For many years, university laboratories were the primary sources of significant advancements in robotics. However, with substantial investments flowing into the sector of physical AI, many researchers are now transitioning to the startup environment.
One recent example is Microagi, a startup based in Germany, which has appointed Animesh Garg as its chief research officer. Garg, a professor at Georgia Tech and former Nvidia researcher, is set to spearhead the company's initiatives in refining general-purpose robotics models for specific applications. Alongside Garg, seven members from his research group will join the company as part of an acquisition involving a previously stealth-mode startup that he had been developing. Details about the acquisition were not disclosed by either Garg or Microagi.
This past summer, Microagi successfully raised $55 million, marking it as the largest seed funding round secured by a German startup to date, as reported by Business Insider. The company collaborates with manufacturers across various sectors such as automotive, logistics, and food, focusing on training robots for industrial roles. Its strategy involves using data harvested from human workers to customize robotics models for individual client operations.
Garg's decision to join Microagi coincides with an influx of capital into the realms of robotics and physical AI, driven by technological advancements, decreasing hardware costs, labor shortages, and a growing impetus for automation and reshoring production processes.
The surge in funding is attracting numerous robotics experts to the startup scene. For instance, Deepak Pathak and Abhinav Gupta, both professors at Carnegie Mellon, cofounded Skild AI, which has amassed over $2 billion to develop a universal AI framework for robots. Similarly, Stanford's Chelsea Finn and UC Berkeley's Sergey Levine founded Physical Intelligence, another venture exploring general-purpose robotic models and raising upwards of $1 billion. Additionally, Russ Tedrake from MIT recently launched Walden Robotics after keeping it under wraps for some time.
Garg expressed that he chose to affiliate with Microagi due to a pivotal shift in robotics focus—from theoretical lab experiments to widespread commercial implementation. "The technology has reached a critical juncture," he mentioned in an interview with Business Insider. "We have arrived at a stage where we can establish viable businesses."
With over 15 years of experience in robotics research, Garg holds a Ph.D. from UC Berkeley and has completed a postdoctoral fellowship at Stanford. He has also served in faculty roles at the University of Toronto and Georgia Tech, and was the chief scientific officer at humanoid robot manufacturer Apptronik from 2024 to 2025. At Nvidia, Garg contributed to the development of simulation technologies absorbed into Isaac Lab, the company's open-source platform for robot training. His recent research has revolved around enabling robots to learn from human demonstrations and enhance their capabilities via experience.
This aligns closely with Microagi's operational methods, as the company gathers information from individuals performing various physical tasks, utilizing equipment like cameras and sensor-laden gloves, to train their robots effectively.
Microagi's CEO, Bercan Kilic, shared their ambitious goal of deploying at least 10 million robots for clients over the next five years.
