Developer Productivity Engineer

Making progress towards the above requires tools that let researchers and engineers move quickly from an idea to a working experiment and, ultimately, to a real robot.

Role Overview

We're looking for an engineer to own and improve that development experience. You'll design, implement, and maintain our build system and developer tooling, working directly with the people who use them. Your impact will show up in faster feedback, reliable builds, easier onboarding, and fewer interruptions across the team.

Responsibilities

• Own & Evolve Build Infrastructure: Own and evolve our Bazel build system, including build rules, toolchains, dependency management, and remote caching.

• Optimize Velocity & Reliability: Make builds and tests faster and more reliable across local development and CI.

• Streamline Engineering Workflows: Improve everyday workflows, from setting up a development environment to debugging a failed test or packaging a training job.

• Build Internal Tooling as a Product: Build tools that make common tasks straightforward and failures easy to understand.

• Collaborate & Measure Impact: Work with researchers and robotics engineers to identify recurring friction, prioritize improvements, and measure whether they help.

• Ensure Multi-Platform Reproducibility: Keep development environments and build outputs reproducible across Linux, macOS, and GPU-enabled containers.

Minimum Qualifications

• Build Systems & DevEx Experience: Track record of building and maintaining build systems, developer tools, or CI infrastructure used by other engineers.

• Product Mindset for Internal Tools: Ability to write clear, maintainable software and treat internal tools as products with users.

• Systems & Build Theory Expertise: Deep understanding of dependency graphs, caching, reproducibility, and the tradeoffs involved in speeding up builds and tests.

• Polyglot & Systems Tooling: Proficiency with Python, Linux, shell tooling, and debugging problems across language and system boundaries.

• Autonomy & Ownership: Ability to take an ambiguous problem, work with its users, and carry a solution through implementation and adoption..

Preferred Qualifications

• Hands-on experience with Bazel

• Experience managing C++ toolchains, Python packaging, GPU software dependencies, or machine learning training infrastructure.

• Prior robotics or reinforcement learning experience

Our Tech Stack

• Build & CI/CD: Bazel, Bzlmod, Gazelle, GitHub Actions

• Languages: Python, C++

• Infrastructure & Containerization: Docker, Google Cloud Platform (GCP), Terraform

• Compute & ML Frameworks: JAX, CUDA, MuJoCo/MJX, NVIDIA Warp

Note: You do not need experience in every tool listed above to be successful in this role.

Success means engineers spend less time waiting for builds, repairing environments, and diagnosing infrastructure failures, and more time developing and testing ideas. You'll build the tools that multiply the team's output.