What will your job look like: Operate and maintain multi-node GPU compute and training clusters (SLURM), including scheduling policy, quotas, and job troubleshooting. Manage the NVIDIA GPU software stack across mixed generations, plus the high-speed interconnect fabric. Own storage infrastructure: parallel and network filesystems. IT site owner, POC. Administer enterprise networking: firewall policy, VPN, SD-WAN, and site connectivity. Run identity and directory services, and the SaaS SSO integrations that depend on them. Provision and manage a mixed endpoint fleet across Linux, Windows, and macOS. Handle server-room operations: power, cooling, UPS, and capacity planning. All you need is: 3+ years of experience as an IT Specialist or a similar role Linux systems administration experience (Ubuntu at scale) Identity and directory administration (Active Directory) Comfortable being the sole technical owner and working independently Hands-on HPC experience: SLURM or a comparable scheduler, GPU cluster operations, storage and network experience Supporting developing teams Nice to have: Scripting for automation (Bash, Ansible) NVIDIA GPU/driver stack, DKMS, Secure Boot InfiniBand DDN Lustre FortiGate Endpoint management tooling (Intune or similar) Cloud identity platforms (Entra ID, Google Workspace) Experience with infrastructure migrations or datacenter moves macOS fleet management Hebrew and English
