Xin Li

WORK HISTORY

EDUCATION

ABOUT XIN LI

I am a Ph.D. Candidate in Computer Engineering at Arizona State University. My current research centers on vision-language models (VLMs) combined with multi-agent collaboration and workflow systems, along with AI infrastructure optimization. I study how multimodal models can reason and cooperate through agent frameworks, while improving efficiency and scalability at the systems level. Our research is built on applications in medical and brain imaging, where multimodal understanding and precision are critical.Before academia, I spent several years in industry, working in game development and iOS engineering at Tencent and Bilibili. These experiences gave me a strong understanding of graphics, real-time systems, and mobile performance, and continue to influence my research philosophy on efficient, deployable AI.Since 2021, I have been with the Geometry Systems Lab (GSL), starting from computer graphics, then moving into computer vision, and now focusing on VLMs, agents, and system-level AI optimization.In the long term, I aim to continue working in AI — from multi-agent collaboration to AI system design and optimization — building intelligent, scalable, and reliable AI infrastructure. I also remain passionate about AR/VR/XR, where my graphics background continues to inspire new ideas.