HPC Infrastructure Engineer - GPU Clusters

At ElevenLabs, a leading AI research and product company known for its human-like AI voice models and expansive platforms, the HPC Infrastructure Engineer role is part of the Research Engineering team. The company emphasizes rapid innovation, minimal bureaucracy, and a global, impact-driven culture. This position supports the infrastructure that powers AI model training, ensuring high performance and reliability of GPU clusters across various locations including the US, UK, Poland, and Bulgaria.

The engineer will manage NVIDIA GPU clusters, focusing on provisioning, scheduling, monitoring, and capacity planning to optimize training throughput. Responsibilities include automating maintenance tasks, managing the software stack beneath training code, tuning job schedulers, maintaining high-performance storage, and troubleshooting hardware and network issues. The role requires hands-on hardware work and collaboration with datacenter staff, emphasizing end-to-end ownership and operational excellence.

Candidates should have experience running large-scale Linux or GPU environments, strong knowledge of the NVIDIA stack, and comfort with bare-metal servers and high-speed networking. Proficiency in automation using Python, Bash, and infrastructure-as-code tools is essential. Familiarity with ML training workloads, GPU cloud providers, and parallel filesystems is a plus. The position is fully remote with optional office locations, offering a dynamic environment for engineers passionate about AI infrastructure.