Spellbrush
HPC/ML Infrastructure Engineer
San Francisco or Tokyo · On-site · Full-time · Posted Jun 15, 2026
- Visa sponsorship
- Visa sponsorships are available.
About the job
Lead the bring-up, administration, and operation of a large GPU training cluster, supporting researchers and keeping compute, storage, and networking running.
What you'll do
- Bring up and administer HPC infrastructure and bare-metal GPU machines.
- Operate and troubleshoot SLURM jobs, parallel filesystems, networking, and cluster systems.
- Support AI researchers training anime image models and help build out edge datacenters.
What they ask for
- Relevant experience across the modern HPC software stack.
- Traditional Linux systems administration skills.
- Comfort working with physical hardware and small, fast-paced teams.
Nice to have
- On-site collaboration in Tokyo or San Francisco; Bay Area is strongly preferred.
- Interest in anime and anime aesthetics.
Tools
SLURM, Slinky, K8s, warewulf, MAAS, ansible, WEKA, VAST
AI focus
Support researchers training anime image models on a large GPU cluster.
Model-written summary · the original posting is authoritative