Software Engineer, Hermetic Build

Responsibilities

• Own the build and CI experience from top to bottom, migrating the company to a hermetic, deterministic build system (e.g., Bazel or Buck2)
• Audit current build and test pipelines to identify failures in reproducibility, hermeticity, and speed
• Implement and manage remote execution and remote caching to improve CI and local build times
• Pin toolchains and seal dependencies to prevent host environment leakage into builds
• Define and execute the long-term roadmap for build, test, and CI infrastructure as the codebase scales
• Collaborate with application and infrastructure engineers to ensure seamless migration across the stack

Requirements

• Senior or staff-level experience managing large-scale build systems like Bazel, Buck2, or Pants
• Proven experience operating remote execution and remote caching in a production environment
• Proficiency in TypeScript and Rust
• Strong expertise in CI operations, including managing queue health, flake budgets, and build time budgets
• Deep understanding of determinism and reproducibility in software builds
• Ability to work independently and scope complex infrastructure projects without a predefined spec

Preferred Qualifications

• Experience migrating a codebase onto or off of Bazel
• Experience with polyglot or protobuf-heavy monorepos
• Prior experience working on developer infrastructure at an autonomy, robotics, or systems-focused company

Benefits

• Competitive salary ($175K – $250K) and generous equity grants
• 100% covered medical, dental, and vision insurance premiums for employees and dependents
• 401(k) retirement plan with up to 4% matching
• Unlimited paid time off and 10+ observed holidays
• Paid parental leave for biological, adoptive, and foster parents
• Daily lunch provided at the office
• Visa sponsorship available

About the Company

SF Compute runs large-scale GPU clusters (H100s, H200s, B300s) on contracts that allow for liquidity. We operate the entire stack from UEFI up, providing a liquid market for GPU offtake. Our team has managed over $1B of hardware and is building the defining infrastructure marketplace for the AI era.