Software Engineer, Perception (SLAM/VIO)

We are building the perception stack that enables robust three/six degrees of freedom tracking, localization and mapping in complex real-world environments. This role focuses on fusing data from cameras and inertial sensors to power SLAM / Visual-Inertial Odometry pipelines. We need an experienced perception software engineer that can create algorithms and systems that run in real time, handle uncertainty gracefully, and provide the foundation for advanced spatial intelligence on edge devices.

- Develop and optimize sensor fusion algorithms combining IMU and camera data for SLAM and VIO (Simultaneous Localization and Mapping & Visual Inertial Odometry)

- Implement state of the art structure-from-motion techniques using different sensor modalities to generate consistent 3D reconstructions

- Build Calibration and sync pipelines across multiple sensor types

- Evaluate algos against public benchmarks and real-world datasets

- Integrate outputs into higher-level autonomy or collaborative mapping systems (i.e. shared mapping systems)

- Collaborate with HW, robotics and platform teams to ensure end-to-end performance

- Define metrics, testing frameworks and deployment strategies for production-ready perception systems

- BS with 10+ years of academic or industry experience in inertial sensing, computer vision, robotics or related fields with shipped or published work (or MS with 8+ yrs of the above)

- Strong background in linear algebra, SLAM, VIO and structure-from-motion

- Proficient in C++/Python with experience in real-time optimization

- Knowledge of sensor modelling, calibration and noise characteristics

- Experience with real-time processing pipelines on embedded or edge hardware

- Familiarity with probabilistic estimation, filtering (e.g. EKF/ factor-graph optimization)

- Preferred track record of applied research in robotics or AR/VR