NomadicML Inc. said today it has raised $8.4 million in seed funding to address the critical video data management headaches that are unique to autonomous robots and vehicles.

The round was led by TQ Ventures and saw participation from Pear VC, with Google LLC artificial intelligence wizard Jeff Dean and a number of executives from OpenAI Group PBC and Google’s DeepMind investing as angels.

The startup has developed a platform that aims to help robotics and autonomous vehicle developers search through the massive volumes of video data generated by their fleets.

Such companies are racing to develop “physical” artificial intelligence models that will enable robots and cars to operate in real-world environments with much greater autonomy than they currently do now. But there’s a problem that isn’t easily solved. A single car or robot can generate terabytes of data each day, and for those operating fleets that number in the thousands, that means they’re swimming in petabytes of video footage. This unstructured data is generally archived, and most of it sits there untouched, because it simply isn’t possible for humans physically to review and label it all.

The inability to take advantage of this wealth of video data is holding back model development. For instance, a model that operates an autonomous car needs more data for critical edge cases, such as navigating through a construction zone in rainy weather, to improve its performance.

The company will likely have hours of video footage of its vehicles doing this in the real world, but the challenge is finding it among those petabytes of archived files. It’s an old, worn-out adage, but it really is like searching for a needle in a haystack.

Co-founder and Chief Executive Mustafa Bal (pictured, left) said finding this critical data is especially difficult for startups. “Teams are sitting on a goldmine of video and sensor data, but most of it never becomes a usable training signal,” he said.

Nomadic’s solution is a “visual data engine” that transforms hours of raw, unorganized video footage into a searchable library of production-ready AI training data. It relies on advanced computer vision models to do this.

Those models power autonomous teams of agents that literally watch the video on behalf of companies, cataloging everything so it can be found easily. Engineers can then query their video archives in natural language to find relevant material in a matter of seconds, as opposed to spending months trawling through their archives.

Chief Technology Officer Varun Krishnan (right) said the system is more than just a data labeling tool. He describes it as an “agentic reasoning system” that’s designed to figure out the context and actions taking place within a video, so it can surface that footage when a human tries to describe it. This is what enables it to support more complex queries, such as every single incident where a robot’s gripper interacts with a specific type of object.

The funding will enable Nomadic to scale its platform and meet the growing demand from robotics teams that have found themselves drowning in data archives.

According to TQ Ventures co-founder Andrew Marks, Nomadic is building a vital capability that will help determine who rules the physical AI world. “Physical AI is going to be won by the teams that can learn fastest from the real world,” he said. “Nomadic gives robotics and AV builders the most actionable way to understand their data.”

Photo: Nomadic