Senior Data Center Network Engineer
San Jose, CA (On-site)
About the role
We are seeking a Senior Data Center Network Engineer to design, deploy, and operate high performance networks that power our next-generation compute infrastructure. This role focuses on low-latency, high-throughput network fabrics supporting AI and HPC workloads, including InfiniBand (IB) and RoCEv2. You will be responsible for building and maintaining the network architecture that underpins large-scale, self-owned compute clusters, ensuring reliability, scalability, and performance. You’ll work hands-on across network systems, automation, and infrastructure, collaborating closely with hardware and systems teams in a fast-paced, technically demanding environment.
Responsibilities
Design and Implementation: Architect and deploy scalable network infrastructure for HPC and AI clusters using Ethernet, InfiniBand, and RoCEv2 technologies.
Operations and Reliability: Maintain high network uptime and performance through proactive monitoring, configuration management, and troubleshooting.
Performance Optimization: Diagnose and resolve complex issues across Layers 1–3, ensuring low latency and efficient data flow across compute clusters.
Automation and Efficiency: Develop and maintain scripts and automation frameworks (Python, Ansible, etc.) to streamline configuration, deployment, and monitoring tasks.
Capacity and Growth Planning: Analyze utilization trends and forecast capacity requirements to ensure long-term scalability.
Monitoring and Analysis: Implement and use monitoring systems (Prometheus, Grafana, ELK, etc.) and packet analysis tools (tcpdump, Wireshark) to identify and prevent performance bottlenecks.
Documentation and Standards: Produce clear documentation, topology diagrams, and operational runbooks for current and future deployments.
Cross-Team Collaboration: Work closely with systems, hardware, and operations teams to ensure seamless integration and coordinated scaling across the stack.
Vendor Management: Evaluate and collaborate with hardware and software vendors to select networking solutions optimized for high-density compute environments.
Leadership and Mentorship: Act as a senior technical resource, mentoring other engineers and contributing to best practices and design standards.
Required Qualifications
8+ years of experience in data center network engineering, including at least 3 years in a senior or lead role.
Deep understanding of Layer 2/3 network protocols, including BGP, OSPF, VXLAN, PIM, IGMP, and spine-leaf architectures.
Hands-on experience deploying and operating InfiniBand (HDR/NDR) and RoCEv2 fabrics for HPC or AI environments.
Proven ability to design and optimize large-scale, high-performance network fabrics with a focus on reliability and low latency.
Experience with Arista (EOS), Cisco (NX-OS), NVIDIA (Cumulus/Onyx), or SONiC-based network platforms.
Strong Linux/Unix command-line and networking diagnostic skills.
Scripting experience with Python, Bash, or equivalent for network automation or data analysis.
Proficiency in packet analysis and troubleshooting using tools like tcpdump, Wireshark, or similar.
Knowledge of fiber optics technology and cabling standards (1–800 Gbps), including experience selecting transceivers and interconnects.
Strong written and verbal communication skills, including technical documentation and cross-functional communication.
Preferred Qualifications
Experience supporting AI or HPC compute clusters and GPU-based workloads.
Familiarity with network telemetry, zero-touch provisioning, and configuration management tools (e.g., Ansible, Salt).
Knowledge of data center power, cooling, and rack-level design considerations for dense compute environments.
Exposure to public cloud networking (AWS, GCP, Azure) or hybrid designs.
Relevant industry certifications (e.g., CCNP, CCIE, Arista ACE) are a plus.