We launched recruiters.rip, let your AI apply for you. First month free with code onemonthfree

Senior HPC Platform Engineer (HPC / GPU Infrastructure)

Taiga Cloud is Northern Data Group’s AI-cloud division, providing on-demand NVIDIA GPU and bare-metal compute plus managed Kubernetes and SLURM services.

0 current maintainers0 active leads
Distributed

Funding history

About Taiga Cloud

Taiga Cloud operates AI infrastructure for model training, prototyping, and scaling, including self-service GPU compute, bare metal, managed Kubernetes, SLURM, and NVIDIA AI Enterprise-related tooling. The 2024 annual report identifies Taiga Cloud Ltd. (Ireland) as the group company providing cloud services to external customers worldwide.

View jobs by Taiga Cloud

Skills

Candidate Availability

About the Role

You will build, run and harden the bare-metal GPU platform behind an AI cloud. This is a hands-on Linux systems role at the intersection of HPC, GPU virtualisation and high-performance fabrics, where you own the servers, OS, GPUs, DPUs and InfiniBand fabric. You will tune GPU hosts, operate BlueField DPUs and InfiniBand with OpenSM and UFM, deliver GPU virtualisation with KubeVirt or libvirt/QEMU, automate provisioning with PXE, cloud-init, Redfish and Ansible, write Bash and Python tooling, and lead incident response across European datacentres.

Requirements

  • Typically 7+ years in Linux systems or infrastructure engineering, with recent hands-on GPU or HPC platform experience
  • Deep knowledge of Linux OS and kernel, Debian/Ubuntu in production
  • Strong understanding of NVIDIA GPUs: CUDA, Fabric Manager, NVLink/NVSwitch, MIG, DCGM, GPUDirect RDMA
  • GPU virtualisation experience: passthrough, SR-IOV, vGPU, KVM/QEMU/libvirt or KubeVirt
  • Extensive hands-on experience with NVIDIA BlueField-2/3 in DPU mode (must have)
  • InfiniBand operation with OpenSM and UFM (must have)
  • Strong IPv6/IPv4 networking fundamentals, DHCP, DNS, NTP, BGP
  • Storage networking knowledge: NVMe-oF, RoCE, NFS
  • Excellent Bash and Python
  • Working knowledge of Ansible
  • BMC, Redfish, IPMI, PXE, preseed, cloud-init
  • Docker/containerd and Kubernetes including Operators, CNIs and CSIs
  • Proven ability to lead production incidents to root cause
  • Clear communicator in English

Responsibilities

  • Build, tune and maintain Linux bare-metal GPU hosts for HPC and AI workloads, from BIOS/BMC through kernel, drivers and firmware
  • Deploy, configure and operate BlueField-2/3 DPUs in DPU mode, including BFB images, DOCA services, OVS offload and network/storage offload
  • Operate and troubleshoot InfiniBand fabrics with OpenSM and UFM, including partitions, routing, health monitoring and NCCL/RDMA performance analysis
  • Deliver GPU virtualisation: PCIe passthrough, SR-IOV Virtual Functions and GPU VM integration into Kubernetes via KubeVirt or libvirt/QEMU
  • Build and support physical and virtual GPU clusters interconnected over InfiniBand
  • Run IPv6-first host networking and support storage networks carrying NVMe-oF, RoCE, NFS, block and object traffic
  • Automate provisioning and lifecycle with PXE, preseed, cloud-init, Redfish/IPMI and Ansible
  • Write production-quality Bash and Python tooling for deployment, validation, health checks and fleet operations
  • Lead incident response across multiple European datacentres, perform root-cause analysis and drive permanent fixes
  • Work with networking, storage, Kubernetes and customer-facing teams to bring new hardware generations into production

Benefits

  • Work flexibly from home with an international virtual team
  • Hardware wishes fulfilled
  • Regular wellbeing initiatives