Bosch Chassis Systems India Private Ltd. is seeking a Traffic Light Vision Perception Algorithm Engineer to design, develop, and optimize vision perception algorithms for autonomous vehicles. You will ensure recognition accuracy and real-time performance in complex urban driving scenarios.
What You'll Do
- Design, develop, and iteratively optimize traffic light and traffic sign visual perception algorithms to ensure accuracy and real-time performance in complex urban scenes.
- Lead the full pipeline algorithm development for traffic light detection, classification, state recognition, tracking, and prediction, solving challenges like occlusion, backlight, rain, and night conditions.
- Design and implement perception fusion algorithms based on multi-sensor visual data.
- Train, evaluate, quantize, compress, and optimize perception models to meet the computational power and latency requirements of automotive embedded platforms.
- Collaborate with system, localization, and planning & control teams for algorithm integration, issue diagnosis, and real-world vehicle road test problem resolution.
- Track industry-leading technologies, lead algorithm solution selection and implementation, and continuously improve system stability and robustness.
What We're Looking For
- Bachelor's degree or higher in Computer Science, Electronic Engineering, Automation, Robotics, or a related field.
- 3-5 years of experience in autonomous driving visual perception algorithm development.
- Expertise in traffic light/traffic object detection, classification, and tracking algorithms.
- Proficiency with deep learning frameworks like PyTorch and the ability to independently train, tune, and analyze models.
- Solid computer vision fundamentals, familiar with traditional vision algorithms and deep learning object detection methods.
- Proficient in Python programming with strong engineering implementation, code optimization, and debugging skills.
- Excellent problem-solving, teamwork, and technical closure capabilities.
Nice to Have
- Experience with city NOA (Navigate on Autopilot) traffic light perception implementation.
- Experience in model development for automotive embedded platforms, especially Horizon Robotics platforms.
Technical Stack
- PyTorch
- Python
