Neura robotics partners with aws to scale cognitive robots globally

Tuesday 21 April 2026 - 12:00

By: Dakir Madiha

Neura robotics partners with aws to scale cognitive robots globally

NEURA Robotics and Amazon Web Services have announced a strategic partnership to accelerate the development and deployment of physical artificial intelligence, combining NEURA’s cognitive robotics platform with AWS cloud and AI infrastructure. The agreement, unveiled on April 21 at NEURA’s headquarters in Metzingen, aims to address a core limitation in robotics: the lack of real-world training data.

Under the partnership, AWS will serve as NEURA’s primary cloud provider, hosting its Neuraverse platform, a system designed to connect robots, humans and data in a continuous learning network. NEURA Gym, the company’s training environment where robots practice complex tasks alongside high-fidelity simulations, will be integrated into Amazon SageMaker to streamline AI training pipelines and improve scalability.

The collaboration also includes potential deployment of NEURA robots within Amazon logistics centers. These facilities would provide large-scale real-world environments where robots can operate, generate data and refine performance. The data collected would be fed back into the Neuraverse, enabling continuous improvement across NEURA’s robot fleet. Chief executive David Reger said physical AI requires constant validation in real conditions to reach operational maturity.

The agreement builds on NEURA’s broader ecosystem strategy. The company has been forming partnerships across cloud computing, artificial intelligence and semiconductors to support large-scale deployment. Its network already includes industrial players such as Kawasaki, Schaeffler and Bosch, with the goal of deploying millions of cognitive robots by 2030.

AWS said the partnership reflects NEURA’s open platform approach to solving the data gap in robotics. Jason Bennett noted that AWS infrastructure will provide the global foundation required to support real-time intelligence sharing across robotic systems as deployment scales.

The collaboration highlights intensifying competition to move physical AI from controlled environments into commercial use. While large language models benefit from vast datasets sourced from the internet, robots rely on limited real-world training inputs. Access to operational environments such as warehouses could accelerate progress by providing continuous streams of data necessary for perception, reasoning and autonomous action in complex settings.