HPC Systems Administrator (Europe)

HPC Systems Administrator / Systems DevOps Engineer Alliance Services Plus - GECI International | Permanent Contract | France & Benelux About Alliance Services Plus A specialist in High Performance Computing (HPC) and Artificial Intelligence for over 15 years, Alliance Services Plus operates on the largest national and industrial HPC infrastructures in France and Europe (CEA, CNRS, TotalEnergies, SAFRAN). The Role You will be responsible for deploying, configuring, and managing HPC infrastructures (on-premise, cloud, hybrid). You will work on critical systems supporting large-scale scientific and industrial computing. Main Responsibilities Deployment and administration of HPC clusters (CPU/GPU) Configuration of computing environments (MPI, OpenMP, CUDA, OpenACC) Management of on-premise, cloud (AWS, Azure, GCP), and hybrid infrastructures Automation and DevOps: CI/CD, containerization (Docker, Singularity), orchestration 24/7 monitoring and supervision of HPC systems Storage management and parallel file systems (Lustre, GPFS) System and network performance optimization (InfiniBand, high-performance Ethernet) Level 2/3 technical support for teams and users Technical documentation and knowledge transfer Required Profile Education: Master's degree (Bac+5) in Computer Science, Systems Engineering, or equivalent Required Technical Skills: Linux system administration (RHEL, CentOS, Ubuntu) in HPC environments Job schedulers (Slurm, PBS Pro, LSF) Scripting: Bash, Python, Ansible Virtualization and containerization technologies High-performance networking (InfiniBand, RoCE) Assets: Experience with GPU architectures (NVIDIA, AMD) Knowledge of cloud HPC platforms Experience in critical production environments Certifications (RHCE, CKA, AWS/Azure) Personal Qualities: Rigor, autonomy, teamwork, ability to work on critical environments Conditions Permanent contract based in France (Paris region or client sites) or Benelux Salary according to profile and experience Varied assignments on leading infrastructures Continuous training and technology watch