Senior HPC Platform Engineer — Cloud-Native Compute (Paris)

Arlequin AI; Paris; Hybride; EUR 90 000 - 130 000; Plein temps

Résumé du poste

Arlequin AI is hiring an experienced HPC Engineer to design and operate a cloud-native, sovereign compute platform that powers our AI products and research. You will work on GPU/CPU clusters, orchestration with Kubernetes, and end-to-end performance optimization in a hybrid/remote/onsite Paris setup.

The role focuses on scalable infrastructure, IaC, and platform engineering to enable researchers and engineers to ship experiments efficiently while maintaining security and reliability.

Qualifications

• 5+ years of experience in infrastructure / HPC / compute platforms.
• Advanced proficiency with Kubernetes in production.
• Hands-on experience with GPU workloads and batch schedulers.
• Experience operating production services with latency and availability constraints.
• Strong IaC experience (Terraform / OpenTofu, Terragrunt).
• Understanding of distributed computing and large-scale batch processing.

Responsabilités

• Design, deploy, and maintain infrastructure on Scaleway and automate with IaC (OpenTofu, Terragrunt).
• Operate high-performance storage and I/O for training and inference.
• Design, deploy, and operate a Kubernetes compute cluster for heterogeneous workloads; manage scheduling and resources.
• Identify bottlenecks across compute, network, and storage; run model serving and large-scale pipelines.
• Monitor performance, latency, and costs; implement alerts and right-size resources.
• Collaborate with research/product to evolve the platform roadmap and provide self-service abstractions.

Connaissances

Kubernetes

GPU workloads

IaC

Terragrunt

OpenTofu

Terraform

Multi-node compute

Outils

NVIDIA GPU Operator

CUDA

DCGM

SGLang

Kueue

NCCL

MPI

Description du poste

Arlequin AI is hiring an experienced HPC Engineer to design and operate a cloud-native, sovereign compute platform that powers our AI products and research. You will work on GPU/CPU clusters, orchestration with Kubernetes, and end-to-end performance optimization in a hybrid/remote/onsite Paris setup.

The role focuses on scalable infrastructure, IaC, and platform engineering to enable researchers and engineers to ship experiments efficiently while maintaining security and reliability.