--- title: "microagi Raises Germany's Biggest Seed Round to Teach Factory Robots Real Jobs" subtitle: "microagi raises a record $55M seed round to teach factory robots by filming human workers — no robots or models required" url: "https://alphasignal.ai/news/microagi-raises-germany-s-biggest-seed-round-to-teach-factory-robots-real-jobs" author: "AlphaSignal Newsroom" author_url: "https://alphasignal.ai/editorial-team" company: "microagi" company_url: "https://alphasignal.ai/company/microagi" published: "2026-07-16T08:08:20.000Z" updated: "2026-07-16T12:26:02.840Z" type: "News" topics: - "Business" - "Robotics" subtopics: - "Manipulation" - "Vla Models" upvotes: 275 read_time: "6 min" source_url: "https://x.com/microagi/status/2077666630507594022" comments: 0 --- # microagi Raises Germany's Biggest Seed Round to Teach Factory Robots Real Jobs > microagi raises a record $55M seed round to teach factory robots by filming human workers — no robots or models required [microagi](https://x.com/microagi/status/2077666630507594022), a Munich-based startup founded just ten months ago, has closed a **$55 million seed round** , what it claims is the largest seed round ever raised by a German company. The round was led by Hummingbird, with participation from Northzone, LocalGlobe, Village Global, and redalpine. The valuation was not disclosed. The founder's backstory is almost too cinematic to be real. Bercan Kilic landed his dream job as an aerodynamics engineer for Red Bull Racing in 2023, joining the Formula One team at the height of its dominance , but the thrill quickly wore off. He wanted to put his skills toward something bigger: deploying robots in factories to bring about a world of abundant, low-cost goods and services. So he left F1 and started microagi in a Munich hacker house. ## What microagi actually does microagi helps manufacturers train robots for specific jobs. Rather than building its own robots or AI models, the company records workers using cameras and sensor-equipped gloves, then uses that footage to teach existing robotics models how to perform specific tasks inside a customer's factory. Think of it as the integration layer between frontier robot AI and the messy reality of a factory floor. The business model is a two-sided flywheel: - **For AI labs:** microagi provides the labs with data, the labs provide models, and then microagi layers on proprietary customer data to make its customers happy. - **For manufacturers:** microagi adapts its partners' models using each customer's factory data, then helps deploy them on robots. Its customers span the automotive, logistics, and food industries. Kilic declined to name microagi's model partners, though companies building this type of technology include Physical Intelligence, Skild AI, and Generalist AI. The platform is hardware-agnostic, meaning it can work with different robot bodies. ## The hidden arm that went viral You may already know microagi without knowing it. Its consumer-facing arm, **shift**, went viral for offering free apartment cleanings in New York in exchange for recording first-person footage of cleaners doing dishes, mopping floors, and folding laundry. This week, shift began offering free private chefs in San Francisco. This is not a marketing stunt. It is the core of microagi's data strategy. microagi originally planned to do deployment only, until the team found that most robotics models were not good enough to start from. Kilic's analogy is training a new hire: an adult picks up a factory job in a week, a child may never get it , and existing robotics models are still children. So they built shift to generate the training data the field desperately needs. shift now operates in 15 countries and pays more than 20,000 people to record themselves performing physical tasks, selling the footage to the labs building robot brains. It competes with Scale AI, Turing, and micro1, all doing versions of the same thing. ## The data gap nobody talks about Large language models were trained on vast amounts of text and images from the internet, but no comparable trove exists for robots. UC Berkeley roboticist Ken Goldberg has called this the "100,000-year data gap," referring to the disparity between the training data available to robots and chatbots. microagi is betting that whoever closes that gap first wins the physical AI era. The macroeconomic tailwinds are real and urgent. China installed 295,000 factory robots in 2024, accounting for 54% of the global total, compared with 34,200 in the US, according to the International Federation of Robotics. Meanwhile, the EU's median age hit 44.9 in 2025, up from 39.6 two decades earlier, and the European Commission estimates the bloc could lose 18.8 million workers by 2050. US manufacturers are projected to leave 2.1 million jobs unfilled by 2030. ## Chasing a GPT-3.5 moment Kilic believes robotics is approaching its "GPT-2 moment" , for language models, that was the point when researchers found that adding more data and computing power could produce predictable improvements. Robotics companies are still working out that recipe, but he believes the industry is not far from its "GPT-3.5 moment," when the technology becomes useful enough for broad deployment. Nvidia CEO Jensen Huang echoed this framing at CES, declaring "the ChatGPT moment for physical AI is here." Over the last two years, about $20 billion has been invested in physical AI, with applications spanning warehouses, logistics, trucking, construction, aviation, and defense. Citi said the winners in physical AI will likely be firms that own proprietary real-world data, solve specific labor bottlenecks, and use Robotics-as-a-Service models to reduce upfront costs for customers. microagi is explicitly targeting all three. ## Who wins, who loses, and what's at stake The team behind microagi is stacked with domain credibility. Kilic was an aerodynamics engineer at Red Bull Racing. His four cofounders include a former Mercedes F1 engineer, a former Alan Turing Institute researcher, an RWTH Aachen engineer, and a serial entrepreneur. Hummingbird managing partner Firat Ileri, who led the investment, said he was struck by microagi's intensity during a visit to its Munich headquarters, where employees rarely seemed to leave the office. Ileri said microagi stood out because of the founders' ambition and the startup's focus on automating European factories as the continent's workforce ages and shrinks. The competitive picture is nuanced. microagi is not competing with robot hardware makers like Boston Dynamics or Figure AI , it needs them to succeed. Its real competition is in the intelligence and integration layer: - **Data collection:** Scale AI, Turing, micro1 (all collecting human task data) - **Model deployment:** Any system integrator or robotics software company that helps manufacturers adopt robots - **Foundation models:** Physical Intelligence, Skild AI, Generalist AI (microagi's partners, but potentially future competitors) The conference reinforced the view that physical AI is a decade-long buildout, with durable value accruing to companies that own the data flywheel, solve real deployment problems, and meet the highest safety standards. microagi's bet is that the data flywheel , built through shift , becomes a durable moat that model providers cannot easily replicate. ## What's real, what's ambition The gap between the vision and today's reality is worth noting. Kilic told Business Insider that five companies are collecting data through its platform, with one preparing to deploy robots in a factory. microagi will use the new funding to pay for the computing power needed to train robotics models, expand shift's data collection network, and grow its US presence from New York. microagi employs 37 people globally, while shift has about 75. The ambition, however, is not modest. "In five years, if we haven't deployed more than 20 million or 30 million robots, it's a big failure," Kilic said. That is a long way from one customer on the verge of a factory floor deployment , but it is exactly the kind of conviction that raises $55 million ten months after founding. The physical AI race is on, and microagi is betting that the team who teaches the robots wins. ## Sources - [X post by @microagi](https://x.com/microagi/status/2077666630507594022) - [A Red Bull engineer got bored with Formula One. His robotics startup just raised $55 million.](https://dnyuz.com/2026/07/16/a-red-bull-engineer-got-bored-with-formula-one-his-robotics-startup-just-raised-55-million/) - [An F1 aerodynamicist just raised $55m to teach factory robots, using footage of people doing chores](https://thenextweb.com/news/microagi-55m-seed-robot-training-data-shift) - [Physical AI Steps in as Global Workforce Shrinks](https://www.pymnts.com/news/artificial-intelligence/2026/physical-ai-steps-in-as-global-workforce-shrinks/) - [2026 Robotics Update: The Physical AI Ecosystem](https://etfdb.com/artificial-intelligence-content-hub/physical-ai-ecosystem/) - [Ramping Up: Current State Of Robotics In 2026](https://www.technocracy.news/ramping-up-current-state-of-robotics-in-2026/)