Senior Software Engineer, Fabric Networking - GPU

NVIDIA NVIDIA · Semiconductors · Bangalore, India +5 · Remote

Senior Software Engineer for GPU Fabric Networking team at NVIDIA, focusing on high-performance communication software for deep learning and HPC workloads. Responsibilities include designing, developing, and maintaining system-level software for GPU-to-GPU communication, collaborating with hardware/software teams, and contributing to scalable fabric architecture.

What you'd actually do

  1. Design, develop, and maintain system-level software to support GPU-to-GPU communication.
  2. Collaborate with cross-functional hardware and software teams to build next-generation networking solutions.
  3. Contribute to scalable and reliable GPU fabric architecture for large compute clusters.
  4. Align software development with customer needs and real-world deployment environments.

Skills

Required

  • C/C++
  • debugging
  • system-level problem-solving
  • software that interacts with hardware and device drivers
  • system architecture
  • operating systems
  • kernel internals
  • multi-threaded systems development
  • distributed systems development
  • Linux development
  • networking fundamentals
  • high-performance interconnects
  • OS virtualization technologies

Nice to have

  • CUDA programming model
  • NVIDIA GPU architecture
  • memory consistency and coherence models
  • static/dynamic code analysis
  • fuzz testing
  • fault injection techniques
  • Windows development

What the JD emphasized

  • deep learning
  • HPC
  • GPU-to-GPU communication