Platform
From robot model to deploy-ready policy. One integrated environment for simulation, training, evaluation, and export.
Model Studio
Drop any URDF, MJCF, or SDF robot model into the platform. The Model Studio automatically parses the kinematic chain, identifies joints and actuators, and generates an interactive 3D view.
Every revolute and prismatic joint rendered with range-of-motion indicators and real-time angle readouts.
Forward and inverse kinematics computed on import. Full Jacobian matrix available for each end-effector.
Toggle collision geometry overlays. Detect self-collisions before they become training artifacts.
Torque curves, velocity limits, and gear ratios extracted and displayed per-joint. Ready for simulation.
Interactive joint control
Training Engine
Explore how hyperparameters shape the training curve. The reward autopilot tunes these automatically — here you can see why each parameter matters.
Hyperparameter explorer
The training engine runs thousands of parallel physics simulations on GPU. Each simulation instance generates experience data that feeds the policy optimizer in real time.
Controls the step size of gradient updates. Too high causes instability; too low slows convergence. The autopilot uses a cosine annealing schedule.
Number of experiences per gradient step. Larger batches reduce variance but require more memory. Automatically scaled to hardware.
Encourages exploration by penalizing policy certainty. High entropy prevents premature convergence but slows final performance.
Architecture
Four layers from browser to bare metal. Data flows from user interaction through API orchestration into GPU training clusters and back.
System architecture
Pipeline
GPU-parallel RL with autonomous reward iteration and multi-seed evaluation
Aggregate scoring across seeds with safety metrics and video rendering
ONNX, PyTorch, TFLite — versioned artifacts with metadata and checkpoints
Push to real hardware with domain randomization for sim-to-real transfer
Get access to the full platform. Import your robot, train a policy, and deploy — all from your browser.