Software Engineer, Robotics Platform

About the Role

We're hiring a Software Engineer, Robotics Platform to help us scale our fleet of robots. This role is on-site five days a week in San Francisco — it's how our small, high-ownership team moves.

The robot platform is the foundation layer our entire robotics stack — and the ML models we run on the robot (our "forward-deployed" models) — runs on: robot compute, OS, middleware, and networking. You'll own it, making it a stable, performant base for our perception and robotics algorithms. The work spans everything from low-level system software to high-bandwidth sensor logging, debugging, and performance analysis, in close partnership with senior robotics and perception engineers.

Our next-generation platform is greenfield: we're taking everything we've learned from the platform running in production today and designing the next one from a clean sheet — run as an applied hypothesis → experiment → learn loop. You'll be central to that.

We want architects who work full-stack — from system design down to the metal — not armchair architects. Engineers who design a system and live with its low-level consequences, who feel the pain a design inflicts downstream and account for it from the start.

And while we hire you for a specific set of strengths, we won't pigeonhole you into a narrow lane. You're free — encouraged — to contribute anywhere in the company you can make it better.

What You’ll Do:

• Architect and build our next-generation robotics platform — the foundation layer our robotics stack and forward-deployed models run on, spanning compute, OS, middleware, and networking — taking the lessons from the platform in production today into a clean-sheet design.
• Build the infrastructure and frameworks that carry a robot across its whole software lifecycle — provisioning a new instance so it joins the fleet, then keeping its software updated from there on — working with manufacturing to do this at scale.
• Lead an upgrade of our autonomy software OS and middleware.
• Help integrate new robot hardware, sensors, and cameras, and monitor performance.
• Investigate, debug and fix difficult low-level systems software problems and performance, e.g. with drivers, networking, Linux configuration.

Qualifications

• Strong experience and expertise in Linux systems — building complex software in a Linux environment.
• Strong systems programming in C, C++, or Rust (Python a plus).
• Excellent software engineering fundamentals.
• Experience leading and shipping complex software projects.
• Experience deploying and managing software across networked edge devices (robot fleets, embedded systems) and/or with ROS 2 middleware — the DDS-based publish/subscribe (topics) layer that connects a robot's software components.

Nice-to-Have

• Robotics experience.
• Experience with the broader robotics tech stack, e.g. OpenCV, CUDA.
• Understanding of network and time protocols, e.g. TCP, UDP, PTP.
• Experience in systems engineering.
• Experience with basic mechatronics and electronics.
• Experience leading projects or people in a tech-lead or similar capacity.
• A Master's or PhD — not required, but we value the depth of reasoning that sustained research develops.