A robotics startup based in Watertown is building what it calls the “largest robot data factory in the United States.” Inside the newly renovated facility, around 100 humanoid robots are learning how to identify, pick up, and manipulate everyday objects using artificial intelligence and computer vision technology.
The robots, known as Sonny, are equipped with multiple cameras positioned on their heads, chests, and robotic claws. Employees monitor their progress and correct mistakes through smart devices, helping the machines gradually improve their understanding of real-world tasks.
Although the robots are still in the early stages of development, the company believes the technology could eventually support large-scale industrial automation. Intelligence CEO Josh Gruenstein said the robots are being trained to perform complex logistics and warehouse operations for future enterprise customers.
Tutor Intelligence Builds Advanced Robot Learning Platform
Tutor Intelligence recently moved into its new headquarters located in a historic building that once served as a cotton mill and later housed Boston Scientific. The startup’s facility, called Data Factory 1 (DF1), focuses on teaching robots how to learn autonomously through a combination of AI models and human supervision.
The company’s first vision-language-action model, Ti0, powers the robots’ learning process. In addition to onsite employees, remote workers across the United States, Mexico, and the Philippines monitor robot performance and help correct behaviors in real time.
Founded in 2021 by researchers from the Massachusetts Institute of Technology, Tutor Intelligence aims to create scalable robot learning systems capable of performing general-purpose industrial tasks. The startup has already raised $42 million, including a $34 million Series A funding round completed last year.
Cost-Effective Robotics and Cloud-Powered AI Development
Rather than relying heavily on expensive sensors and specialized hardware, Tutor Intelligence uses a cost-effective combination of cameras, AI-powered software, and cloud computing infrastructure. Company executives believe this approach can reduce robotics costs while improving scalability for future deployments.
The startup also participated in a robotics accelerator program backed by Amazon Web Services, NVIDIA, and MassRobotics. Through the program, Tutor received engineering support and $200,000 in AWS cloud credits to accelerate its AI development efforts.
Executives at AWS said Tutor stood out because of its goal to create adaptable robot intelligence capable of functioning across multiple environments. The company connects all of its robots to AWS cloud infrastructure, giving them the computing power required for large-scale training and real-time learning.
Real : Locus Robotics Launches Locus Array for Smart Warehousing
Tutor Expands Real-World Robot Deployments
While the Sonny robots remain in development, Tutor Intelligence has already deployed its industrial robot, Cassie, at manufacturing and logistics facilities across the United States. The company said the 2,000-pound robot performs tasks such as palletizing and case-picking with performance levels comparable to or better than human workers.
Several businesses testing the technology reported improvements in operational efficiency and labor cost savings. BetterBody Foods said its facilities reduced costs by 36% after implementing Tutor’s robots for loading and unloading operations.
Industry leaders believe robotics and automation will continue reshaping warehouses and manufacturing environments over the next decade. Tutor Intelligence hopes its growing robot data factory and AI-powered learning systems will position the company as a major player in the future of industrial automation.
