NVIDIA Network Engineer (InfiniBand / Ethernet)

Job Summary:

We are looking for an experienced NVIDIA Network Engineer to build, manage, and optimize high-performance networks supporting large-scale NVIDIA GPU clusters and AI/ML workloads.

Key Responsibilities:

• 
  

  Design and manage InfiniBand and Ethernet networks for GPU clusters.

  
  


• 
  

  Configure and optimize RoCE v2, network performance, and connectivity.

  
  


• 
  

  Troubleshoot network congestion, latency, and critical network issues.

  
  


• 
  

  Monitor network performance and ensure high availability.

  
  


• 
  

  Automate network configurations using Ansible/Terraform.

  
  

Required Skills:

• 
  

  Strong experience in Data Centre Networking.

  
  


• 
  

  Deep expertise in InfiniBand (HDR/NDR) and RoCE.

  
  


• 
  

  Knowledge of GPU cluster networking and NCCL communication.

  
  


• 
  

  Experience in network performance tuning and troubleshooting.

  
  


• 
  

  Familiarity with Ethernet, BGP/EVPN, and network automation.

  
  

Qualifications:

• 
  

  BE/B.Tech in Computer Science, Electronics & Communication, or equivalent.

  
  


• 
  

  7–15 years of relevant Data Centre Networking experience.