Microagi Names Animesh Garg Chief Research Officer and Acquires His Team’s Stealth Startup

German robotics startup Microagi has named Animesh Garg its chief research officer and bought the stealth startup his research group had been building. Garg, a Georgia Tech professor and former Nvidia researcher, brings seven members of that group with him. Neither side disclosed what Microagi paid.

‍

What Garg Will Run at Microagi

‍

Garg takes over post-training, which means tuning broad robot models until they can handle the jobs in a particular plant. Microagi's raw material for that is human motion: it records employees doing the work, using cameras and sensor-fitted gloves, then trains robots on the recordings. Its early customers are automotive and logistics companies, plus food producers.

‍

The company raised $55 million in seed funding earlier this year. CEO Bercan Kilic has said Microagi wants to help customers deploy at least 10 million robots over the next five years, which for now is a company goal.

‍

From Robot Learning Research to the Factory Floor

‍

Garg's recent research focuses on teaching robots from human demonstrations and letting them improve through practice, which lines up closely with how Microagi collects its training data.

‍

Garg was at Nvidia from 2018 to 2024, and his simulation work there became part of Isaac Lab, the company's open-source platform for training robots virtually. He spent 2024 to 2025 as chief scientific officer at humanoid maker Apptronik and has led the People, AI and Robotics group at Georgia Tech since 2024.

‍

Giving Post-Training a Chief Research Officer

‍

Post-training is the step that stands between a signed customer and robots doing real work, and Microagi has handed it to a research executive with a C-suite title. That choice tells candidates and customers where Microagi thinks its advantage sits, and it gives Garg authority over the part of the business that gates each sale.

‍

Professors who left to found their own robotics companies have drawn some of the largest checks in the field, and Skild AI's latest round valued it above $14 billion. A smaller startup recruiting senior robot-learning researchers one at a time is bidding against that kind of money. Buying a lab's startup got Microagi eight people who already work together, in a single deal.

‍

At German sites with a works council, Microagi's data-collection model could add a step to each deployment. Section 87 of Germany's Works Constitution Act gives works councils co-determination rights over technical systems designed to monitor employee behavior or performance, and cameras and wearable sensors that record workers could fall under it, depending on how the system is set up. For a research chief, that puts employee representatives in the path of the training data.

‍

Kilic's 10 million robot goal puts another hiring challenge outside research. Reaching even a fraction of it would take people who can deploy robots at customer sites and keep fleets running after installation. Microagi has now named the executive responsible for adapting its models to customer work, and field engineering and deployment operations are the roles that grow as those installations do.

‍