Machine Learning Engineer - Computer Vision and Perception

About the Role

Under the mission of "Making Robots Intelligent," Seoul Robotics is developing the world's best perception technology for autonomous driving. We are looking for talented software engineers who can design, implement, and productize technologies that precisely perceive the world by fusing data from various sensors such as LiDAR and cameras. You will develop and optimize algorithms that process vast amounts of LiDAR point clouds and high-resolution camera images in real-time using deep learning and various cutting-edge technologies.

Responsibilities

• Develop and implement 3D object detection, tracking, segmentation, and classification algorithms using LiDAR and camera data.
• Convert and integrate deep learning models developed in PyTorch, TensorFlow, etc., into C++ based high-performance inference engines.
• Profiling and optimization for real-time processing.
• Implement Sensor Fusion algorithms for autonomous driving systems.
• Build data pipelines.
• Verify algorithms through simulation.

Qualifications

• Bachelor’s degree or higher in Computer Science, Robotics, Mechanical Engineering, Electrical Engineering, or a related field.
• Proficiency in Python or Modern C++ (C++14/17 or higher) programming.
• Fundamental knowledge of computer science such as data structures, algorithms, and operating systems.
• Related mathematical knowledge such as linear algebra, probability and statistics, and geometry.
• Basic knowledge of machine learning and deep learning (CNN, PointNet, Transformer, etc.).
• Business English proficiency (Collaboration with a multinational team is essential).

Preferred Qualifications

• Experience in LiDAR point cloud processing or 3D computer vision projects.
• Experience implementing tracking algorithms using Kalman Filter, etc.
• Experience in parallel computing and acceleration using CUDA, TensorRT, etc.
• Experience using Git, Docker, CMake, and Linux environments.