Senior AI Inference Systems Engineer - GPU HPC

NVIDIA; Schweiz; Vor Ort; CHF 140.000 - 210.000; Vollzeit

Zusammenfassung

NVIDIA is seeking highly skilled software engineers to build AI inference systems that scale efficiency across multi-GPU and multi-cloud environments. You will architect inference stacks, optimize GPU kernels and compilers, and collaborate across inference, compiler, scheduling, and performance teams.

You will contribute features to vLLM, benchmark HPC-style workloads, and push the frontier of accelerated computing for AI, with opportunities to publish research and influence NVIDIA products.

Qualifikationen

• Bachelor's degree (or equivalent) in CS/CE/SE with 7+ years experience; or Master's with 5+ years; or PhD with publications in ML Systems/GPU.
• Strong programming skills in Python and C/C++; Go or Rust a plus; solid CS fundamentals.

Aufgaben

• Contribute features to vLLM and optimize inference framework.
• Develop and benchmark GPU kernels with fusion and autotuning.
• Define benchmarking methodologies and participate in MLPerf Inference submissions.
• Architect scheduling and orchestration for containerized large-scale deployments on GPUs.
• Publish original research and integrate ideas into NVIDIA products.

Kenntnisse

Python

C/C++

Go

Rust

Parallel programming

GPU programming

Performance optimization

DS/ML frameworks

Ausbildung

Bachelor's degree in CS/CE/SE

Master's degree in CS/CE/SE

PhD in ML Systems / GPU architecture

Tools

Nsight Systems/Compute

Docker/Kubernetes

Slurm

MLIR/LLVM

Jobbeschreibung

NVIDIA is seeking highly skilled software engineers to build AI inference systems that scale efficiency across multi-GPU and multi-cloud environments. You will architect inference stacks, optimize GPU kernels and compilers, and collaborate across inference, compiler, scheduling, and performance teams.

You will contribute features to vLLM, benchmark HPC-style workloads, and push the frontier of accelerated computing for AI, with opportunities to publish research and influence NVIDIA products.