TMLS 2025: AI Engineering Workshop Day

Welcome to AI Makerspace's comprehensive AI Engineering workshop! Today we'll explore three critical areas of modern AI development: RAG in Practice, Agents with LangGraph, and Model Context Protocol (MCP).

🚀 Day Overview

This workshop is designed to take you from foundational concepts to production-ready implementations across three key AI engineering domains:

📚 Session 1: RAG In Practice

Production-Ready RAG Applications for 2025

While 2025 might be the year of agents for AI Engineers, it's the year of practical RAG for enterprise and AI Engineering leaders. We'll cover the minimum viable production-ready LLM app stack:

• Two Implementation Tracks:
  • RAG-In-Practice-2025 - Our recommended enterprise stack
  • RAG-In-Practice-2025-OSS - Open-source focused implementation

Key Technologies:

• 🎺 Orchestration: LangChain's LangGraph
• ↗️ Vector Database: QDrant
• 📊 Reranking: Cohere's Rerank
• 📐 Evaluation: RAGAS

🤖 Session 2: Agents

Building Intelligent Agents with LangGraph

Dive deep into agentic AI systems with hands-on LangGraph development:

01_Our_First_Agent_with_LangGraph

Create your first agentic RAG application covering:

• Tool belt creation
• State management
• Graph compilation and execution
• LangSmith evaluation

02_Multi_Agent_with_LangGraph

Advanced multi-agent systems and coordination patterns

🔌 Session 3: MCP (Model Context Protocol)

Extending AI Capabilities with Custom Tools

Learn to build custom MCP servers for enhanced AI interactions:

MCP-Session-Code

• Web search integration with Tavily API
• Custom tool development
• MCP server configuration in Cursor

⚡ Environment Setup Instructions

🔧 IMPORTANT: UV Environment Usage

Each session directory contains its own isolated uv environment. You MUST navigate to each specific subdirectory and use its environment for that session:

# For RAG sessions
cd 01_RAG_In_Practice/RAG-In-Practice-2025/
# or
cd 01_RAG_In_Practice/RAG-In-Practice-2025-OSS/

# For Agent sessions  
cd 02_Agents/01_Our_First_Agent_with_LangGraph/
# or
cd 02_Agents/02_Multi_Agent_with_LangGraph/

# For MCP session
cd 03_MCP/MCP-Session-Code/

🐍 Environment Activation

In each directory, you can:

• Sync the environment: uv sync

📋 Prerequisites

• Python: 3.11+ (3.13+ recommended for MCP)
• UV Package Manager: Installation Guide
• API Keys: You'll need various API keys (OpenAI, Cohere, Tavily, etc.) - check individual session READMEs

Windows Users

For the MCP session specifically, ensure you have:

• winget install astral-sh.uv
• winget install --id Git.Git -e --source winget

🗓️ Suggested Schedule

1. Morning: RAG In Practice (Choose your track)
2. Midday: First Agent with LangGraph
3. Afternoon: Multi-Agent Systems or MCP Integration

🔗 Quick Navigation

Session	Directory	Focus
RAG (Enterprise)	01_RAG_In_Practice/RAG-In-Practice-2025/	Production RAG stack
RAG (OSS)	01_RAG_In_Practice/RAG-In-Practice-2025-OSS/	Open-source RAG
First Agent	02_Agents/01_Our_First_Agent_with_LangGraph/	Basic agentic RAG
Multi-Agent	02_Agents/02_Multi_Agent_with_LangGraph/	Advanced agents
MCP Tools	03_MCP/MCP-Session-Code/	Custom AI tools

💡 Getting Help

• Each subdirectory contains detailed setup and usage instructions
• Check individual README files for session-specific requirements
• Ensure you're using the correct uv environment for each session

🎯 Learning Outcomes

By the end of today, you'll have:

• ✅ Built production-ready RAG applications
• ✅ Created intelligent agents with LangGraph
• ✅ Developed custom MCP tools
• ✅ Understood evaluation and monitoring strategies
• ✅ Gained hands-on experience with the latest AI engineering tools

---

Ready to build the future of AI? Let's get started! 🚀