Senior GPU Networking Architect: Optimize AI Communication

NVIDIA; Zürich; On-site; CHF 180,000 - 260,000; Full time

Job summary

NVIDIA is seeking a Senior GPU Networking Architect to join our networking software group in Switzerland. You will craft, optimize, and validate GPU communication kernels that support large-scale AI systems, working with GPU-aware primitives and device-side APIs.

You will apply deep CUDA and GPU architecture knowledge to reduce latency, improve throughput, and coordinate with cross-functional teams to align networking strategies with GPU execution patterns.

Qualifications

• 5+ years of hands-on CUDA programming and GPU kernel optimization.
• M.S. or equivalent in computer science, computer engineering, or related field.
• Strong understanding of GPU architecture: warp scheduling, shared memory, L2 cache, memory coalescing, occupancy, async execution.
• Experience with systems-level C/C++ in performance-critical environments.
• Familiarity with GPU data movement mechanisms such as GPUDirect RDMA and GPU-initiated communication.
• Ability to read and interpret GPU performance profiles (Nsight Compute/Sys) and translate observations into optimizations.
• Strong collaboration skills in a multi-national, interdisciplinary environment.

Responsibilities

• Build, implement, and optimize GPU communication kernels for collective and point-to-point operations in large-scale AI systems.
• Leverage GPU architecture knowledge to improve kernel efficiency, reduce latency, and overlap compute with communication.
• Develop GPU-resident communication primitives and device-side APIs for data movement across nodes and accelerators.
• Profile and tune GPU kernels end-to-end, identifying bottlenecks at compute/memory/network boundaries.
• Collaborate with network software, hardware, and AI framework teams to align communication with GPU execution patterns.
• Build proofs-of-concept, run experiments, and model new strategies before production adoption.
• Contribute to GPU-aware programming model evolution for application developers.

Skills

CUDA programming

GPU architecture

C/C++ development

Performance profiling

GPUDirect RDMA

Parallel computing

Collaboration

Education

M.S. or equivalent in CS/CE

Tools

Nsight Compute

Nsight Systems

NCCL / NVSHMEM

PyTorch / TensorRT-LLM

Job description

NVIDIA is seeking a Senior GPU Networking Architect to join our networking software group in Switzerland. You will craft, optimize, and validate GPU communication kernels that support large-scale AI systems, working with GPU-aware primitives and device-side APIs.

You will apply deep CUDA and GPU architecture knowledge to reduce latency, improve throughput, and coordinate with cross-functional teams to align networking strategies with GPU execution patterns.