Software Engineer to develop calibration systems for the sensors and articulated geometry of our autonomous construction machines. Every machine we retrofit is different. There is no factory line to calibrate it on, and construction work shakes and knocks sensors out of place. You9ll build the algorithms and procedures that calibrate each machine, verify calibration accuracy to qualify it for autonomous operation, and detect when calibration degrades in the field.
Depending on your background, you may focus on sensor calibration, articulated-machine geometry, or validation and monitoring. You9ll write production software, test on real machines, and make sure field teams can run calibration independently across machine types. We9re hiring at mid and senior levels, with scope matched to experience.
What you9ll do- Develop and deploy intrinsic and extrinsic calibration algorithms for our sensors (cameras, lidars, GNSS, IMU, encoders), and procedures for reasoning about machine geometry
- Build calibration procedures that field teams can run without engineering involvement, on machines of varied types in active use at job sites
- Own calibration verification methods used to qualify machines for autonomous operation, distinguishing a good numerical fit from accurate calibration through independent measurements and system-level testing
- Define quantitative calibration quality metrics and acceptance criteria tied to machine performance
- Develop methods to detect degraded calibration on working machines, and run these checks across the fleet
- Diagnose calibration failures across sensor and kinematic models, timing, mounting, and data quality,
- Test calibration improvements using recorded data and physical machines, working with estimation, perception, hardware, and field teams to resolve field issues
What we9re looking for- 4+ years of professional engineering or applied research experience in calibration, robotics, computer vision, or a related field, or a proven track record of delivering calibration systems for real-world applications
- Strong foundations in linear algebra, 3D geometry, nonlinear optimization, and numerical methods
- Hands-on experience developing sensor or kinematic calibration methods, including understanding the underlying models and what makes their parameters observable
- Experience validating calibration accuracy beyond an optimizer9s residuals, and investigating failures using real sensor or machine data
- Strong production software skills in Rust or modern C++. Our production stack is primarily Rust, and we will support experienced C++ engineers as they ramp up
- A degree in Robotics, Computer Science, Electrical Engineering, Mechanical Engineering, Applied Mathematics, or a related field, or equivalent practical experience
Ways to stand outOne or more of the following:
- Experience shipping software on robots, vehicles, or other machines in the field
- Experience with in-field, targetless, or online calibration, or with calibrating articulated machines
- Experience formulating and solving large nonlinear optimization problems, such as bundle adjustment or joint estimation of calibration, timing, and motion
- Production experience with Rust
Other special aspects of the role- Based in the Bay Area with the ability to be onsite at our SF office on a daily/weekly basis