Euporia / Robotics

General-purpose robot intelligence - developed on Earth, deployed in orbit

We build software that lets existing robots learn new physical tasks from human demonstrations, handle real-world variation, and adapt with less specialist programming.

See how it works Get in touch

Existing robots. New skills. Less task-specific programming.

How it works

From demonstration to autonomy

Demonstrate

A human demonstrates the task through teleoperation, giving the robot grounded examples of what good behaviour looks like.

Train

The demonstrations become training data for a vision-language-action policy that connects what the robot sees to the right motion.

Deploy

Once trained, the robot repeats the skill autonomously and handles more variation than a rigid scripted workflow.

What we are building

Intelligence that makes robots useful in changing work

Teach through demonstration

Give robots new skills through examples instead of specialist programming.

Adapt to real-world variation

Help existing robots adjust to new tasks, objects, and changing environments.

Automate the learning process

Reduce the human effort required for data collection, training, validation, and deployment.

Why it matters

Automation should adapt when the work changes

Many valuable tasks remain difficult to automate with conventional robot programming. Every new product, object, or workstation adds engineering time and cost, leaving people to perform work that is repetitive, physically demanding, unpleasant, or dangerous.

Learning from demonstrations can make shorter production runs more economical, help manufacturers adapt faster, and extend automation to tasks that are too variable or costly to program conventionally.

Europe faces labour shortages, demographic pressure, and a widening technology gap with the United States and China. Building robotics and physical AI in Europe is essential to technological sovereignty: maintaining industrial capability and economic resilience while ensuring that these technologies reflect European priorities such as well-being, sustainability, and social responsibility.

Factories first / Orbit next

One learning stack. Many robotic bodies.

Industrial deployments on Earth build the software, operational experience, and data needed for more capable robotics. The same learning process can support future inspection, servicing, and assembly in orbit.

Founding team