Robotics has a data problem. A robot can’t learn to grip a bat or time a swing by watching a video — it needs records of what real-world action actually feels like. That’s the gap Singapore startup Ropedia is building to fill.
The company recently crossed $30 million in total funding, following a $22 million Pre-Series A round. As Physical AI moves out of research labs and into factories, workplaces, and eventually homes, questions about its impact are growing louder.
In an interview, Ropedia Co-Founder and CTO Fangzhou Hong told TNGlobal about why Asia is emerging as a hub for Physical AI, how Singapore can take the lead, whether data or hardware will determine the winners, and how robots entering everyday life could reshape employment.
Below are the edited excerpts:
Is Asia becoming the global center of Physical AI? Why?
Yes, I think Asia is becoming one of the most important centers of Physical AI. Physical AI requires much more than foundation models. It needs manufacturing, supply chains, hardware, real-world deployment, and large-scale data collection.
Asia has a unique combination of all of these strengths. The region already manufactures much of the world’s electronics, sensors, and robotics hardware, while also providing dense industrial ecosystems where AI systems can be deployed, tested, and improved in real-world environments. That creates a much faster feedback loop between building, deploying, and learning than many other regions can offer.
At Ropedia, we’re taking advantage of this ecosystem by building the data infrastructure layer for Physical AI.
How can Singapore become a global leader when it comes to training robots?
Singapore can win by becoming the best place not only to train them, but also to evaluate and deploy them. Evaluation and deployment are two essentials for a good robotics foundation model. We have strong universities, trusted regulations, great infrastructure, and real demand across logistics, construction, hospitality, and public services.
Those are exactly the environments where robots can be deployed and where valuable real-world data can be collected. Singapore has the opportunity to become a living testbed for Physical AI and the fastest place to connect everything needed to bring robots into the real world.
Why are investors investing in robotics? Where do you think the sector is headed?
Investors see robotics as the next major computing platform. The previous waves of AI changed how we process information. Physical AI extends intelligence into the physical world, creating opportunities across manufacturing, logistics, healthcare, construction, and eventually our homes. At the same time, labor shortages, aging populations, and advances in AI, robotics, and computing are all coming together.
That’s why investors believe Physical AI has the potential to become one of the biggest technology shifts over the next decade. That’s also why companies like Ropedia are attracting investor interest. We focus on one of the most fundamental pieces of the Physical AI stack — the data infrastructure layer. As robots become more capable, the ability to collect, process, and scale high-quality real-world data becomes increasingly valuable.
Is data, not hardware, going to determine the winners of the robotics race?
I don’t think it’s data versus hardware. The winners will need both, but the bottleneck is shifting. Hardware determines what a robot can physically do. Data determines how many useful things it actually knows how to do. That’s exactly where Ropedia comes in. We’re building the data infrastructure layer for Physical AI, helping robots learn from diverse, high-quality real-world data so they can handle a much broader range of behaviors, tasks, and environments.
The earlier $8 million got Ropedia to this point, with $30 million in total. Now you’re scaling manufacturing, data collection, and R&D all at the same time. Can you share more details about your future plan?
Although manufacturing, data collection, and R&D look like three different priorities, they’re really one compounding data engine. As we scale production of our wearable capture device, HOMIE, we can scale our data collection operations. More high-quality data feeds our annotation pipeline and our research on perception and robot learning.
Those research results then help us design better hardware and build more efficient data collection systems. Every part strengthens the others, so it’s not three separate businesses, it’s one ecosystem that keeps improving itself.
Looking ahead, our roadmap is quite focused: better hardware, scaled manufacturing, more sensing modalities, and higher-quality data annotation. Everything we build is aimed at strengthening the data infrastructure layer for Physical AI.
You’ve described how robots can move from factories into homes and workplaces. How do you see Physical AI affecting employment?
I believe Physical AI will complement people rather than replace them. It can help address labor shortages in jobs that are repetitive or difficult to staff, while also creating new careers in robot operations, maintenance, data collection, and training. As the technology matures, we’ll see entirely new professions emerge around it.