Light is a team of creative technologists on a mission to redefine how automobiles perceive the environment using multiview passive optics. We are a venture-backed, pre-launch startup headquartered in Redwood City, CA.
You will work in a collaborative team environment to develop cutting-edge deep learning models aiding in scene understanding and depth estimation in a multiview camera system designed for ADAS applications. You'll participate in the process from start to end, from developing data collection practices and working directly with raw data to developing, training, and productizing state-of-the-art models for realtime inference. Your efforts will play a critical role as we radically enable automobiles to see like humans.
- MS in Machine Learning/similar field or 2+ years industry experience with deep learning
- Hands-on experience with TensorFlow/Keras (preferred) or PyTorch
- Python programming proficiency
- Good communication skills and experience collaborating with cross-functional teams
- Experience with image and/or video-based machine learning tasks, e.g. semantic segmentation and object tracking
- Experience with edge inference platforms (e.g. TensorFlow Lite, TensorRT) and model optimization (quantization and pruning)
- Strong software engineering skills
- Experience with autonomous driving-related tasks
- Contributions to open source projects
- C++11 programming experience