AI Cluster Networking Performance Lead

Cisco; Milpitas (CA); On-site; USD 168,000 - 245,000; Full time

Benefits offered by this job

Medical, dental, vision insurance

401(k) with company match

Paid parental leave

Disability coverage

Life insurance

Holiday and vacation benefits

Job summary

Cisco is seeking a highly skilled engineer to join the CVIS team in Milpitas, CA. You will investigate kernel failures, boot-time anomalies, PCIe issues, and network latency to validate AI clusters before deployment.

You will tune RDMA/NCCL, optimize SmartNIC/DPU paths, read traces and logs, write code in C/C++/Python, and partner with hardware and software teams to remove end-to-end bottlenecks and deliver reliable, scalable performance.

Qualifications

• Bachelor's degree with 7+ years related experience (or equivalent).
• 4+ years Linux kernel, networking, GPU, and distributed-systems debugging experience.
• Proficient in C/C++/Python; experienced with user-mode and kernel-mode drivers.

Responsibilities

• Investigate kernel-level failures, boot-time anomalies, PCIe link issues, driver behavior, and network latency.
• Diagnose and tune RDMA/RoCEv2, NCCL, SmartNIC/DPU, NIC queues, congestion, throughput, and collective communications.
• Read and correlate unified hardware traces across CPUs, PCIe switches, SmartNICs, and GPU streams.
• Capture and analyze Wireshark packets, system logs, NVIDIA logs, traces, profiles, and observability data.
• Partner with hardware and software teams to eliminate end-to-end bottlenecks.

Skills

Linux kernel debugging

C/C++/Python

RDMA/NCCL

Networking performance

Packet capture analysis

Education

Bachelor's degree in CS/EE

Tools

Nsight profiling

Wireshark

PCIe tracing

Job description

Cisco is seeking a highly skilled engineer to join the CVIS team in Milpitas, CA. You will investigate kernel failures, boot-time anomalies, PCIe issues, and network latency to validate AI clusters before deployment.

You will tune RDMA/NCCL, optimize SmartNIC/DPU paths, read traces and logs, write code in C/C++/Python, and partner with hardware and software teams to remove end-to-end bottlenecks and deliver reliable, scalable performance.