Ethertechnology raises $15M angel round from Huawei, Hillhouse for spatial-AI infrastructure
What's the deal? Chinese startup EthertechnologyDealroom has a profile for this one. Try Dealroom → has raised US$13.8M in an angel round, backed by an unusually broad syndicate spanning Huawei Hubble, Hillhouse, robotics firms AgiBot and X Square Robot, and industrial players including iSoftStone and JLC. The company is building what it calls a "physical-space engine" — infrastructure to turn changing real-world environments into reusable, computable spatial data.
What's the endgame? Founded in December 2025, Ethertechnology aims to continuously compile physical spaces into data that robots, world models, and industry systems can call on. Its engine layers sensor inputs, spatial-state processing, and reusable "spatial assets" that capture objects' positions, physics, and interactions.
Who's behind it? Founder and chief executive officer Yang Yeqing previously served as chief of staff at HTC VIVEDealroom has a profile for this one. Try Dealroom → and led its metaverse and software units in China. The core team draws from HTC VIVE, Microsoft, JD, Meituan, and universities including Peking and Hong Kong University of Science and Technology.
What's the pitch? Yang argues the engine is not simple 3D reconstruction or digital twins, which capture static geometry or state. "It not only restores space, but makes space computable, runnable, and callable," he said (translated from Chinese), positioning the product as infrastructure beneath world models rather than a competing generator.
Why now? The round lands amid a cluster of early-stage bets on embodied-AI infrastructure. Hangzhou startup Gewu Tongzhi recently raised nearly US$13.8M for a world-model-based robotics system — a sign that capital is reaching beyond robot makers toward data and enabling technologies.
Ethertechnology says it is already working on applications in robotics and world models, with early focus on capturing human hand interactions — grasping, assembling, and tool use — to record not just motion but contact, force, and object-state changes at sub-millimetre precision.
What could go wrong? The startup's broad ambitions and stated collaborations are not yet evidence of commercial traction. The key test is whether it can deliver repeatable deployments and prove its spatial assets are genuinely reusable across customers and industries.
The signal: At roughly $15 million, the round sits near the 99th percentile for angel rounds in the region all-time, and the breadth of strategic investors signals conviction that physical-space data is becoming foundational plumbing for robotics and world models.
Read more: sohu.com
Image credit: Generated with Gemini
