Singapore Physical AI Startup Ropedia Raises US2M for Embodied AI Data Infrastructure

Singapore's startup ecosystem is proving that the most interesting AI companies aren't always the ones building the next chatbot. While the world's attention has been fixed on generative models and reasoning agents, a quieter — but potentially more transformative — category of AI infrastructure has been taking shape in the city-state's research labs and early-stage studios. Physical AI, the branch of artificial intelligence that deals with robots, autonomous systems, and machines that interact with the real world, requires fundamentally different kinds of training data than the text and images that power large language models. And a Singapore startup born out of NTU is building the data pipeline that the embodied AI industry desperately needs.

Ropedia, a Singapore-based company founded in 2025, announced on July 23 that it has raised a US2 million pre-Series A round, bringing its total funding to US0 million. The company builds wearable hardware and software that captures first-person human experience data — video, depth, motion, and audio — from the same physical viewpoint that a robot would operate in. This egocentric data is critical for training embodied AI systems, vision-language-action models, and robotic agents that need to understand and navigate real-world environments. The startup plans to use the new capital to expand data collection operations across Southeast Asia and North America, grow its teams in Singapore and the US, and scale manufacturing of its wearable capture devices.

The funding arrives at a moment when the physical AI sector is accelerating rapidly. Egocentric datasets — which align video, depth, motion, and audio from a single wearable device — are becoming the gold standard for training robots that can operate in unstructured human environments. Unlike synthetic data or third-person video recordings, first-person data captures the exact sensory inputs a robot would encounter when performing tasks like cooking, assembling components, or navigating crowded spaces. Ropedia's approach of combining custom hardware with annotation tools, quality analytics, and compliance infrastructure positions it as a full-stack data partner for companies building everything from warehouse robots to household assistants.

What makes Ropedia particularly interesting for Singapore's tech ecosystem is its deep ties to local research talent. Co-founder Ziwei Liu holds a faculty position at Nanyang Technological University and has authored more than 50 papers across computer vision, graphics, machine learning, and robotics. The startup's earlier backers include investors and angel investors connected to Google, a16z, Nvidia, and Amazon — signalling that global tech heavyweights are watching Singapore's physical AI space closely. The company's decision to base its hardware manufacturing and data collection expansion in both Southeast Asia and North America also reflects Singapore's growing role as a bridge between Asian and Western AI supply chains.

Why it matters for Singapore: Physical AI represents one of the highest-value opportunities in the next wave of artificial intelligence, and Singapore is quietly building the infrastructure to support it. Ropedia's funding round demonstrates that Singapore-based startups can attract significant international capital for hard-tech AI ventures — not just software and services. For the broader ecosystem, every successful physical AI company that emerges from Singapore strengthens the city-state's position as a serious player in embodied AI, robotics, and advanced manufacturing, sectors where the government has been strategically investing through initiatives like the National Robotics Programme and the Singapore Smart Nation ambition. The question is no longer whether physical AI will transform industries, but which hubs will supply the data that makes it work.