A former Formula One aerodynamics specialist stepped away from high-speed racing to pursue a grander goal: enabling robots to handle complex tasks on factory floors around the world.

His company, microagi, based in Munich, has now closed a $55 million seed round, the biggest such funding for any German startup to date. Hummingbird led the investment, joined by Northzone, LocalGlobe, Village Global, and redalpine. The firm chose not to reveal its valuation.

Bercan Kilic joined Red Bull Racing in 2023 during a period of strong performance for the team. Although the technical work proved impressive, he soon sought a larger impact and turned his attention to robotics that could support widespread production of affordable goods.

microagi records factory employees using cameras and sensor gloves to capture detailed movements. The resulting data trains existing robot systems for precise operations at client sites without developing hardware or base models in-house.

Five manufacturers currently gather information through the platform, and one is preparing its first robot deployment on the production line.

While microagi may be new to many, its consumer service shift gained attention earlier this year by providing complimentary home cleanings in New York in return for first-person video of routine chores. The offering has now expanded to free private chef experiences in San Francisco.

The funding round reflects growing investor interest in robotics and physical AI, fueled by advances beyond language models, lower hardware expenses, workforce gaps, and efforts to bring supply chains closer to home.

Kilic views the technology as a solution to labor shortages rather than a threat to workers. He points to Europe and the United States needing automation to match progress elsewhere, noting that China installed 295,000 industrial robots last year, representing 54 percent of the worldwide total, while the United States added 34,200.

"If you run factories, the math is already on your desk," Kilic said. "Your most experienced people retire this decade, and their replacements were never born. Reshoring only works if the robots do."

Why microagi created shift

The company began last year in a Munich hacker house with the aim of moving robotic systems from labs into real workplaces. Its name signals the belief that AI models must shrink in size to support deployment across millions of machines, as larger systems would demand excessive computing resources.

microagi initially intended to focus only on adapting and installing robots using customer data. However, the team discovered that current models lacked sufficient capability as a foundation for practical use.

Kilic compared the situation to teaching new employees: an experienced adult might master a role quickly, but a child could struggle indefinitely despite clear guidance. Existing robotics models remain closer to the latter stage, he explained.

This led to the launch of shift, which now operates in 15 countries and compensates more than 20,000 individuals to record physical activities. The footage is sold to AI and robotics developers working on machine learning systems.

Although microagi does not name its model partners, it adapts offerings from various labs with proprietary factory data before assisting with robot integration. Clients come from automotive, logistics, and food sectors.

"We provide the labs with data, they provide us with models, and then we layer on proprietary data to make our customers happy," Kilic said.

shift addresses a major challenge in the field. While language models drew from enormous online text and image collections, robotics lacks equivalent resources. UC Berkeley roboticist Ken Goldberg has described this as the "100,000-year data gap."

The service competes with firms such as Scale AI, Turing, and micro1 that similarly gather task recordings from paid contributors.

Chasing a ChatGPT moment

With the new capital, microagi plans to invest in computing resources for model training, broaden shift's global data network, and strengthen its operations in the United States starting from New York.

The company has 37 employees worldwide, while shift employs around 75, according to Kilic.

He sees the industry nearing its "GPT-2 moment," when scaling data and compute began yielding reliable gains for language systems. Robotics teams are still refining the approach, yet Kilic expects a "GPT-3.5 moment" soon, when the technology reaches practical utility for widespread adoption.

Hummingbird managing partner Firat Ileri, who led the round, highlighted the team's dedication after visiting the Munich office. Hummingbird previously backed Lovable, Kraken, and Etched.

Ileri was drawn to the founders' drive and the emphasis on automating European manufacturing amid an aging population. The EU median age hit 44.9 in 2025, up from 39.6 twenty years prior, per Eurostat data. The European Commission projects a potential workforce reduction of up to 18.8 million people by 2050.

Kilic's cofounders include former Mercedes F1 engineer Yoan Iliev, ex-Alan Turing Institute researcher Anton Poletaev, RWTH Aachen engineer Nico Nussbaum, and entrepreneur Artjem Weissbeck.

The group has set an ambitious target of becoming the largest company globally within five years by powering tens of millions of robots.

"In five years, if we haven't deployed more than 20 million or 30 million robots, it's a big failure," Kilic said.