Staff Mapping and Localization Engineer

Staff Engineer - Maps & Localization team

About Us & The Role

We are at the frontier of building the AI foundation for autonomous driving, with a commercial focus on autonomous trucking and further applications ahead. The Maps & Localization team is technically entwined with the hands-on activity of the organization ; where you will maintain technical leadership and ownership across mapping, and the localization stack.

You will be the ones writing, debugging, and shipping the code, where you will enjoy real ownership of systems that sit on the critical path of a driverless truck, working close to the planning and autonomy teams that consume your output.

Scope & Requirements :

• Staff-level engineer, with experience (5-10 years+) who writes production code today, not a team lead or manager.
• A track record of owning complex systems end to end and shipping them to production, with enough depth to set technical direction for your area.
• Deep domain strength in 3D geometry. Autonomous-vehicle experience is not required; adjacent fields such as robotics, AR/VR, or geospatial are welcome.
• Strong software fundamentals in C++ and Python within very large code bases, with experience on real-world data.
• Strong programming and scripting proficiency : C++ & Python.
• Hands-on LiDAR or radar localization, state estimation, and sensor fusion, with production real-time software.
• Own large-scale testing and validation for localization. This is the software engineering side of the localization problem.
• Code first: You write production code every week. Your work is measured in code that ships and runs on the vehicle, not in documents or delegation.
• Full ownership: You own a system end to end: design, implementation, testing, deployment, and on-road or on-data debugging when it misbehaves.
• Decide and build: You make the architecture and algorithm decisions for your area, and you implement them yourself rather than handing them to someone else.
• Real data, real bugs: You dig into real-world data, reproduce failures, and trace them to root cause in very large C++ and Python code bases.
• Lead by contributing: You raise the bar through code review, design review, and example. You influence others by being the strongest hands-on contributor in the room, not by managing them.
• Design and ship LiDAR- and radar-based map matching running in real time in production.
• Develop state estimation and sensor fusion across radar, IMU, camera, and LiDAR. Our localization is sensor-based rather than GNSS-centric.
• Use ground-truth vehicle position data to evaluate and improve matcher robustness under weather and degraded sensing.