Senior GPU Networking Architect: AI-Scale Kernel Innovator

NVIDIA; United Kingdom; On-site; GBP 90,000 - 140,000; Full time

Benefits offered by this job

Competitive salary

Comprehensive benefits package

Job summary

NVIDIA is seeking a Senior GPU Networking Architect to advance GPU communication kernels and integration with AI frameworks. You will meld GPU computing with networking, optimizing kernels for latency, throughput, and scalability across large AI systems.

The role emphasizes in‑depth CUDA work, GPU architecture mastery, and collaboration with cross‑functional teams to push the boundaries of GPU‑driven networking.

Qualifications

• 5+ years of hands‑on CUDA programming.
• M.Sc. or equivalent in CS/CE or related field.
• Strong understanding of GPU architecture fundamentals.
• Experience with systems‑level C/C++ development.
• Familiarity with GPU data movement mechanisms (GPUDirect RDMA).
• Ability to read and reason about GPU performance profiles (Nsight).
• Strong collaboration skills in a multi‑national environment.

Responsibilities

• Build, implement, and optimize GPU communication kernels for large-scale AI systems.
• Leverage GPU architecture knowledge to improve kernel efficiency and reduce latency.
• Develop GPU-resident communication primitives and device-side APIs.
• Profile and tune GPU kernels end‑to‑end to identify bottlenecks.
• Collaborate with network software, hardware, and AI teams to align strategies.
• Build proofs‑of‑concept, run experiments, and model results.
• Contribute to GPU-aware programming models for developers.

Skills

CUDA programming

GPU architecture

C/C++ development

Nsight profiling

collaboration

Education

M.Sc. or equivalent

Tools

Nsight Compute

Nsight Systems

GPUDirect RDMA

Job description

NVIDIA is seeking a Senior GPU Networking Architect to advance GPU communication kernels and integration with AI frameworks. You will meld GPU computing with networking, optimizing kernels for latency, throughput, and scalability across large AI systems.

The role emphasizes in‑depth CUDA work, GPU architecture mastery, and collaboration with cross‑functional teams to push the boundaries of GPU‑driven networking.