Purpose of the Job:

Contribute to the development, testing, and optimization of perception algorithms for Brightskies’s Level 3 and Level 4 autonomous driving systems, with strong hands-on coding skills in C++ and Python for multi-sensor environmental perception.

Responsibilities and Duties:

• Develop and refine algorithms for 2D/3D object detection, classification, tracking, and segmentation using LiDAR, camera, radar, and ultrasonic data.
• Implement multi-sensor fusion methods.
• Assist in the integration of perception models into real-time embedded processing pipelines.
• Perform model training, evaluation, and optimization to improve accuracy and efficiency.
• Contribute to sensor calibration and synchronization procedures.
• Execute perception modules in both simulation and real-world environments, logging and analyzing results.
• Conduct data preprocessing, cleaning, and annotation validation for model development.
• Apply model optimization techniques such as quantization and TensorRT deployment.
• Participate in code reviews, debugging, and CI/CD pipeline integration.
• Keep up to date with perception-related research, frameworks, and tools.

Education:

Bachelor’s degree in Computer Science, Electrical Engineering, Robotics, or a related field.

Experience:

• 0–3 years of experience in computer vision, perception, or machine learning
• Proficiency in both C++ (modern standards) and Python for algorithm implementation and integration.
• Solid understanding of LiDAR, camera, radar, and ultrasonic sensor data formats and processing methods.
• Hands-on experience with deep learning frameworks (TensorFlow, PyTorch) for perception tasks.
• Familiarity with ROS / ROS2 and real-time system constraints.
• Knowledge of probabilistic state estimation (e.g., Kalman filters) and basic tracking algorithms.
• Experience working with datasets for perception model training and validation.

Additional:

• Proficient with Git, Linux development environments, and Agile workflows.
• Familiarity with GStreamer or similar high-throughput streaming frameworks is a plus.
• Exposure to embedded deployment workflows and GPU acceleration is an advantage.
• Strong interest in autonomous vehicle technology and real-time robotics applications.

Skills and Abilities:

• Programming: C++17/20, Python 3.x (strong coding ability required)
• Frameworks & Libraries: ROS, ROS2, TensorFlow, PyTorch, OpenCV, PCL
• Sensors: LiDAR, camera, radar, ultrasonic
• Optimization Tools: TensorRT, CUDA (basic usage)
• Version Control: Git, GitLab/GitHub
• Soft Skills: Problem-solving, adaptability, attention to detail, ability to work in a collaborative team