🤖 AI Agents & Workflows - Code Repository

---

📚 About This Repository

This repository contains all the code examples and implementations from the book on Building AI Agents and Workflows. Learn how to create, optimize, and deploy intelligent AI agents using cutting-edge frameworks and methodologies.

What You'll Learn

• 🎯 Agentic vs Non-Agentic Workflows - Understand the differences and when to use each approach
• 🔧 CrewAI Framework - Build multi-agent systems that collaborate to solve complex tasks
• 🎓 DSPy Optimization - Automatically optimize your AI agents for better performance
• 🔌 Model Context Protocol - Create MCP servers for seamless AI tool integration

---

🚀 Getting Started

Prerequisites

• Python 3.8 or higher
• OpenAI API key
• Additional API keys (see Environment Setup)

Installation

1. Clone the repository

   git clone https://github.com/yourusername/Code-for-book.git
   cd Code-for-book
   

2. Create 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. Set up environment variables

   cp .env.example .env
   # Edit .env with your API keys
   

📖 Units Overview

Unit 2: Agentic vs Non-Agentic Workflows

Learn the fundamental differences between agentic and non-agentic approaches through a practical AI news research and Twitter posting system.

Key Topics:

• Multi-agent collaboration with CrewAI
• Tool integration (Exa search, image generation, Twitter API)
• Task delegation and sequential processing
• Comparison of workflow approaches

Notebooks:

• agentic_workflow.ipynb - CrewAI multi-agent system
• non_agentic.ipynb - Traditional sequential approach

Unit 3: DSPy Optimization

Discover how to optimize AI agents automatically using DSPy's advanced prompting and few-shot learning techniques.

Key Topics:

• DSPy signatures and modules
• ReAct agent pattern
• MIPROv2 optimization
• Custom tool integration
• Evaluation metrics

Files:

• dspy.ipynb - Complete DSPy implementation for nutrition analysis
• foodDB.json - Sample nutrition database
• prompt.txt - Example prompt structure

Unit 4: Model Context Protocol (MCP)

Build MCP servers to enable AI models to interact with external tools and services.

Key Topics:

• MCP server architecture
• FastMCP implementation
• Tool registration and execution
• Transport methods (stdio, SSE)

Files:

• mcp_server.py - FastMCP server implementation
• tool.py - Nutrition API tool example

---

✨ Features

🎯 Agentic Workflows (Unit 2)

• Multi-Agent Collaboration: Researcher, Image Generator, and Tweet Creator agents working together
• Semantic Search: Using Exa API for intelligent web research
• AI Image Generation: Creating visuals with Hugging Face models
• Social Media Automation: Automated Twitter posting with media

🧠 DSPy Optimization (Unit 3)

• Automatic Prompt Engineering: Let DSPy find the best prompts
• Few-Shot Learning: Optimize with minimal training examples
• ReAct Pattern: Reasoning and acting in iterative loops
• Custom Metrics: Define and optimize for your specific use case

🔌 MCP Integration (Unit 4)

• Flexible Transport: Support for stdio and Server-Sent Events
• Easy Tool Creation: Simple decorators for tool registration
• Type Safety: Full type hints for better development experience
• Extensible Architecture: Add new tools and capabilities easily

---

🛠️ Tech Stack

Category	Technologies
AI Frameworks	CrewAI, DSPy, LangChain
LLM Providers	OpenAI (GPT-4, GPT-4o-mini)
Search & Research	Exa API
Image Generation	Hugging Face Inference API
Social Media	Tweepy (Twitter API)
MCP	FastMCP
Utilities	Python-Decouple, Python-Dotenv, Pillow

🤝 Contributing

Contributions are welcome! Here's how you can help:

1. Report Issues: Found a bug? Open an issue
2. Suggest Improvements: Have ideas? Share them in discussions
3. Submit Pull Requests:
   • Fork the repository
   • Create a feature branch (git checkout -b feature/amazing-feature)
   • Commit your changes (git commit -m 'Add amazing feature')
   • Push to the branch (git push origin feature/amazing-feature)
   • Open a Pull Request

---

📄 License

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

---

🙏 Acknowledgments

• OpenAI for providing the GPT models
• CrewAI for the amazing multi-agent framework
• DSPy team for revolutionizing prompt engineering
• FastMCP for making MCP servers simple
• All the open-source contributors who make this possible

---

📬 Contact & Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Email: your.email@example.com

---