Senior AI Cluster Network

Responsibilities

• Design, deploy and maintain high-performance network infrastructure supporting large-scale GPU clusters and distributed AI/ML workloads.
• Implement InfiniBand and/or high-performance Ethernet/RoCE fabrics according to the target architecture.
• Configure and troubleshoot high-speed network switches, network interfaces, optical transceivers and fabric connectivity.
• Deploy and support spine-leaf network architectures for AI data centers.
• Troubleshoot RDMA connectivity, network congestion, packet loss, latency and throughput issues.
• Support NVIDIA GPU infrastructure and high-bandwidth communication between compute nodes.
• Collaborate with GPU infrastructure, Linux, Kubernetes, storage, security and data center operations teams.
• Develop network monitoring, troubleshooting procedures, technical documentation and operational runbooks.
• Support incident response, root-cause analysis and network reliability improvements.

Requirements

• Strong hands-on experience in large-scale data center networking, HPC networking or AI cluster network infrastructure.
• Experience with InfiniBand or RoCE-based Ethernet fabrics.
• Strong understanding of RDMA, high-bandwidth networking, and low-latency communication.
• Hands-on experience deploying and troubleshooting high-speed network switches.
• Knowledge of network performance monitoring, congestion management, and fabric troubleshooting.
• Strong Linux networking and troubleshooting skills.
• Experience with network configuration, deployment automation, and infrastructure validation.
• Experience collaborating with infrastructure engineering and data center operations teams.
• Experience supporting distributed AI training environments.
• Strong knowledge of NVIDIA Quantum InfiniBand, Spectrum Ethernet, or equivalent networking platforms.
• Proficiency in Arista, Juniper, or NVIDIA/Mellanox switching technologies.
• Strong understanding of BGP, EVPN, and VXLAN networking protocols.
• Hands-on expertise in 400G/800G networking and optical infrastructure.
• Proficiency in Python, Ansible, and zero-touch provisioning (ZTP).
• Strong understanding of Kubernetes networking and GPU cluster integration.
• Knowledge of Palo Alto, Fortinet, or equivalent network security technologies.

Qualifications

Bachelor's degree or equivalent combination of education and experience.