GPU Compute Infrastructure Engineer – Paris (Hybrid)

H Company is seeking an Infrastructure Engineer to own the GPU compute powering our model training, evaluation and serving systems. You’ll run Kubernetes clusters, manage GPU nodes, and collaborate with research, product and platform teams in a hybrid Paris/London setup.

You’ll design scalable GPU compute, debug OS and driver issues, and optimize scheduling and capacity with cloud providers, while delivering reliable performance across training and inference workloads.