Kubernetes Scale Systems Engineer for AI Infra

NVIDIA; France; On-site; EUR 90,000 - 140,000; Full time

Job summary

NVIDIA DGX Cloud is seeking a Systems Software Engineer to advance distributed AI infrastructure. You will own performance and scale characterization across the software stack, from Kubernetes control planes to GPU components, and collaborate with researchers and customers on automated tests and real workloads.

You will diagnose complex issues, build monitoring tools, and contribute to a high-velocity CI/CD framework while engaging with upstream CNCF/NVIDIA projects to shape design decisions.

Qualifications

• 2+ years of experience in computer architecture, networking, storage or accelerators.
• Expertise with Kubernetes and related CNCF projects.
• Background in large-scale parallel and distributed accelerator systems.
• Experience optimizing performance for AI workloads on big-scale systems.
• Proficiency in Golang and Python.
• Familiarity with NVIDIA software stack for training and inference.

Responsibilities

• Drive end-to-end performance and scale characterization for the DGX Cloud software stack.
• Collaborate with researchers and customers to develop automated tests simulating real workloads.
• Investigate and resolve root causes in complex distributed systems.
• Design monitoring and analysis tools for performance testing across GPU/CPU resources.
• Triage and debug Kubernetes clusters at ultra-large scale for reliability and efficiency.
• Support CI/CD pipelines enabling continuous performance testing.

Skills

Kubernetes

Golang

Python

Distributed systems

Computer Architecture

Performance modeling

Networking

Cloud platforms

Education

Bachelor's degree in Engineering

Master's preferred

Job description

NVIDIA DGX Cloud is seeking a Systems Software Engineer to advance distributed AI infrastructure. You will own performance and scale characterization across the software stack, from Kubernetes control planes to GPU components, and collaborate with researchers and customers on automated tests and real workloads.

You will diagnose complex issues, build monitoring tools, and contribute to a high-velocity CI/CD framework while engaging with upstream CNCF/NVIDIA projects to shape design decisions.