Lyceum · Zurich
You will work with the engineers who bring Lyceum’s GPU capacity online and keep it running. The team’s aim is a platform that operates without manual intervention: every new cluster should go live through the same automated, repeatable process, and changes should reach production without customers noticing.
You will bring new GPU clusters online for Lyceum, from the technical specification of a cluster to the moment it serves its first customer workload. As we add GPU capacity across European data centres, you make sure every new cluster is specified correctly, set up fast, and handed over in a state customers can rely on.
Technical specification: Define cluster designs together with our hardware and data-centre partners: server configuration, InfiniBand and Ethernet networking, storage, power and cooling. Review supplier specifications against what our products need.
Cluster bring-up: Own the setup from racked hardware to production: firmware, BIOS, drivers, OS images, subnet manager and network configuration, burn-in and acceptance testing.
Troubleshooting: Diagnose hardware, network and driver issues during deployment and in production, and resolve them with vendors and data-centre staff.
Automation: Turn every deployment into repeatable tooling and runbooks, so the next cluster goes live faster than the last one, and integrate new clusters into our orchestration and fleet management.
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.
Requirements
Nice to have