Understanding Pipelines

Pipelines are a core feature of CoursePipelines that allow you to create automated data flows between courses and chatbots. This guide will help you understand how pipelines work and how to use them effectively.

Data Flow Visualization

The diagram above shows how data flows in a typical pipeline:

1. Course A sends quiz scores to an AI Chatbot and completion rates to Course B
2. The AI Chatbot analyzes the data and determines:
   • A skill level that affects Course B content
   • Topic mastery that guides Course C progression
3. Each connection represents a variable being passed between components

What is a Pipeline?

A pipeline is a visual representation of data flow between different educational components in your system. It consists of:

• Course Nodes: Represent individual courses in your system
• Chatbot Nodes: Represent AI-powered chatbots that can interact with learners
• Connections: Show how data flows between components through variables

Components

Course Nodes

Course nodes represent individual courses in your system. Each course can have:

• Input variables: Accept data from other courses or chatbots
• Output variables: Send data to other courses or chatbots
• Multiple connections: Can connect to multiple components

Chatbot Nodes

Chatbot nodes represent AI assistants that can:

• Receive input from courses
• Process learner interactions
• Generate output variables based on conversations
• Connect to multiple courses

Creating Connections

To create a connection between components:

1. Click and drag from an output handle (right side) of one component
2. Drop onto an input handle (left side) of another component
3. The connection will be created, establishing the data flow

Connection Rules

• Course inputs can only accept one connection
• Outputs can connect to multiple destinations
• Connections can be deleted by selecting them and pressing Delete

Managing Pipelines

You can:

• Create new pipelines with custom names
• Auto-arrange nodes for better visualization
• Delete connections and nodes as needed
• Save changes automatically
• Track the flow of learning data across your learner data system

Best Practices

1. Organize Logically: Arrange nodes in a way that makes the data flow clear
2. Use Descriptive Names: Give your pipelines and variables clear, meaningful names
3. Validate Connections: Ensure your connections make sense for your learning objectives
4. Regular Review: Periodically review your pipelines to ensure they're working as intended

Advanced Pipeline Patterns

Here's an example of a more complex learning pathway with multiple chatbots and bidirectional data flow:

This advanced pipeline demonstrates:

• Parallel learning tracks (Programming and Theory)
• Chatbot collaboration (skill exchange between bots)
• Prerequisite relationships between courses
• Bidirectional feedback loops for continuous assessment

Common Pipeline Patterns

Basic Linear Flow

Feedback Loop

Question Integration

Here's how questions integrate into the pipeline flow:

This diagram shows how:

• Courses present questions to learners
• Question responses are stored as variables
• Variables feed into chatbot analysis
• Chatbots provide adaptive content
• All data flows into reporting