HPC-insinööri

Verda is building a full-stack AI cloud spanning data centers, hardware, and our cloud platform. Headquartered in Helsinki with offices in London and San Francisco, Verda is growing rapidly and offers a chance to impact the world through infrastructure. Role: HPC Engineer will own the baremetal and virtualized clusters behind the AI cloud - from InfiniBand fabric and shared filesystems up through the workload orchestration layer. Responsibilities include Administer baremetal GPU/HPC clusters end to end; Administer virtualized clusters; Design, deploy, and tune InfiniBand fabrics; Deploy and operate shared/parallel filesystems; Troubleshoot and resolve issues across the full fabric stack; Partner with remote-hands and data center teams; Operate and tune Slurm workloads; Keep IPAM/DCIM records; On-call rotations; Collaborate with platform, network, and storage teams. Required competencies include Linux, InfiniBand, IB clustering, shared filesystems, NCCL/CUDA/DOCA, Slurm, and scripting. Nice to have RoCEv2, GPU diagnostics, MAAS/Foreman, GPU virtualization. Next steps: apply via Careers page.