Embodied AI startup OriginFlow has successfully secured multiple rounds of funding—including Angel, Strategic, and Pre-A1 rounds—within just five months of its inception, raising a total exceeding 500 million yuan. Its founder, Qin Shentao, is a PhD candidate at Tsinghua University born after 2000. The core technological innovation of the team centers on the development and implementation of the NeuroScale data acquisition framework. Leveraging neuromuscular electrical signals as a medium, this framework encodes data using proprietary kits and the foundational model PULSE, facilitating the reconstruction of continuous, multimodal hand information. This approach effectively eliminates asynchronous data discrepancies and surpasses the constraints of conventional methods. Furthermore, it boasts high acquisition efficiency and low costs (with equipment priced at approximately a thousand yuan). Currently, the company is pushing forward with industrial applications in industrial manufacturing (where it collaborates with leading companies to jointly develop scenarios) and home services (working with the 58 Group to gather household service data). Looking ahead, its technology is poised to extend into next-generation human-machine interaction interfaces, with potential applications in robotics, smart hardware, and beyond.