The San Francisco Compute Company is building a liquid market for GPU offtake to de-risk the largest infrastructure build-out in history, addressing the risk associated with financing GPU clusters and enabling smaller companies to access compute resources.
Requirements
- Experienced in a systems programing language (C, C++, Rust)
- Understand command and control patterns (control loops, reconciliation, state management.)
- Designed data models in SQL and implemented the corresponding APIs (gRPC, REST) that expose them.
- Experience with Rust
- Experience working on large-scale distributed scheduling problems
Responsibilities
- Develop novel systems that integrate our compute market with the orchestration software managing virtual machines running on cutting-edge HPC hardware worth billions of dollars.
- Develop user facing APIs that integrate control plane primitives with our compute market.
- Designing scheduling algorithms that translate marketplace contracts into hardware allocations.
- Extending the interface between the orchestration service and hypervisor to support new virtual machine provisioning capabilities and lifecycle operations.
Other
- High agency engineer
- Engineer who cuts through ambiguity to ship fast.
- Comfortable translating product ideas into design proposals and iterating on feedback.
- Previously worked at or founded a startup
- Generous equity grant
- Visa Sponsorships
- Unlimited paid time off
