• Ph.D. or master’s degree in Robotics, Computer Vision, Electrical Engineering, or related field.
• 1-4 years’ of demonstrated experience in developing and deploying perception systems for robots.
• Strong proficiency in programming languages such as Python, C/C++
• Hands-on knowledge of data structures and algorithms for point cloud processing and are familiar with ROS/ROS2, OpenCV and PCL
• Expertise with tools and libraries like OpenCV, PCL, Tensor RT, P y Torch.
• Experience tracking-by-detection, and multi-sensor calibration.
• Hands-on experience with 3D cameras and Li DARs for perception applications.
• Strong communication and leadership skills
•