When Bercan Kilic walked away from the high-octane world of Red Bull Racing, he wasn’t just chasing a new hobby; he was answering a call to apply the precision of aerodynamics to a far more pressing human challenge: making factories smarter, faster, and more humane.
His decision to launch microagi, a Munich-based robotics venture, culminated in a staggering $55 million seed round—the largest ever recorded for a German startup at that stage. Backed by Hummingbird, Northzone, LocalGlobe, Village Global, and redalpine, the infusion signals strong investor confidence that the next wave of AI will move beyond chatbots into the physical realm of assembly lines and warehouses.
This round arrives at a moment when falling sensor costs, matured machine-learning frameworks, and urgent demand for reshored supply chains converge to create fertile ground for practical robotics. Kilic’s journey from the pit lane to the factory floor exemplifies how elite engineering talent can be redirected toward solving systemic economic bottlenecks.
Microagi’s core innovation lies not in fabricating new robotic arms but in curating a high-fidelity data pipeline that turns ordinary workers into teachers for machines. By equipping employees with synchronized video cameras and sensor‑laden gloves, the startup captures the nuanced kinematics and tactile feedback involved in tasks ranging from intricate electronic assembly to bulk material handling.
To fuel its data engine, microagi launched Shift, a consumer‑facing initiative that flips the traditional data‑collection model on its head. Rather than paying factories to record their own processes, Shift offers complimentary services—such as apartment cleanings in New York or private chef experiences in San Francisco—in exchange for users willingly recording themselves performing everyday chores.
This gamified approach has yielded impressive scale: Shift now operates in fifteen countries and compensates more than twenty thousand contributors for capturing a diverse library of human motion. The resulting footage spans a wide spectrum of domestic activities, from dishwashing and floor mopping to laundry folding and meal preparation, providing a rich testbed for teaching robots general manipulation skills.
Early experiments revealed a sobering truth: many off‑the‑shelf robotics models available today behave more like novices than seasoned technicians when confronted with unfamiliar tasks. Kilic likens this gap to attempting to teach a complex trade to a child who lacks the foundational cognitive scaffolding; no amount of demonstration will yield mastery if the underlying model cannot abstract and generalize from the examples.
This insight forced microagi to pivot from a pure deployment play to an active role in shaping the training process itself. The company now works closely with its model‑partner labs, supplying them with curated, factory‑specific demonstration streams that enable the base models to acquire the fine‑grained control and adaptive responsiveness required for high‑mix, low‑volume production.
Microagi’s revenue model hinges on a symbiotic partnership structure that separates data provision from model generation while adding a proprietary layer of customization. The startup supplies its partners with high‑quality, task‑specific demonstration footage harvested through Shift and direct factory collaborations. In return, those partners provide microagi with access to their state‑of‑the‑art base models, which microagi then fine‑tunes with the customer‑specific data streams to adapt the model’s weights and biases to the unique dynamics of a given factory layout, tooling, and product mix.
The resulting customized model is deployed on the client’s existing robotic hardware, with microagi offering integration support, safety validation, and performance monitoring. This approach allows microagi to avoid the capital‑intensive burden of manufacturing robots or inventing new algorithms from scratch, while still capturing value through service fees, licensing arrangements, or outcome‑based contracts tied to productivity gains.
The urgency behind microagi’s mission is amplified by demographic trends that are eroding the traditional labor base upon which Western manufacturing has depended. According to Eurostat, the median age of the European Union’s population reached 44.9 years in 2025, a significant increase from 39.6 just two decades prior. This aging trajectory implies that a growing share of the workforce is approaching retirement, while the inflow of younger entrants has not kept pace—a imbalance reflected in the European Commission’s projection that the bloc could lose as many as 18.8 million workers by 2050.
While Europe wrestles with an aging workforce, the global landscape of industrial robotics reveals a stark asymmetry that further underscores the stakes for Western competitiveness. Data from the International Federation of Robotics shows that China deployed a staggering 295,000 factory robots in 2024, accounting for over half of the worldwide total. In contrast, the United States installed merely 34,200 units during the same period—a figure that highlights a substantial gap in automation intensity between the two economic superpowers.