Senior AI Inference Systems Engineer

NVIDIA; Italia; In loco; EUR 120.000 - 180.000; Tempo pieno

Descrizione del lavoro

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

You’ll drive industry benchmarks, contribute to vLLM, SGLang, and related tooling, and publish research that advances ML systems.

Competenze

• Bachelor's degree in CS/CE/SE or equivalent with 7+ years of experience; or Master's with 5+ years; or PhD with top-tier publications.
• Strong programming skills in Python and C/C++, Go or Rust a plus; solid CS fundamentals.
• Knowledgeable about performance engineering in ML frameworks (PyTorch) and inference engines (vLLM, SGLang).
• Familiar with GPU programming: CUDA, memory hierarchy, streams, NCCL; profiling/debug tools Nsight Systems/Compute.
• Experience with containers and orchestration (Docker, Kubernetes, Slurm); Linux namespaces and cgroups.

Mansioni

• Contribute features to vLLM and optimize inference framework with modern GPU features and parallelism.
• Develop, optimize, and benchmark GPU kernels using fusion, autotuning, and memory/layout optimization.
• Define and build inference benchmarking methodologies; contribute to MLPerf Inference submissions.
• Architect scheduling and orchestration of containerized large-scale inference deployments on GPU clusters across clouds.
• Publish original research and translate ideas into NVIDIA software products.

Conoscenze

Python

C/C++

Go or Rust

Algorithms & data structures

Operating systems

Parallel programming

Distributed systems

Deep learning theories

Performance engineering

Python ML frameworks

Formazione

Bachelor's degree in CS/CE/SE

Master's degree in CS/CE/SE

PhD in ML Systems or related fields

Strumenti

CUDA

Nsight Systems/Compute

Docker

Kubernetes

Slurm

Descrizione del lavoro

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

You’ll drive industry benchmarks, contribute to vLLM, SGLang, and related tooling, and publish research that advances ML systems.