Machine Learning Engineer -Ethernet/IP networking, including L2/L3 forwarding, BGP, ECMP, EVPN/VXLAN, VRFs, ACLs, QoS, 5 to 8 years
Core
Plan, execute, and analyze AI/ML network benchmarks to validate high-performance Ethernet fabrics and translate results into reference architectures and deployment guidance.
Role type
Senior IC infrastructure networking engineer (AI/ML benchmarks & validation)
Builds
Validated L3 leaf-spine fabric designs, benchmark reports, configuration templates, and deployment recommendations for AI/ML data center networks.
Domain
Data center networking, AI/ML infrastructure, Ethernet/IP, RoCE
Deliverable
production ML models | product features | dashboards & analysis | research | client delivery | infrastructure
Required skills
L3 leaf-spine fabric design, BGP (eBGP/iBGP, unnumbered, EVPN/VXLAN), Ethernet/IP networking (L2/L3 forwarding, ECMP, VRFs, ACLs, QoS), Linux systems administration, Python scripting, network benchmarking methodology, performance analysis (throughput, latency, congestion, failover), automation (Ansible, Bash, CI/CD), RDMA/RoCEv2 concepts, switch OS configuration (SONiC, NX-OS)
Preferred skills
Switch ASIC behavior validation, SRv6/segment routing, AI-assisted automation workflows, technical presentation to executives and customers
Technologies
eBGP, iBGP, EVPN, VXLAN, VRF, ECMP, SRv6, SONiC, NX-OS, NCCL, RCCL, ib_write_bw, ib_read_bw, ib_send_bw, ib_write_lat, netperf, iperf, MPI, OSU, Python, Ansible, Bash, Git
Responsibilities
Plan and execute AI/ML network benchmarks covering throughput, latency, congestion, and failover; Design and validate L3 leaf-spine fabrics using BGP and overlay technologies; Define benchmark methodology, KPIs, and reporting standards; Create reference architectures and deployment recommendations; Run performance tests using industry-standard tools; Troubleshoot network issues in AI/ML infrastructure environments
Seniority
Senior, hands-on IC
