Blog
Updates on the development of our robot and the company
Planning instead of programming: how our robot breaks down tasks itself
A language model plans, a library of tested skills executes. This lets us bring new tasks to the robot without reprogramming it every time.
Read more
Training data without teleoperation: our Agentic Data Flywheel
The biggest bottleneck in robot learning is training data. Our pipeline generates it on its own and gets better with every round.
Read more
Symbiosika erhält das push!-Stipendium 2026 des Landes Hessen
Symbiosika gehört zu den 22 Start-ups, die in dieser Runde mit dem push!-Stipendium des Hessischen Wirtschaftsministeriums gefördert werden.
Read more
One platform, many robots: our Robot Management System
A robot in a hotel has to be monitored, updated and controlled remotely. We have built our own platform for that.
Read more
First steps: from a remote-controlled to an AI-controlled robot
We have set up a simulation environment, teleoperated robots and moved one with an AI agent for the first time.
Read more
Why we are teaching robots what language models can already do
Starting now, we are bringing our experience with language models into robotics. Why we are doing it and what we have planned for 2026.
Read more