Sr Ethernet AI Network Engineer
Core
Design, deploy, and maintain high-performance Ethernet fabrics supporting AI and HPC environments, ensuring network infrastructure handles distributed training workloads and large-scale GPU cluster communications.
Role type
Senior IC Ethernet Network Engineer (AI/HPC)
Builds
High-bandwidth Ethernet fabrics for AI and HPC clusters
Domain
Data center networking, AI infrastructure, HPC
Deliverable
production ML models
Required skills
Spine-leaf architecture, BGP, EVPN/VXLAN, MLAG, ECMP, QoS, PFC, RoCE, optical transceivers, switch CLI, network automation, packet analysis, documentation
Preferred skills
AI networking fabrics, SONiC/Cumulus Linux, streaming telemetry, Prometheus, Grafana, SRE models
Technologies
BGP, EVPN, VXLAN, MLAG, ECMP, QoS, PFC, RoCE, SONiC, Cumulus Linux, Prometheus, Grafana
Responsibilities
Deploy and validate Ethernet network fabrics for AI/HPC clusters; Troubleshoot network issues across Layers 1-4 (optics, cabling, VLANs, routing); Validate network readiness for distributed AI training; Diagnose performance issues (buffer pressure, microbursts, congestion); Collaborate with Linux/storage teams to isolate host vs network issues; Execute network change plans (firmware, topology); Analyze packet-level data and telemetry; Establish operational standards for fabric health.
