The San Francisco-based company announced it is developing a new two-armed robot system designed to assemble meals at prep tables, tackling tasks like building burgers, burritos, and other made-to-order dishes that are still largely done by hand.
Unlike Chef Robotics’ existing systems, which focus on high-volume food production in manufacturing facilities, the new platform is aimed at lower-volume environments where a single worker typically handles the entire assembly process. That includes ghost kitchens, fast-casual restaurants, airline catering operations, schools, hospitals, military facilities, prisons, stadiums, corporate cafeterias, and hotels.
To handle these more complex jobs, the system will use two robotic arms working together. The setup is designed to mimic the coordinated movements of human arms and hands, allowing the robot to manipulate ingredients and utensils with greater dexterity.
The new platform will be powered by Chef Robotics’ Food Foundation Model, an AI system trained specifically for food manipulation.
According to the company, traditional vision-language-action models are not well suited for handling food because ingredients are often sticky, wet, irregular, and constantly changing in shape. Chef’s model is designed to understand these variables and perform a wide range of tasks using a single AI system.
That includes identifying trays and compartments, picking up ingredients, scooping food, and assembling complete meals. The model learns through demonstration, meaning workers can show the robot how to build a specific dish rather than manually programming each movement.
Chef Robotics says this approach will eventually allow robots to adapt to new ingredients with minimal additional training and improve their performance over time.
The company is also building the system specifically for commercial kitchens. It will be food-safe, washable, and able to operate in different temperature and humidity conditions. The robot is intended to work alongside human employees and can be controlled using natural language prompts.
“We started Chef by focusing on high-throughput food manufacturing, but a large part of the industry still relies on manual prep table assembly,” said founder and CEO Rajat Bhageria. “With this new physical AI system and our Food Foundation Model, we will extend physical AI to handle those real-world conditions and unlock a much broader set of applications in the food industry.”
Chef Robotics says its technology has already helped produce more than 100 million servings in commercial operations.
Join the Discussion