Summary
Xin Liu is an applied scientist with nine years of experience specializing in efficient deep learning for video, computer vision, and vision-language models, currently working at Amazon. With a PhD background from TU Delft, Xin has led research and product efforts ranging from lightweight mobile segmentation and 3D reconstruction for health diagnostics to diffusion models for packaging and medical imaging. He combines strong academic publication pedigree with hands-on deployment experience across startups and industry (Danone, Veridi, Tencent, ING), bridging research innovations and production constraints. Notably, his work emphasizes training and architecture optimizations for temporal transformers and resource-efficient video models—skills that enable real-time, mobile-capable systems. Based in Delft, he is actively seeking full-time roles where he can scale cutting-edge video and multimodal ML into impactful applications.
9 years of coding experience
4 years of employment as a software developer
Bachelor of Engineering (B.E.) Electrical Engineering and Automation, Bachelor of Engineering (B.E.) Electrical Engineering and Automation at Harbin Institute of Technology
PhD candidate Computer Vision Deep Learning, PhD candidate Computer Vision Deep Learning at Delft University of Technology
Electrical and Electronics Engineering, Electrical and Electronics Engineering at 성균관대학교(SKKU)
Chinese, English, Dutch