Ropedia raises $22M to build the data layer for physical AI
What's the deal? Singapore startup Ropedia has raised a $22 million pre-Series A round, bringing its total funding past $30 million. The company builds what it calls the data infrastructure layer for physical AI — the systems that let robots learn from real-world action rather than video.
Why now? Robots can't learn to grip an object or time a movement by watching footage; they need records of how real actions feel. Co-founder and chief technology officer Fangzhou Hong argues that data, not hardware, is becoming the bottleneck in robotics.
What's the endgame? Ropedia will use the funds to scale manufacturing, data collection, and R&D simultaneously. An earlier $8 million round carried the company to this point.
"Hardware determines what a robot can physically do. Data determines how many useful things it actually knows how to do," Hong said.
What's the geography? Hong sees Asia emerging as a hub for physical AI, citing its concentration of manufacturing, supply chains, hardware, and dense industrial ecosystems for real-world testing. He argues Singapore can lead by becoming a place to train, evaluate, and deploy robots, pointing to its universities, regulations, and demand across logistics, construction, hospitality, and public services.
The signal: Investors are treating robotics as the next major computing platform, driven by labor shortages, aging populations, and advances in AI. Ropedia's raise reflects a bet that as robots grow more capable, the ability to collect and scale high-quality real-world data becomes the fundamental — and most valuable — piece of the stack.
Read more: TechNode Global
Image credit: jurvetson
