Summary
Yanbei Chen is a staff research scientist with a decade of experience advancing deep learning and computer vision, currently at Meta after leading multimodal foundation-model efforts at Amazon. He specializes in semi-supervised, unsupervised, and cross-domain visual learning and has driven production-facing multimodal systems—contributing to the Amazon Nova family and Rekognition APIs—bridging research and engineering at scale. Yanbei’s work spans scalable multimodal training recipes, efficient architectures, and multimodal embeddings for unified semantic search across text, images, video, and audio. Trained with a PhD in computer vision and a background across top research labs in Europe and industry, he combines rigorous academic methods with product-driven deployment. An often-overlooked strength is his track record of translating synthetic-caption and hyperbolic-learning ideas into open-world detection and large-scale multimodal capabilities.
10 years of coding experience
5 years of employment as a software developer
Master's degree Computer Vision (Systems Control and Robotics), Master's degree Computer Vision (Systems Control and Robotics) at KTH Royal Institute of Technology
Engineering Summer Program - Robotics, Engineering Summer Program - Robotics at Tohoku University
Doctor of Philosophy - PhD Computer Vision Deep Learning Machine Learnin, Doctor of Philosophy - PhD Computer Vision Deep Learning Machine Learnin at Queen Mary University of London
Bachelor's degree Electrical Engineering, Bachelor's degree Electrical Engineering at Zhejiang University
English, Chinese