Principal Network Engineer

About the role

TensorWave is looking for a Principal Network Engineer (L7) to lead the design, implementation, and operation of large‑scale RoCEv2 data‑center networks that power next‑generation AI and ML workloads. You will collaborate with the network architect and infrastructure leadership to ensure the network can support thousands to over 100,000 GPUs while maintaining performance, reliability, and operational sanity.

Key responsibilities

• Own end‑to‑end network architecture and make high‑impact design decisions for AI/ML clusters.
• Define, evolve, and standardize RoCEv2 fabrics, including congestion‑management strategies (PFC, ECN, DCQCN).
• Establish automation, validation, and observability patterns to prevent misconfiguration and reduce manual work.
• Act as the technical escalation point for complex failures, scaling limits, and architectural trade‑offs in always‑on, multi‑tenant environments.
• Stay hands‑on with high‑speed optics, switching, routing, and cabling in production clusters.

Required profile

• Bachelor’s degree in Computer Science, Electrical Engineering, or related field, or equivalent practical experience.
• Deep experience designing and operating RDMA/RoCEv2 networks at large scale for AI or HPC workloads.
• Proven ability to make hard technical calls with incomplete data and take responsibility for systems under sustained load.

Required skills

• Expert knowledge of Arista, Juniper, and custom SONiC‑based switching platforms.
• Hands‑on experience with congestion management (PFC, ECN, DCQCN) and performance tuning.
• High‑speed optics and cabling expertise (400G, 800G, AEC, AOC, DAC, structured cabling).
• Automation skills using Python, Ansible, Terraform, Git, and production observability tooling.

What we offer

• Stock options.
• 100% paid medical, dental, and vision benefits.