About the Role
We are seeking a Founding Software Engineer to manage the data flow from AI models to actuators. This role involves working closely with hardware, from kernel and firmware to controls and perception, to manage real-time pipelines with low latency. The objective is to enhance the determinism of the end-to-end inference pipeline.
Responsibilities
- Ship performance-critical code to real robots daily.
- Own the full data flow, from camera frames to control commands and actuator feedback.
- Manage the real-time control loop, ensuring deterministic operation alongside vision inference.
- Build the perception data path for high-bandwidth camera streams with minimal latency.
- Reduce latency and optimize the stack for speed, determinism, and reliability.
- Develop tooling and instrumentation for observing and debugging real-time behavior in controls and vision.
Requirements
- 5+ years of software engineering experience in production systems close to hardware (drivers, kernels, embedded, robotics, or similar).
- Proficiency in Rust, Python, C++, operating systems, and multithreading.
- Experience with Bazel, Nix, HIL testing, and CI/CD.
- Experience with Linux kernel hacking and embedded systems.
- Proven ability to debug timing and concurrency issues.
- Experience owning and resolving ambiguous problems.
- Ability to work on-site in San Francisco 5 days/week, with potential for 6 days/week during crunch times.
Skills
- Rust
- Python
- C++
- Operating Systems
- Multithreading
- Bazel
- Nix
- HIL testing
- CI/CD
- Linux kernel
- Embedded systems
- Real-time control systems
- Computer vision pipelines
- Tracing
- Profiling
- Linux I/O
- DPDK
- SPDK
- io_uring
- AI tools
Location
- San Francisco
Work Type
- On-site
Experience Level
- 5+ years
