Senior, hands-on IC
