HPC Infrastructure Engineer - GPU Clusters
ElevenLabs United Kingdom
Owns and operates NVIDIA GPU clusters across bare metal and rented capacity, handling provisioning, scheduling, monitoring and capacity planning. Builds automation for node health checks, draining, remediation and burn-in, and manages the stack beneath training code including OS images, CUDA, NCCL, InfiniBand/RoCE and Slurm. Also benchmarks rented GPU capacity and does hands-on datacenter hardware work.