Gao Jiyang

Gao Jiyang is the founder of Xinghaitu and the central guest in 132. 对星海图创始人高继扬的3小时访谈：鲶鱼、曾国藩、Waymo与Momenta的两面、一只狼与许华哲的离开. The source presents his path as a sequence from physics competition and Tsinghua training, through early deep-learning exposure at SenseTime, engineering work at Waymo, mass-production autonomous-driving work at Momenta, and finally Embodied AI entrepreneurship.

Key Views

• AI should be applied where it is a bottom-level variable of the industry, which is why autonomous driving became attractive before robotics.
• High-performing technical work still depends on induction, decomposition, measurement, schedule discipline, and practical execution.
• Waymo gave him strong engineering training, while Momenta gave him customer-facing, result-oriented operator training.
• Xinghaitu has to build both the robot body and the intelligence layer because the whole machine is the product and the data-collection carrier.
• The long-term moat in robotics comes from Physical World Data Flywheel, Real Robot Data Strategy, and the ability to turn machines into useful production workers.
• Algorithm innovation matters, but it has to be judged inside Embodied AI Value Chain and by customer value rather than by isolated publication or demo speed.

Connections

• Xinghaitu — company he founded.
• Waymo and Momenta — autonomous-driving companies that shaped his engineering and operating style.
• Zhao Hang, Yang Zeyi, Xu Huazhe, and Cao Xudong — collaborators or comparison figures in the episode.
• Embodied AI, Wheel-Based Dual-Arm Robots, and Production Robot Scenario Selection — robotics strategy he describes.