TensorWave
Principal Network Engineer
Remote · Remote · Principal · Full-time · Posted Jul 31, 2026
- Education
- Bachelor’s degree in Computer Science, Electrical Engineering, or a related technical field, or equivalent practical experience.
- Visa sponsorship
- Employment offers are contingent upon verification of identity and authorization to work in the United States, as required by law.
About the job
Own and evolve large-scale RoCEv2 data center network architecture for AI and ML infrastructure, supporting clusters scaling from thousands to over 100,000 GPUs.
What you'll do
- Set network architecture and technical direction, and standardize RoCEv2 networks for AI/ML clusters.
- Define and validate RDMA congestion management strategies, including PFC, ECN, and DCQCN.
- Establish automation, validation, and observability practices; serve as the technical escalation point for complex failures and design tradeoffs.
What they ask for
- Deep experience designing and operating RDMA and RoCEv2 networks in large-scale production AI or HPC environments.
- Expert knowledge of switching hardware and NOS, including high-speed Ethernet fabrics.
- Hands-on congestion management and performance tuning with PFC, ECN, and DCQCN.
- Experience with high-speed optics and cabling, and automation tools including Python, Ansible, Terraform, and Git.
Tools
RoCEv2, RDMA, PFC, ECN, DCQCN, Python, Ansible, Terraform
AI focus
Design and operate high-scale networks that power AI and ML infrastructure and workloads.
Model-written summary · the original posting is authoritative