Develops multimodal computer vision and machine learning systems to index, tag, and understand video.