About Me

I'm Coco Xu, originally from Shanghai, China and currently based in Palo Alto. I'm passionate about leveraging AI and computer vision to solve meaningful problems.

Research Focus: AI systems, computer vision, reinforcement learning, full-stack development, and human-computer interaction.
Current Work: At Stanford AI Lab, I'm developing a long-context self-improving memory agent for LLMs focused on code generation. Previously, I transformed a CLI tool into a scalable web interface at Meta's Superintelligence Lab, and led frontend development for the Stanford Deliberation Platform.
Technical Skills: Python, C++, Swift, JavaScript, React, PyTorch, Computer Vision, ML/AI, and more (see Experiences for full technical stack).