HPC Performance Engineer — Large-Scale GPU Clusters

NVIDIA AI; Val-de-Travers; Vor Ort; CHF 120.000 - 180.000; Vollzeit

Zusammenfassung

NVIDIA AI in Switzerland seeks a performance engineer to drive optimization of HPC communication libraries and large-scale GPU workloads. You will characterize performance on multi-GPU clusters, study hardware-software interactions, and guide roadmap choices.

You will implement C/C++ micro-benchmarks, debug across HW/SW, and script with Python. Containers, Kubernetes, and SLURM enable scalable testing in a collaborative, global team.

Qualifikationen

• MS or PhD in Computer Science or related field with performance engineering and HPC experience.
• 3+ years of experience with parallel programming and a communication runtime (MPI, NCCL, UCX, NVSHMEM).
• Experience benchmarking and triaging performance on large HPC clusters.
• Strong understanding of computer system architecture and OS principles.
• Proficiency in C/C++ micro-benchmarks and code modification when required.
• Proficient in Python scripting and debugging performance issues across HW/SW.
• Familiar with containers and scheduling tools (Kubernetes, SLURM, Docker, Ansible).
• Willingness to learn new areas and collaborate across time zones.

Aufgaben

• Conduct in-depth performance characterization on large multi-GPU and multi-node clusters.
• Study interactions of libraries with hardware and software components in the stack.
• Evaluate concepts and perform trade-offs when multiple solutions exist.
• Triage and root-cause performance issues reported by users.
• Collect and analyze performance data; build visualization tools.
• Collaborate with a dynamic team across multiple time zones.

Kenntnisse

Parallel programming

MPI

NCCL

UCX

NVSHMEM

C/C++

Python scripting

Docker

Kubernetes

SLURM

Ansible

Ausbildung

MS or PhD in CS or related field

Tools

MPI

NCCL

UCX

NVSHMEM

C/C++

Python

Docker

Kubernetes

SLURM

Ansible

Jobbeschreibung

NVIDIA AI in Switzerland seeks a performance engineer to drive optimization of HPC communication libraries and large-scale GPU workloads. You will characterize performance on multi-GPU clusters, study hardware-software interactions, and guide roadmap choices.

You will implement C/C++ micro-benchmarks, debug across HW/SW, and script with Python. Containers, Kubernetes, and SLURM enable scalable testing in a collaborative, global team.