Staff Network Engineer (AI Fabric, Datacenter and Edge Networking) - Radian Arc
JobgetherRemotely
bgpvxlanroceevpngpunvidiardmamellanox
Job Description
📋 Description
- Design, deploy, and operate GPU networking fabrics for distributed AI workloads (RoCE, RDMA
- Optimize AI GPU communication patterns; improve latency, throughput, and reliability.
- Own Layer 2/3 datacenter networking (BGP, ECMP, EVPN/VXLAN, VLAN/VRF, OVS/OVN, Linux).
- Design north-south connectivity, WAF protections, TLS termination, and DDoS mitigation.
- Plan inter-datacenter private connectivity, DWDM, inter-site BGP routing, redundancy.
- Lead infrastructure initiatives from architecture to production rollout.
🎯 Requirements
- Hands-on datacenter networking with BGP, OSPF, ECMP, EVPN/VXLAN.
- AI networking expertise for GPU fabrics; RoCE/RDMA experience.
- Strong NCCL knowledge and GPU communication patterns (all-reduce, all-gather, etc.).
- RoCE optimization experience; PFC/ECN congestion management; diagnose stalls and congestion.
- Production infrastructure experience with high-speed Ethernet, NVIDIA/Mellanox gear.
- Systems troubleshooting across hardware, firmware, Linux kernel, and distributed apps.
🎁 Benefits
- Full-time or contract engagement; 100% remote within Europe.
- Autonomy, collaboration, and international work culture.
- Chance to shape networking architecture for AI infra at scale.
- Exposure to AI fabrics, RoCE/RDMA, GPU infrastructure, edge computing.
Back to all jobs