GL Ventures Invests in XintLabs, an 'AI + Human Body Motion Model' Research Company in Shenzhen
Recently, Shenzhen XintLabs Technology Co., Ltd. announced the completion of tens of millions of yuan in angel round financing, exclusively invested by GL Ventures. The funds will be mainly used for AI model research, product development, and market promotion.
According to GL Ventures' official WeChat account, XintLabs' research direction is the "human body model". This model breaks away from the traditional framework of simple posture recognition and motion scoring, integrating dimensions such as posture, strength, balance, control, fatigue, intention, and their change trajectories into a computable, predictable, and optimizable time-varying quantification system. In XintLabs' view, for AI to truly improve the relationship between humans and the world, it must not only study the world but also study humans.
XintLabs first chose the sports scenario because sports provide a complete closed loop for building human body models: clear goals, high-frequency feedback, continuous data, and verifiable results. From skiing and golf to more sports training scenarios, each user's movement forms an observable, feedback-driven, and iterative data loop.
XintLabs' core team comes from universities such as Hong Kong University of Science and Technology, Imperial College London, Chinese University of Hong Kong, and University of Hong Kong, as well as technology and hardware companies like DJI, ePropulsion, BYD, and Speediance. The team's capabilities cover AI algorithms, robot perception, software-hardware integration, sports science, engineering R&D, and product implementation.
Founder and CEO Liu Yuefeng graduated from Hong Kong University of Science and Technology with a degree in Mechanical and Aerospace Engineering. He previously joined ePropulsion as an early core member, leading the establishment of the product system and the definition, R&D, and commercial implementation of multiple core products. He has long been deeply involved in skiing, freediving, sailing, golf and other sports, and even worked as a ski instructor. His cross-sport learning and teaching experience made him realize: The key to improving ordinary people's athletic ability is not to get more standard answers, but to build a feedback system that truly understands one's own body.
Founder and CTO Wang Siao is a humanoid robot perception expert and computational neuroscientist, long focused on neuromorphic computing, event-driven perception, and robot decision-making mechanisms. He earned his bachelor's degree in Mathematics from Hong Kong University of Science and Technology, followed by a master's in Robotics from the University of Bristol, a PhD in Bioengineering and Robotics from the University of Genoa, and postdoctoral research in event-driven perception at the Italian Institute of Technology (IIT). His technical expertise covers high-speed low-latency perception, dynamic system modeling, brain-like computing, and robot control decision-making, highly aligned with XintLabs' technical path for building human body models.
XintLabs believes that the next step for AI entering the physical world is not only to make machines understand the world better, but also to make AI understand humans better. If the world model is the foundation for AI to understand the external world, then the human body model is the cornerstone for AI to understand human behavior. XintLabs is starting from sports training to build a Body Operating System for everyone, allowing AI to truly understand how the body perceives, acts, learns, and evolves.
| Source: GL Ventures official WeChat account
| Image source: GL Ventures official WeChat account; Unsplash