Staff Network Engineer (AI Infrastructure)

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Staff Network Engineer (AI Infrastructure) based in Canada. The Staff Network Engineer will design and build the networking automation that powers next-generation AI cloud infrastructure. This highly technical role focuses on automating the provisioning and management of high-performance InfiniBand and Ethernet network fabrics for large-scale AI workloads. You will develop software-driven solutions that configure network switches, maintain consistent network states, and support reliable connectivity across tenant clusters. Working within a small, collaborative R&D environment, you will help shape the technical direction of networking capabilities across the platform. You will also test solutions in real production environments, turning operational insights into scalable product improvements. This is an opportunity to influence the future of AI infrastructure while taking ownership of a critical engineering domain. Accountabilities: - Design and develop automation systems for provisioning and configuring NVIDIA switches and InfiniBand network fabrics, including integration with NVIDIA UFM. - Build network configuration management and synchronization systems that maintain consistent switch, VLAN, and VXLAN states, detect configuration drift, and safely reconcile differences in production. - Develop NAT gateway and Layer 4 load-balancing capabilities to provide reliable connectivity for tenant clusters, with robust failover mechanisms. - Define network architecture and behavior at scale, addressing BGP route table limits, convergence, failover, and resilience when links, switches, or gateways fail. - Deploy and test networking automation in real AI cloud environments, identify operational challenges, and translate lessons learned into product improvements. - Establish technical direction for networking capabilities, determining which solutions should be developed internally and which should integrate with partner technologies.

- Produce technical design documents, requirements, and engineering standards to guide networking development across the platform. - Collaborate with engineering stakeholders to build scalable, reliable, and automated networking solutions for demanding AI workloads. Requirements: - Proven experience in network automation, including developing software that programmatically configures switches, routers, or load balancers for production platforms or internal infrastructure. - Hands-on experience with NVIDIA networking technologies such as NVIDIA UFM, Spectrum or Quantum switches, Cumulus Linux, or NVUE, including working with their APIs. - Deep understanding of networking technologies and protocols, including InfiniBand, VLAN, VXLAN, EVPN, and BGP. - Experience operating large-scale production networks, troubleshooting complex failures, and implementing automated solutions rather than relying solely on manual runbooks. - Strong software engineering skills, with production coding experience in Go or Python. - Familiarity with version control, code reviews, and infrastructure or network configuration managed as code. Additional assets: - Experience optimizing network fabrics for GPU training or inference workloads, including RoCEv2, congestion control, rail-optimized topologies, or Spectrum-X. - Background in network automation at a GPU cloud provider, hyperscaler, or large content delivery network. - Experience developing Layer 4 load balancers or NAT gateways, including technologies such as eBPF/XDP or DPDK. - Contributions to open-source networking or Kubernetes projects such as SONiC, FRRouting, Cilium, or MetalLB. Benefits: - Competitive annual salary ranging from CAD $180,000 to $210,000 . - Equity awards and potential bonus compensation.