GPU Platform Engineer

About the job Your mission You will enable the expansion of Lyceum’s bare-metal and managed service GPU offering, taking our EU-sovereign cloud offering to the next step. This exciting role includes both a technical customer-facing function and the opportunity to develop and contribute to a novel software stack – changing the way GPUs are being used today. Your focus: This role has both a customer and a developer component. On the customer side, you’ll be responsible for technical customer requests, both during the onboarding process and later in the lifecycle. On the developer side, you’ll help to build out and improve our managed services such as Slurm or Kubernetes. Your KPIs: Response time / error resolving Customer satisfaction Develop and improve software offering on top of bare metal offering Your profile We consider candidates from diverse backgrounds, with a deep love for technical challenges and the desire to take on ownership beyond what’s reasonably expected. What we’re looking for: 3+ years of experience working with GPU infrastructure (bare metal, VM, cloud) Experience with Kubernetes and/or Slurm cluster deployment and management Hands-on experience with the NVIDIA software stack (drivers, CUDA, DCGM, GPU) Operator Familiarity with high-performance networking (InfiniBand, NVLink, RoCE) Strong Linux systems administration skills Solid networking fundamentals (IP addressing, firewalls, VPNs, load balancing) Familiarity with IaC tools (Ansible, Terraform, or similar) Strong written and verbal communication in English Nice to have: Experience with confidential computing technologies (Intel TDX, AMD SEV-SNP) Experience with NVIDIA DGX/HGX systems or multi-node GPU clusters Knowledge of MIG (Multi-Instance GPU) partitioning and GPU virtualisation Background in HPC or AI/ML workloads Experience in a customer-facing technical role – you can explain complex infrastructure clearly and handle urgent requests professionally