Canvas
The Canvas is the core interface for designing, configuring, and deploying AI agents within Ardor Cloud. It provides a visual environment where users can build agents by defining their architecture, connecting components, and managing various operational settings.
Key Features
Agent Management
-
Project Selector: Users can create or select an existing project from the dropdown at the top-left.
-
Save Agent: Allows users to save the current state of the AI agent architecture.
Navigation Tabs
At the top, several tabs allow users to manage different aspects of the AI agent:
The main design interface, where users can drag-and-drop components, connect them, and build the agent workflow
The main design interface, where users can drag-and-drop components, connect them, and build the agent workflow
Displays the API URL for interacting with the agent
Provides a history of events and executions for debugging
Configuration settings for different deployment environments
Tools for monitoring agentic solution performance
Manages versions and deployment history
General configuration options for the project
Deploy button: Allows users to deploy the current state of the AI agent.
Architecture Tab
The Architecture tab is the primary workspace where users visually design their AI agent using a drag-and-drop component, copilot or code editor.
Toolbar
-
Component Library: Allow to chose components from the library and drag n drop to the canvas
-
Copilot: Allows to use the copilot to design the agent in a conversational way
-
PRD: Stores product requirements document for the agent that could be generated by the copilot or provided by the user
-
Canvas as a code: Allows to see the agent architecture as a code and edit it manualy
-
Test menu: Contain test cases, and options to dry run agent and run test cases on deployed agent
-
Canvas activity log: Allow to see the history of changes made to the agent architecture
-
Undo/Redo: Helps in reverting or reapplying changes
-
Grid: Fits all components on the canvas to the screen
-
Zoom Controls: Adjusts the canvas view
Component Library
-
Provides a collection of AI components, including data processors, models, API connectors, and logic handlers.
-
Drag-and-Drop Interface: Users can pick components from the library and place them on the canvas.
-
Node-based UI: Components can be connected using directed edges, representing the flow of data between modules.
Components
Function nodes are the fundamental building blocks similar to LEGO, you can build a customized LLM orchestration flow, a chatbot, an agentic app with all the integrations available in Ardor or build your own custom nodes. They represent various microservices with specific functionalities.
Interactive Editing
-
Move, connect and delete components
-
Customize properties and settings of each node
-
Customize source code of each node (if necessary)
-
Establish relationships by linking components via connection points
You could perform all the actions above by using the copilot.
Ardor Copilot (AI Assistant)
The right-side panel contains Ardor Copilot (Cerebrum), a built-in AI assistant to help users design and configure their agentic solutions.
To open copilot, click on the “Copilot” button in the toolbar.
-
Users can enter a prompt in the text box to describe the AI agent they want to create
-
Clicking “Ask Copilot” will generate recommendations or automatically configure an agent setup
Deployment Controls
Deploy Button (Top-right): Once the agent is ready, users can deploy it with a single click.
Ardor will deploy your solutions to the cloud in a serverless manner and provide you with a unique API endpoint or URl to interact with your solution. And Cerebrum will take care of maintanence and operations.
Vendor Flexibility
Ardor allows users to maintain full control over their AI deployments:
-
No Vendor Lock-In – Users can bring their own code, customize deployments, and migrate workloads without restrictions.
-
Scalable Architecture – Easily scale AI Agents up or down based on workload requirements.