Skip to main content

​
Step 1: Create Your First Project

Cyberwave is in Private BetaTo get started, sign up for our Private Beta and get a token!

​
Join the Cyberwave Private Beta

We’re inviting a select group of pioneers to test Cyberwave in both virtual and real-world environments. Apply now to secure your spot.

No credit card required • Cancel anytime

Projects help you organize your robotics work by grouping related environments, assets, and simulations together.
  • Click “New Project” from the dashboard
  • Give it a name like “My First Robot Project”
  • Add a description of what you’re building

Go to Dashboard

Create your first project from the dashboard

​
Step 2: Create an Environment

Environments are 3D spaces where your robots live and interact. Think of them as virtual worlds for your simulations.
  • Click “New Environment” from your project
  • Pick a name for your environment

​
Step 3: Add Robot Assets

Browse our catalog of pre-built robots or upload your own URDF files to bring your robots into the simulation.
  • Browse the asset catalog for robots
  • Add robots to your environment
  • Position and configure your robots

Browse Catalog

Explore our robot catalog

​
Step 4: Control with Python SDK

Take full control of your robots using our Python SDK. Write code to move robots, read sensors, and implement complex behaviors. 
# Install the Cyberwave SDK
pip install cyberwave

# Control your robot
import cyberwave as cw

# Connect to your robot
robot = cw.twin("the-robot-studio/so101")

# Move the robot to a position
robot.move(x=1.0, y=0.0, z=0.5)
robot.rotate(yaw=90)

# Actuate the robot's joint to 30 degrees
robot.joints.set("1", 30)

​
Step 5: Record and Create Datasets

Collect training data by recording your robot operations. These datasets will be used to train AI models for autonomous behavior.
  • Switch your environment to Live Mode
  • Turn on the camera to capture video feed
  • Start recording your robot operations
  • Perform the task you want the robot to learn multiple times
  • Stop recording when you have enough demonstrations
Record 10-15 demonstrations of the same task with slight variations for better model training results.

​
Step 6: Export Your Dataset

Process your recordings into structured datasets ready for training.
  • Review your recorded sessions
  • Trim recordings to create episodes (one per task completion)
  • Select quality episodes for your dataset
  • Export the dataset
  • Access your dataset from the Manage Datasets tab

​
Step 7: Train an AI Model

Use your dataset to train a machine learning model that can control your robot autonomously.
  • Navigate to AI → Training in your environment
  • Click New Training
  • Select your exported dataset
  • Choose model architecture (e.g., VLA model)
  • Configure training parameters or use defaults
  • Click Start Training
  • Monitor training progress in the Training tab
# Training happens in the cloud - no local setup required!
# Monitor progress in the Cyberwave dashboard

​
Step 8: Deploy Your Model

Deploy your trained model as a controller policy to enable autonomous robot control.
  • Go to AI → Deployments in your environment
  • Click Deploy Model
  • Select your trained model
  • Give your deployment a name
  • Configure deployment settings
  • Click Deploy

​
Step 9: Use AI to Control Your Robot

Control your robot autonomously using natural language prompts powered by your deployed AI model.
  • Switch to Edit Mode in your environment
  • Click Assign Controller Policy for your robot
  • Select your deployed model
  • Save the configuration
  • Switch to Live Mode
  • Enter a natural language prompt (e.g., “Pick up the object”)
  • Watch your robot execute the task autonomously!
Ensure the workspace is clear and the robot has safe operating space before running autonomous operations.