Cluster Ops

We manage thousands of GPUs today and need to grow this with reliability, security and performance in mind.

You’ll be working on ops for multi-provider GPU clusters.

• GPU type and count you’ve managed
• Providers you’ve worked with. Eg Hyperscalers, neoclouds, on prem.
• Interconnect you’ve managed.
• What tooling you used eg. for provision, scheduling, storage, monitoring, cost management etc.
• What tooling you developed.

• 🚀 We move fast. We ship weekly—new features, improvements, and fixes go live fast. Our infra runs cluster scale up tests daily.👥 We test big. Every month, we stress test with large groups of users face to face, get real-world feedback, and iterate rapidly.
• 💻 We build together. Weekend hackathons push boundaries, drive innovation, and help us level up as a team.
• 🔄 We iterate relentlessly. Direct user feedback shapes our roadmap—we release, test, refine, and keep moving.
• ✈️ We travel when needed. Engineers may travel between SF and Sydney to run events, attend conferences, and meet with clients.