Senior Network Engineer AI/HPC Fabric

Kanshe Infotech Inc.; Cary (NC); On-site; USD 120,000 - 180,000; Full time

Job summary

Kanshe Infotech Inc. is seeking a Senior Network Engineer to design, validate, and optimize high-performance fabrics for large-scale AI/GPU and HPC clusters using NVIDIA Spectrum-X, Quantum InfiniBand, BlueField-3, RoCEv2, and RDMA.

The role emphasizes architecture development, runbook creation, and performance benchmarking, along with troubleshooting latency and congestion and supporting deployment of BlueField-3 networks. Strong documentation and collaboration across teams are essential.

Qualifications

• Bachelor’s degree in CS/EE or related field is required.
• 10-12 years of network engineering experience in AI/HPC environments.
• Strong expertise with Spectrum-X, InfiniBand, RoCEv2, RDMA and BlueField-3.

Responsibilities

• Design and validate AI/HPC network fabrics at scale.
• Develop reference architectures, runbooks, and performance baselines.
• Troubleshoot latency, congestion, RDMA, and network performance issues.
• Conduct network benchmarking, validation, and capacity testing.
• Support BlueField-3 and high-speed networking deployments.
• Collaborate with field, delivery, architecture, and customer teams.

Skills

NVIDIA Spectrum-X

Quantum InfiniBand

BlueField-3 / DPU

RoCEv2 / RDMA

AI/HPC and GPU cluster networking

High-speed Ethernet/InfiniBand

Network troubleshooting, optimization

Reference architecture

AI/HPC fabric experience

Education

Bachelor’s degree in Computer Science, Electrical Engineering, Networking, or related field

Job description

Senior Network Engineer - AI/HPC Fabric

Location: Cary, NC ( Local highly preferred)

Experience: 10-12 Years

Duration: Long term

Job Summary:

Seeking a Senior Network Engineer to design, validate, and optimize high-performance fabrics for large-scale AI/GPU and HPC clusters using NVIDIA Spectrum-X, Quantum InfiniBand, BlueField-3, RoCEv2, and RDMA.

Job Summary:

Seeking a Senior Network Engineer to design, validate, and optimize high-performance fabrics for large-scale AI/GPU and HPC clusters using NVIDIA Spectrum-X, Quantum InfiniBand, BlueField-3, RoCEv2, and RDMA.

Mandatory Skills:

• NVIDIA Spectrum-X & Quantum InfiniBand
• BlueField-3 / DPU
• RoCEv2 & RDMA
• AI/HPC and GPU cluster networking
• High-speed Ethernet/InfiniBand
• Network troubleshooting, optimization & benchmarking
• Reference architecture and runbook development
• 3-5 years hands-onAI/HPC fabric experience

Key Responsibilities:

• Design and validate AI/HPC network fabrics at scale.
• Develop reference architectures, runbooks, and performance baselines.
• Troubleshoot latency, congestion, RDMA, and network performance issues.
• Conduct network benchmarking, validation, and capacity testing.
• Support BlueField-3 and high-speed networking deployments.
• Collaborate with field, delivery, architecture, and customer teams.

Required Qualifications:

• Bachelor’s degree in Computer Science, Electrical Engineering, Networking, or related field.
• 10-12 years network engineering experience.
• Strong expertise in Spectrum-X, InfiniBand, RoCEv2, RDMA, and BlueField-3.
• Experience with large-scale AI/HPC/GPU environments.
• Strong communication and technical documentation skills.
• Relevant NVIDIA/Cisco/Juniper certifications preferred.