vSLAM Algorithm Engineer (Modern C++)

About the Role

We are looking for a passionate engineer to develop the core algorithms of our vSLAM (visual Simultaneous Localization and Mapping) system. You will design and implement advanced perception, localization, and mapping algorithms using modern C++.

Responsibilities

- Develop and optimize real-time vSLAM algorithms (feature extraction, loop closure, optimization, sensor fusion).

- Design high-performance C++ code for perception and localization pipelines.

- Integrate algorithms with robot systems and validate in simulation and field environments.

- Collaborate with robotics and embedded engineers to ensure end-to-end performance.

Requirements

- Bachelor's or Master's degree in Computer Science, Robotics, or related field.

- Strong skills in C++17 or newer, with good knowledge of data structures and concurrency.

- Solid foundation in linear algebra, geometry, and estimation theory.

- Experience in one or more of: visual odometry, SLAM, structure-from-motion, or bundle adjustment.

- Experience with ROS/ROS2 and Linux development.

- Strong curiosity and learning skill on new technologies.

- Experience with GPU/CUDA optimization or computer vision frameworks (OpenCV, Eigen, Sophus, g2o, etc.).

- Familiarity with Python for prototyping and analysis.

- Proficient in English and Mandarin in business standards as you are required to liaise with Mandarin speaking stakeholders and work in projects with them.