Glad to see Axis inspiring the next generation to build in the physical world. Great things start small.

We are now open-sourcing the AxisDataCleaning pipeline. Github repo: github.com/AxisAIOrg/AxisDat… Browser teleoperation is one of the most scalable paths for robot data generation. Raw human input, however, is not yet model-ready: ▪️ Idle pauses ▪️ Micro-jitters ▪️ Low & Variable frame rates Raw web data alone is not enough for reliable policy training. Here is how our backend turns noisy human demonstrations into usable trajectories for downstream policy training. 🧵👇

Domain Randomization (DR) is a key component of the data augmentation pipeline at Axis Robotics. By applying DR, we are able to scale verified, high-quality human trajectories by 10x to 100x. During training, we systematically introduce variances in environmental parameters. This prevents the model from relying on spurious visual correlations. The objective is to ensure the policy learns rather than overfitting. To demonstrate the necessity and effectiveness of this approach, we evaluated both DR and No-DR models on Task 74 (pour_water_into_mug). The empirical results show a definitive impact on real-world deployment reliability: integrating DR into the pipeline increased the success rate from 0% to 90% (Fig. 1). This divergence stems from how the respective policies process visual observations (Fig. 2). The baseline (No DR) model overfits to the static visual background. It essentially memorizes the poses from the training dataset but fails to generalize when subjected to the inevitable variances of real-world deployment. Consequently, it cannot execute the correct manipulation on the target object. Conversely, the DR-trained model learns to extract essential geometric features and physical constraints, filtering out superficial visual noise. This leads to significantly higher robustness in dynamic environments. The structural difference in execution is clearly reflected in the end-effector trajectory data: These real-world deployment recordings further illustrate this difference (Videos 1 and 2). Scaling Physical AI requires turning raw trajectory data into robust policies, and a rigorously engineered DR infrastructure is an essential bridge to close the Sim2Real gap.

Update: We have added two new training tasks: 'Hang the Hanger' and 'Straight Row Arrangement'. These training tasks are designed to help you become fully comfortable with the control panel. You can use them to practice when no regular training tasks are available. Please note that these training tasks can be completed an unlimited number of times and will not earn any future rewards.

Axis is on @base Discovery hosted by @cityprotocolHQ Drop by our page on @baseapp, dive into robotics intelligence, and test our product — that’s your ticket to the baseapp reward pool.

Another educational visual from my @PrismaXai series. Human operators help robots complete real world tasks, and those interactions provide valuable data for future AI improvements.

Axis Community AMA: Enter the Machine Economy What exactly happens to the data you generate? How does a browser click translate to a physical robot's action? We are hosting our first community AMA to dive deep into the Axis pipeline, data infra, and the future of the Machine Economy. 🗓 Date: Tuesday, April 7 ⏰ Time: 12:00 PM - 1:30 PM UTC | 8:00 PM - 9:30 PM SGT 🎙 Hosted by: @Cunn_19 🗣️ Speakers: @chris_anm01 @0xsexybanana @0xRyzzu @dj673285379 @imsharonw @0xveins @iamlogtun Set your reminders below⬇️

🎙️ LIVE AMA: AI x Robotics in the Real World 6 April · 9:00 – 10:00 PM (GMT+7) · Live on X Space We’re bringing together @axisrobotics, @StrikeRobot_ai, @shadowcleague, @NeroX_AI for a deep dive into AI & Robotics technology, real-world business use cases. Hosted and moderated by @base_vietnam ➡️Join here Space: x.com/i/spaces/1kJzDMQjPeDKv 🔗 Set your reminder and come ask anything about AI & Robotics with Base Vietnam!