Two founders, honestabout what's real.
Pre-alpha. Two founders (TUM / LMU). The design is ahead of the code — and we say which is which.
Robots are tested for the happy path in simulation, then discover the nasty edge cases in the field — where failures are slow to reproduce, expensive, and sometimes dangerous.
The autonomous-vehicle world already solved pre-deployment testing with adversarial scenario testing. General robotics — humanoids, AMRs, arms, industrial autonomy — has no equivalent, and its tooling is fragmented across Gazebo, Isaac and ROS 2.
We add the adversarial layer, the independent oracle, and the failure-pattern flywheel on top of the stack you already run. The engine is open source; we sell scale, judgment and memory.
- availableExists today — docs, contracts, scaffolding, dev container.
- plannedDesigned and on the roadmap, not implemented yet.
- experimentalBeing prototyped; expect it to change.
Two builders, honestabout what's real.
AI systems & agentic LLM workflows. Builds the fault-injection engine and the independent oracle.
LinkedIn →
Aerospace & CAD at TUM. Builds the scenario format and robot/sim integration.
LinkedIn →