Staff Software Engineer, Build Systems

Anthropic•San Francisco, CA

4h•Hybrid

About The Position

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems. About the Role Anthropic is seeking talented and experienced Software Engineers to join our Builds team within the Developer Platform organization. The Builds team owns the build infrastructure that every engineer and researcher at Anthropic depends on—from hermetic build environments and container image pipelines to remote build execution, build caching, and dependency management across our monorepo. By joining this team, you'll have a direct impact on the velocity and reliability of Anthropic's research and product development. As Anthropic scales its engineering organization, research ambitions, and the volume of code contributed by both humans and AI agents, build infrastructure is becoming a critical bottleneck and a critical lever. We're building next-generation systems—scalable remote build execution with content-addressable storage, fine-grained incremental builds, and hermetic reproducible environments—to ensure that build times shrink even as the codebase grows by orders of magnitude. Our monorepo spans Python, Rust, and Go, with builds targeting multiple accelerator platforms (TPUs, Trainium, GPUs) and complex SDK dependencies. This creates uniquely challenging problems in hermeticity, caching, cross-language interop, and dependency management that few organizations encounter. If you've built or operated large-scale, hermetic build infrastructure at a company with a large monorepo and are excited about designing systems that keep thousands of engineers productive, this is the role for you.

Requirements

Have 10+ years of relevant industry experience building and operating large-scale build systems and infrastructure
Have deep experience with hermetic, reproducible build systems (Bazel, Buck, Nix, or similar) in a large monorepo environment
Understand build caching strategies, remote execution, and content-addressable storage deeply
Have experience with containerized build environments and image build pipelines
Have strong proficiency in at least one of Python, Rust, or Go
Have excellent communication skills and enjoy supporting internal partners to improve their development experience
Are excited about designing foundational systems and are comfortable working independently on ambiguous, high-impact technical challenges

Nice To Haves

Experience with Nix/NixOS and managing large Nix package sets at scale
Experience with Rust build toolchains (cargo, maturin) and Python-Rust interop
Experience building or operating remote build execution systems (Bazel Remote Execution API, BuildBarn, BuildBuddy, or similar)
Experience managing accelerator SDK builds (CUDA, Neuron SDK, JAX/XLA)
Experience with fine-grained dependency graph analysis and incremental build optimization at monorepo scale

Responsibilities

Design and build scalable, hermetic build infrastructure including remote build execution (RBE) and content-addressable storage (CAS) systems that support all source and image builds
Own and evolve Nix-based build environments and package management, keeping nixpkgs and language runtimes current and compatible
Drive fine-grained dependency management across the monorepo to enable incremental builds and eliminate unnecessary rebuilds
Build and maintain language-specific build toolchains and accelerator SDK/runtime builds
Reduce container image build times through architectural improvements to build pipelines, layer caching, and content-addressable storage
Partner with CI, Languages, and Developer Acceleration teams to ensure build infrastructure integrates seamlessly with the broader developer platform