Senior Gpu Networking Architect

NVIDIA

Poland
Base: 292,500 pln - 650,000 pln (level 4/5); bonus...
5+ years hands-on cuda programming
Deep knowledge of gpu architecture fundamentals
Systems-level c/c++ development experience
This role involves building and optimizing GPU communication kernels that underpin collective and point-to-point operations in large-scale AI systems

Job Summary

  • This role involves building and optimizing GPU communication kernels that underpin collective and point-to-point operations in large-scale AI systems.
  • Candidates will collaborate with network software, hardware, and AI framework teams to co-design communication strategies aligned with GPU execution patterns.
  • NVIDIA offers highly competitive salaries and a comprehensive benefits package for this position.

Matching Summary

Match Score: 85

This role involves building and optimizing GPU communication kernels that underpin collective and point-to-point operations in large-scale AI systems.

Salary

Base: 292,500 PLN - 650,000 PLN (Level 4/5); Bonus/Equity: Not specified; Benefits: Comprehensive package including www.nvidiabenefits.com

Skills & Requirements

Must-have

  • 5+ years hands-on CUDA programming
  • Deep knowledge of GPU architecture fundamentals
  • Systems-level C/C++ development experience
  • Proficiency in GPU data movement mechanisms
  • Ability to read and optimize performance profiles

Nice-to-have

  • Experience with NCCL or NVSHMEM libraries
  • Understanding of distributed deep learning parallelism
  • Background in RDMA and InfiniBand networking
  • Experience with kernel pipelining techniques
  • Optimization of LLM training workloads

Key Requirements

  • M.Sc. or equivalent in Computer Science or Engineering
  • Strong understanding of warp scheduling and memory hierarchy
  • Experience with Nsight Compute and Nsight Systems profiling tools

Work Rights

Not specified

Tailored Resume

Cover Letter