AI/HPC Network Engineer

Circle Of Care; Faridabad District; On-site; INR 4,000,000 - 7,500,000; Full time

Job summary

Circle Of Care in India seeks an AI/HPC networking expert to design and deploy GPU/AI cluster network fabrics. You will work on InfiniBand, RoCE, NVLink, NVSwitch, ConnectX and related NVIDIA networking technologies to optimize performance across large-scale AI systems.

Ideal candidates have hands-on experience with high-speed Ethernet and fabric management, and a track record of troubleshooting GPU/AI cluster networks in data centers or hyperscale environments.

Qualifications

• Hands-on experience with InfiniBand and high-speed interconnects across GPU/AI clusters.
• Proven background in RDMA/RoCE for data center networking in AI/HPC contexts.
• Direct exposure to NVIDIA networking products (ConnectX, NVLink/NVSwitch) and related fabric tech.
• Experience deploying and troubleshooting GPU fabric connectivity in large-scale environments.
• Familiarity with fabric management tools and performance optimization for AI workloads.

Responsibilities

• Design and deploy AI/HPC network fabrics for GPU clusters.
• Configure, troubleshoot, and optimize high-speed interconnects (InfiniBand, NVLink/NVSwitch).
• Coordinate with data-center teams to ensure scalable, reliable GPU networking for AI workloads.

Skills

InfiniBand

RDMA RoCE

NVIDIA Networking

ConnectX

NVIDIA Quantum

NVLink

NVSwitch

GPU fabric connectivity

UFM fabric management

High-speed Ethernet

AI/HPC network design

Fabric troubleshooting

Networking in AI clusters

Job description

Looking for candidates with hands-on experience in:

• InfiniBand


• RDMA / RoCE


• NVIDIA Networking


• ConnectX


• NVIDIA Quantum / Spectrum


• NVLink / NVLink Switch / NVSwitch


• GPU fabric connectivity


• UFM / fabric management


• High-speed Ethernet


• AI/HPC network fabric design and deployment


• Fabric configuration, troubleshooting and optimization


• Networking within large-scale GPU/AI clusters

Important: This is an AI/HPC networking role, not a conventional enterprise networking role. Candidates should have actual GPU/HPC/AI cluster networking experience.

Candidate Profile

Candidates should preferably have 8+ years of relevant experience and come from backgrounds such as:

NVIDIA | AI Infrastructure | GPU Infrastructure | HPC | AI Servers | Data Centers | Semiconductor | Server OEM/ODM | Hyperscalers | GPU Cloud | AI Hardware

Candidates with direct experience in NVIDIA Blackwell / GB200 / GB300 / B200 / B300 / HGX / NVL72 will be given priority.