We’re building tools for the next era of physical automation.
Munich
Joined March 2025
- Tweets17
- Following4
- Followers1K
- Likes16
The Robot Learning Company retweeted
Weekend project completed: light box upgrade for transferable policies; 💙 to @RobotLearningCo for sharing the BOM. My hope is that this will allow reliable deployment on different systems with similar setup - towards a 🤖 DK-1 policy AppStore
The Robot Learning Company retweeted
I 💙 my @RobotLearningCo DK-1 - today first rollouts of training from Wan 2.2 + ActionExpert - no VLA base .. with ~3h teleop demos. Currently training on 2 more hours I recorded today. Many things to improve - but feels like 🪄 magic. Super 😊 right now.
The Robot Learning Company retweeted
New control stack for TRLC-DK1:
Joint-space PD + gravity compensation
Before vs. after:
The Robot Learning Company retweeted
TRLC-DK1 v0.3 is out! We did some cool upgrades:
- USB3.0 cams → USB3.1 can easily handle 3x 720p60 MJPEG streams
- LeRobot plugin support → it will auto-detect the trlc_dk1 package
- More robust mechanical parts & electronics
Full changelog: github.com/robot-learning-co…
The Robot Learning Company retweeted
We (@FeduniakS, @PashaRizali) have been basically living in the lab working non-stop to push physical AI fordward. We are building a robotics stack of the future for continual and reliable deployment of physical AI systems.
Thanks a lot @JannikGrothusen for lending us the arms!
The Robot Learning Company retweeted
Data collection while the office above having "drumming team building" is a test of patience.
The Robot Learning Company retweeted
`crazy_random_light` is one of the domain randomization options in RoboTwin 2.
The clip features DK1 arms which I recently ordered from TRLC (@JannikGrothusen). I now created a URDF for synth data collection and visualization.
The Robot Learning Company retweeted
Thanks for delivering the robots safe n sound @JannikGrothusen
Check them out at @RobotLearningCo 🤖
The Robot Learning Company retweeted
Today I fixed the follower arms to a profile bar and started experimenting with new leader–follower setups.
This speaker desk I found on Amazon is the first of three.
I like the laptop‘s position in the bottom tray: Doesn’t block view and still accessible for quick code changes.
What I still haven’t figured out is the ideal horizontal distance between both the leader and follower arms to make teleop feel more natural.
The Robot Learning Company retweeted
One of the most affordable and high quality bimanual platform out there ;)
The Robot Learning Company retweeted
Introducing TRLC-DK1-X: An accessible humanoid data collection platform.
It features two 6+1 DOF leader-follower arm pairs and three ultra-wide FOV cameras, integrated for learning static, bimanual manipulation tasks.
Order now for $4,999.
The Robot Learning Company retweeted
A small but powerful modification to the GELLO handle: mounting it at a 45° angle allows both the robot and the operator to be in a natural position while collecting data.
.STL and .STEP in GitHub repo (🧵)
The Robot Learning Company retweeted
I built an open source dev kit for AI-native robotics:
- it’s all-in-one (includes leader arm and cameras)
- allows to deploy autonomous policies in under a day
- can now be ordered for $2,999
🧵 (1/n)
The Robot Learning Company retweeted
If you're using @trossenrobotics or @ARX_Zhang arms, you shouldn't buy the expensive leader variants.
We're open-sourcing our leader design, which costs under $200 and is kinematically equivalent to both arms at 60% scale.
Designed by the one and only @KasparJanssen
The Robot Learning Company retweeted
The Robot Learning Company (@RobotLearningCo) makes automation accessible beyond factories by building affordable AI-powered robot arms to automate repetitive, stationary tasks.
Congrats on the launch, @JannikGrothusen!
ycombinator.com/launches/Nil…