Ropedia Secures $22 Million Pre-Series A to Scale Physical AI Data Infrastructure

The burgeoning field of physical AI, which enables robots to interact with the real world, is facing a critical data bottleneck. Singapore-based Ropedia has raised $22 million in Pre-Series A funding to address this challenge by scaling its unique data infrastructure for robotics and embodied AI systems.
The development of advanced robotics and embodied AI systems hinges on access to high-quality, real-world interaction data. Traditional data collection methods are often prohibitively expensive and lack the nuanced, first-person perspective essential for training intelligent agents to navigate complex physical environments. Ropedia's approach aims to democratize this crucial data, enabling faster progress in robotics.
The company's core offering revolves around its proprietary wearable device, HOMIE (Human-centric Omni Interaction and Experience). This head-mounted rig captures synchronized egocentric video, depth, motion, and audio as humans perform tasks, providing a rich, multimodal dataset. Ropedia then processes and annotates this raw data, delivering model-ready datasets that are specifically designed for training physical AI models.
This full-stack methodology differentiates Ropedia from conventional data labeling services, which typically work with pre-collected data. By generating its own data from real human experiences, Ropedia claims to reduce data collection costs by up to 50 times. The $22 million Pre-Series A round, bringing total funding to $30 million, will fuel the expansion of HOMIE manufacturing, global data collection efforts across Southeast Asia and North America, and further AI research into data foundation models.
Founded in 2025 by CEO Zhaoxi Chen, CTO Fangzhou Hong, and Chief Scientist Ziwei Liu, Ropedia currently serves over 20 robotics and foundation model companies. The investment underscores a growing recognition that the 'ChatGPT moment' for robotics will require massive, high-fidelity real-world data, moving beyond synthetic environments or limited datasets. This funding positions Ropedia as a key enabler for the next generation of intelligent robots.
INTELLIGENCE BRIEF
WHY IT MATTERS
The ability of physical AI systems to operate effectively in complex, unstructured environments depends heavily on the quality and quantity of their training data. Ropedia's innovation in capturing and processing human-centric interaction data could accelerate the deployment of robots from controlled factory settings into diverse real-world applications, from homes to logistics. This is a crucial step towards making embodied AI pervasive.
WHO IS INVOLVED
Ropedia (Singapore-based startup); Zhaoxi Chen (CEO and Co-founder of Ropedia); Fangzhou Hong (CTO and Co-founder of Ropedia); Ziwei Liu (Chief Scientist and Co-founder of Ropedia); Venture investors with experience in AI, deep technology, and infrastructure in Southeast Asia; Angel investors connected to Google, a16z, NVIDIA, and Amazon.
MARKET IMPACT
Ropedia's funding signals a maturing market for specialized data infrastructure tailored for physical AI. As more companies develop embodied AI, the demand for high-fidelity, real-world data will intensify, potentially creating a new category of 'data-as-a-service' providers for robotics. This could significantly lower the barrier to entry for developing sophisticated robotic applications.
This story was drafted with AI assistance and reviewed by TurkSpark editors before publication. Facts, figures, and names may be inaccurate — verify important details independently.


