Staff Networking Platform Engineer, Infrastructure

This is a fully remote position, open to applicants in California, +2 more states.

📋 Description

• Define how networking functionalities are requested, provisioned, altered, monitored, and decommissioned.

• Automate the lifecycle of tenant and node networks using Kubernetes-native tools, ensuring safe recovery from partial failures.

• Manage BGP, routing policies, VLANs, and overlays, along with VIPs, load balancers, and public IPs via fabric APIs.

• Automate GPU networking, including NVL72, InfiniBand, RoCE, and GPUDirect.

• Validate network isolation and performance metrics.

• Oversee IPAM and inventory integration to ensure that intended configurations, applied states, and inventories remain synchronized.

• Establish segmentation and traffic policies in collaboration with security engineering.

• Enforce standards using GitOps and conduct drift reconciliation.

• Manage network telemetry, troubleshooting, and incident response tasks.

• Develop networking features for a managed Kubernetes platform that transforms bare-metal GPU compute into dedicated, isolated customer clusters.

⛳️ Requirements

• Must reside in Dallas, San Francisco, or New York City.

• Extensive knowledge of networking concepts: TCP/IP, BGP, routing policies, ECMP, segmentation, and interconnects.

• Proven production ownership of bare-metal leaf-spine data center networks utilizing VLANs, VRFs, and EVPN/VXLAN.

• Experience with Kubernetes and Linux networking, including CNI behavior, Cilium or similar technologies, and network policy.

• Skilled in building Kubernetes controllers, operators, or custom resources using Go.

• Familiarity with GitOps practices and controlled rollout processes.

• Experience with multitenant infrastructure or managed Kubernetes solutions.

• Proficient in securing networks that support customer workloads.

• Nice to have: Knowledge of Netris, UFM, or NetBox.

• Nice to have: Experience with NVIDIA AI networking, including InfiniBand, Spectrum-X, RDMA, RoCE, and congestion control.

• Nice to have: Networking expertise for large GPU clusters, including multi-site connectivity.

• Must currently be authorized to work in the U.S. or indicate if sponsorship is necessary.

🏝️ Benefits

• Predominantly cash compensation.

• Bonus integrated into base salary.

• Equity component in addition to base compensation.

• Remote work option available for the time being.

• Hybrid work model planned for the future.