Mini-Projects Overview¶

Hands-On AI mini-projects are designed to provide hands-on learning experiences with large language models across different application types. These projects range from beginner to advanced levels and cover the various modules in Hands-On AI.

Project Structure¶

Each mini-project follows a consistent structure:

Title and Difficulty Level: Clearly indicates the project focus and complexity
Learning Objectives: What students will accomplish and learn
Required Background: Prerequisites and knowledge needed
Implementation Guide: Step-by-step instructions
Extension Activities: Ways to expand on the basic project
Assessment Criteria: How to evaluate project success

A note on type hints¶

The example code in these projects is written without type hints on purpose. For someone still learning the language, annotations like def ask(self, prompt: str) -> str: add reading overhead before the idea itself has landed, so the examples stay deliberately plain to keep the focus on the concept.

This is a teaching choice, not a recommendation against type hints:

For teachers: the plain style is intentional. The hands_on_ai library itself is fully type-hinted and type-checked (mypy), so the package models good practice even though the project snippets keep things simple.
For advanced students: adding type hints is a great extension exercise and reflects industry best practice. Annotate your project, run a type checker (mypy or pyright), and fix what it finds. Because the library ships type information, your editor and the checker will understand its functions too.

Project Categories¶

Chat Module Projects¶

Chat projects focus on creating conversational applications:

Personality Bots: Custom chat personalities for specific purposes
Interactive Applications: Goal-oriented conversation systems
Educational Assistants: Subject-specific tutoring systems
Creative Writing: Story generation and creative applications

RAG Module Projects¶

Retrieval-Augmented Generation projects combine document processing with LLMs:

Custom Knowledge Bases: Creating systems that answer questions from specific sources
Document Analysis: Processing and extracting information from documents
Hybrid Systems: Combining retrieval with additional processing

Agent Module Projects¶

Agent projects implement tool-using, reasoning systems:

Custom Tool Creation: Building specialized tools for agents
Multi-step Problem Solving: Complex reasoning tasks
Domain-specific Assistants: Specialized agents for particular fields

How to Use These Projects¶

Browse the Project Gallery to find a suitable project
Review the project documentation to understand requirements
Follow the implementation guide, making adaptations as needed
Use the provided starter code and templates when available
Evaluate based on the assessment criteria

Project Implementation Workflow¶

Setup: Configure the required environment and dependencies
Planning: Sketch the solution approach and identify key components
Implementation: Code the solution, following the project guide
Testing: Verify functionality with appropriate test cases
Extension: Add additional features or improvements
Documentation: Add comments and documentation for your code

Contributing New Projects¶

If you'd like to contribute a new mini-project:

Use the Project Template as a starting point
Ensure your project includes clear learning objectives
Provide a step-by-step implementation guide
Include sample code and expected outputs
Add assessment criteria and extension activities

Using Projects in the Classroom¶

For educators, these projects can be:

Individual assignments: Students complete projects independently
Group projects: Teams collaborate on more complex implementations
In-class demonstrations: Showcase capabilities and concepts
Course capstones: Combined with other modules for comprehensive projects

See the Education Guide for more information on integrating these projects into your curriculum.