Building Agentic AI with Low-Cost LLMs: A Comprehensive Tutorial

Overview

This tutorial guides you through building agentic AI applications using CrewAI and Ollama, focusing on cost-effective implementation with local LLMs. The tutorial is split into two 4-hour sessions, providing both theoretical knowledge and hands-on experience.

Prerequisites

• Python 3.9+
• Git
• Basic understanding of Python programming
• Basic understanding of AI/ML concepts

Quick Start

1. Clone this repository:

git clone [repository-url]
cd crewai-tutorial

2. Create and activate a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Install Ollama:

• Visit ollama.com
• Download and install for your operating system
• Pull a base model:

ollama pull gemma:2b

Tutorial Structure

Part 1: Foundations & Basic Implementation

• Session 1 Guide
  • Environment Setup
  • CrewAI Fundamentals
  • Ollama Deep Dive
  • Building Your First Agent
  • Multi-Agent Basics

Part 2: Advanced Implementation & Real-World Applications

• Session 2 Guide
  • Advanced Agent Development
  • Building Complex Crews
  • Real-World Application Development

Code Examples

All code examples are organized in the examples directory:

• Basic Examples
• Advanced Examples
• Complete Applications

Resources

• CrewAI Documentation
• Ollama Documentation
• Model Comparison
• Cost Analysis

Contributing

Feel free to submit issues and enhancement requests!

License

This project is licensed under the MIT License - see the LICENSE file for details.