Member of Technical Staff – Datacenter Networking

This is a fully remote position, open to applicants in California.

📋 Description

• Design and implement scalable datacenter network architectures for GPU training, inference, storage, and management traffic.

• Configure and manage high-performance Ethernet/RoCE and InfiniBand networks, adhering to standards for routing, redundancy, and capacity.

• Automate processes for network provisioning, configuration validation, upgrades, and rollback.

• Troubleshoot packet loss, congestion, link failures, and collective communication performance across hosts and switches.

• Benchmark overall network performance with infrastructure and ML teams, converting workload requirements into quantifiable acceptance criteria.

• Develop monitoring systems for port health, errors, utilization, congestion, and fabric topology.

• Enhance incident response procedures and runbooks.

• Collaborate with datacenter operators and hardware vendors on cabling, optics, deployment readiness, and failure resolution.

• Engage directly with customers at the forefront of AI innovation.

• Work alongside the engineering team to support systems that drive AI advancements.

⛳️ Requirements

• A minimum of 3 years of experience in production datacenter networking.

• Comprehensive knowledge of Ethernet, TCP/IP, routing, switching, and redundant network design.

• Practical experience in high-performance GPU networking utilizing InfiniBand or RoCE.

• Proven ability to troubleshoot networking issues across Linux hosts, NICs, switches, and physical connections.

• Proficiency in automating network operations using Python, Ansible, or similar tools.

• Familiarity with leaf-spine architectures, BGP, ECMP, VLANs, and network segmentation.

• Understanding of RDMA concepts, performance tuning, congestion control, and lossless Ethernet considerations.

• Knowledge of Linux networking, including NIC drivers and firmware, packet capture, and throughput/latency testing.

• Expertise in optics, transceivers, cable management, and link-level diagnostics.

• Awareness of safe change management, configuration versioning, telemetry, and alerting processes.

• Experience managing 400G/800G networks or large multi-rack GPU clusters.

• Familiarity with NVIDIA Spectrum or Quantum networking.

• Experience with NCCL performance analysis and troubleshooting distributed training.

• Knowledge of EVPN/VXLAN, SONiC, or network source-of-truth systems.

• Background in network simulation, automated validation, and capacity planning.

🏝️ Benefits

• Equity incentives.

• Remote work option.