Software Engineer, Hermetic Build

The San Francisco Compute Company•San Francisco, CA

3d•Onsite

About The Position

We're building the company which will de-risk the largest infrastructure build-out in history. When people finance GPU clusters, the datacenters housing them, and the infrastructure powering them, they need "offtake" - meaning someone has signed a contract to lease the cluster for a period of time before its even built. Financing a GPU cluster is inherently risky, since margins are thin and volumes are huge. Lenders don't want to take on the risk that cluster developers can't repay their loan, and cluster developers really don't want to risk not selling their cluster. As a result, risk is offloaded to the customer using fixed-price long-term contracts. If you don't mitigate this customer risk, there's a bubble. This isn't SaaS anymore - application layer companies sign multi-year contracts for computer and inference, but sell to customers on monthly subscriptions. If you mess up a purchase, it's game over: a minor shift in your revenue growth rate might mean the difference between profit or bankruptcy. But what if companies could exit their contract by selling it back to the market? Otherwise, as AI scales, compute only becomes available to folks who can effectively take on that risk. A 2-person startup in a San Francisco Victorian can't realistically sign a 5-year take or pay contract on $100m supercomputers. But they may be able to buy the month of liquidity that someone else sold back. So that's what we make: a liquid market for GPU offtake. The San Francisco Compute Company runs large-scale GPU clusters (H100s, H200s, B300s) on contracts you can exit. Need 256 H100s for three days? Buy them at market price, cancel what you don't use. We operate the stack from UEFI up, so you're never paying a reseller markup or waiting on a support ticket. Customers include NVIDIA, MIT, Liquid AI, and Roboflow. We're a small team that has managed over $1B of hardware and is building what we think will be the defining infrastructure marketplace for the AI era.

Requirements

Senior or staff-level experience running Bazel, Buck2, Pants, or a comparable system somewhere the build system genuinely mattered
Experience operating remote execution and remote caching in production
Comfortable across language ecosystems. We run TypeScript and Rust today, with Python showing up.
Strong opinions on determinism and reproducibility, with the judgment to know when full hermeticity is worth the cost and when it isn't
CI ops chops: queue health, flake budgets, real test signal, build time budgets you can defend
Able to scope your own work. There's no spec for what our build system should look like.

Nice To Haves

experience moving a codebase onto Bazel (or off of it)
polyglot or protobuf-heavy monorepos
prior work on developer infrastructure at an autonomy, robotics, or systems company

Responsibilities

Audit the current build and test pipeline (Bun for TypeScript, Cargo for Rust, buf for protobuf, plus Docker and Helm) and write down where it fails on reproducibility, hermeticity, and speed
Pick a build system and migrate us onto it without breaking shipping
Stand up remote execution and remote caching that actually move CI and local build times
Pin toolchains, seal dependencies, and stop the host environment from leaking into builds
Run the long-term roadmap for build, test, and CI as the team and codebase grow
Work alongside application and infrastructure engineers throughout, since the migration touches all of them