Role Overview
We are seeking a Senior Robotics Engineer to lead the development of advanced visual localization solutions for autonomous driving and robotics applications. The primary focus of this role is to design and implement robust map-based localization systems that enable accurate vehicle positioning in complex urban environments.
The successful candidate will be responsible for developing localization frameworks that align live semantic observations with High-Definition (HD) maps, leveraging modern optimization techniques and state-of-the-art machine learning approaches. Working closely with perception, navigation, and control teams, you will help deliver highly accurate and reliable localization capabilities to support autonomous operations.
Key Responsibilities
- Design and implement algorithms that align live semantic detections, such as lane markings, stop lines, road boundaries, and traffic-related features, with vectorized HD map layers in formats such as Lanelet2 and OpenDRIVE.
- Develop localization modules that utilize Bird's Eye View (BEV) feature representations to perform spatial associations and estimate vehicle pose with high accuracy.
- Implement and optimize differentiable pose estimation solvers and neural matching approaches to enhance or replace traditional association and filtering methods.
- Design and develop robust camera-based SLAM and Visual-Inertial Odometry (VIO) solutions for urban and structured environments.
- Integrate localization functionality with navigation, planning, and vehicle control systems.
- Collaborate with multidisciplinary teams to ensure seamless deployment of localization software within larger autonomous systems.
- Research, evaluate, and implement state-of-the-art techniques in visual localization, mapping, state estimation, and machine learning.
- Contribute to the continuous improvement of software architecture, performance, scalability, and reliability.
Qualifications & Skills
- Extensive experience with map-based localization using HD maps, including Lanelet2, OpenDRIVE, or comparable formats.
- Strong background in differentiable optimization, neural pose estimation, and machine learning frameworks such as PyTorch, PyPose, Theseus, or similar tools.
- Proven expertise in non-linear optimization and state estimation using frameworks such as Ceres, g2o, GTSAM, or equivalent.
- Strong understanding of Bayesian filtering techniques and autonomous navigation systems.
- Advanced knowledge of coordinate transformations, Lie groups and Lie algebras (SO3, SE3), and 2D/3D geometric reasoning.
- Expert-level programming skills in modern C++ (C++17/C++20) and Python.
- Strong experience with ROS 2 and software development for robotics platforms.
- Experience designing performance-critical, production-quality software within Linux environments.
Preferred Qualifications
- Experience working with learned Bird's Eye View (BEV) representations and probability-map alignment techniques.
- Familiarity with autonomous driving software stacks and modular autonomy architectures.
- Knowledge of geometric deep learning and spatial data association methods.
- Experience applying machine learning techniques to visual odometry, localization, scene understanding, or perception-related challenges.
- Exposure to large-scale autonomous vehicle or robotics deployments.
Education
- Master's or PhD in Robotics, Computer Science, Computer Engineering, Electrical Engineering, or a related field.
