GPU Compute Infrastructure Engineer (Kubernetes)

H Company; Paris; Hybrid; EUR 90,000 - 150,000; Full time

Job summary

H Company is seeking a hands-on Infrastructure/SRE expert to own the GPU compute stack powering model training, evaluation and serving. You’ll operate Kubernetes at scale, optimize GPU scheduling, and ensure high availability across a globally distributed environment.

You’ll collaborate with research, product and platform teams while coordinating with cloud providers. The role is hybrid, with three days in Paris or London and occasional travel between the two offices.

Qualifications

• Experience running GPU workloads on Kubernetes.
• Ability to design and operate GPU compute clusters for training and inference.
• Strong Python scripting and automation skills.
• Fluent in Linux system administration and troubleshooting.

Responsibilities

• Run Kubernetes clusters for GPU training and inference.
• Improve GPU scheduling and placement strategies.
• Detect, drain and repair unhealthy GPU nodes.
• Manage capacity with cloud providers and multi-cloud tools.
• Debug GPU/Linux/storage/networking failures.

Skills

Kubernetes

GPU workloads on Kubernetes

Python

Linux OS

Tools

Terraform

Kueue

Karpenter

SkyPilot

AWS EKS

SageMaker HyperPod

Job description

H Company is seeking a hands-on Infrastructure/SRE expert to own the GPU compute stack powering model training, evaluation and serving. You’ll operate Kubernetes at scale, optimize GPU scheduling, and ensure high availability across a globally distributed environment.

You’ll collaborate with research, product and platform teams while coordinating with cloud providers. The role is hybrid, with three days in Paris or London and occasional travel between the two offices.