Course Overview: LangChain and LangGraph Practical Tutorial

Welcome to the LangChain and LangGraph Practical Tutorial, an additional reference course designed to enhance your understanding and proficiency with LangChain and LangGraph frameworks. This course is meticulously crafted to provide hands-on experience through a series of practical examples hosted on Kaggle notebooks, allowing you to read, interact with, and practice implementing various concepts in real-world scenarios.

Prerequisites

Before diving into this course, it is highly recommended that you:

• Read the LangChain Doc : Gain a solid foundation in LangChain's core principles, functionalities, and applications.
• Read the LangGraph Doc : Understand the fundamentals of LangGraph, its architecture, and how it complements LangChain.

These textbooks will equip you with the necessary theoretical knowledge, enabling you to maximize the benefits of the practical examples provided in this tutorial.

Course Structure

The tutorial is organized into several comprehensive sections, each focusing on different aspects of LangChain and LangGraph. Below is an overview of each module and the corresponding Kaggle notebooks that serve as practical references:

---

1. LCEL (LangChain Expression Language)

Delve into the extended functionalities of LangChain with quick reference guides that cover essential components and design patterns.

• LangChain: LCEL Quick Reference
• LangChain: Message Design Quick Reference
• LangChain: Output Parser Quick Reference
• LangChain: Prompt Templates Quick Reference

2. Splitter

Master text splitting techniques essential for preprocessing in Retrieval-Augmented Generation (RAG) workflows.

• Markdown Extraction for RAG Workflows (MarkItDown)
• LangChain: Mastering Text Splitting

3. Indexing

Learn comprehensive methods for document indexing, embedding caching, and utilizing vector stores to optimize data retrieval.

• LangChain: Document Indexing Comprehensive Guide
• LangChain: Caching Embeddings Designs
• LangChain: Chroma Quick Reference (Vector Store)

4. Retriever

Explore various retrieval mechanisms to efficiently query and manage your indexed data.

• LangChain: MultiQueryRetriever Quick Reference
• LangChain: MultiVectorRetriever Quick Reference
• LangChain: ParentDocumentRetriever Quick Guide
• LangChain: SelfQueryRetriever Quick Reference
• LangChain: EnsembleRetriever Quick Reference

5. Tool

Understand the schema definitions, structured tools, and building conversational AI tools within LangChain.

• LangChain: The Definitions of Tool Schema
• LangChain: StructuredTool Comprehensive Guide
• LangChain: Structured Outputs from LLM
• LangChain: Building Conversational AI Tools

6. Built-in Tools

Leverage built-in tools within LangChain for web searches, SQL database interactions, and data analysis.

• LangChain: DuckDuckGo Quick Guide (Web Search)
• LangChain: SQLDatabase Built-In Toolkit Guide
• LangChain: Data Analysis with REPL-Tool and LLM

7. Agent Workflow

Dive into the workflow of agents within LangChain and LangGraph, including code generation, custom libraries, and execution planning.

• LangGraph: Pandas Code Generator
• LangGraph: AI Coder with Custom Library
• LangChain: Tool-Integrated with ReAct Agent
• LangGraph: From Planning to Execution
• LangChain: Pandas DataFrame Agent Guide
• LangGraph: SQL Query Generation
• LangGraph: Hierarchical Agent Teams
• LangGraph: Dynamic Chart Generator (Agent Teams)

8. Agentic RAG (Retrieval-Augmented Generation)

Enhance your RAG workflows with agentic capabilities, utilizing Chroma for adaptive, corrective, and self-driven data retrieval.

• LangGraph: Adaptive RAG with Chroma
• LangGraph: Agentic RAG with Chroma
• LangGraph: Corrective RAG with Chroma
• LangGraph: Self RAG with Chroma

---

How to Access the Course Materials

All practical examples and references are available through Kaggle notebooks, providing an interactive environment to experiment and implement the concepts discussed. Access the tutorials and notebooks through the following link:

• LangChain Tutorials

Learning Outcomes

By the end of this course, you will:

• Gain practical experience with LangChain and LangGraph through diverse, real-world examples.
• Understand advanced concepts such as document indexing, retrieval mechanisms, and agent workflows.
• Develop the ability to build and integrate sophisticated AI tools and workflows.
• Enhance your skills in implementing Retrieval-Augmented Generation (RAG) with agentic capabilities.

Recommended Approach

1. Start with the Textbooks: Begin by thoroughly reading the LangChain and LangGraph textbooks to build a strong theoretical foundation.
2. Explore the Tutorials: Navigate through the structured modules, starting with LCEL and progressing to more advanced topics like Agentic RAG.
3. Hands-On Practice: Actively engage with the Kaggle notebooks by running the code, experimenting with modifications, and implementing your own projects.
4. Supplement Learning: Utilize the quick references and comprehensive guides to deepen your understanding and address specific challenges.

Embark on this tutorial to elevate your expertise in LangChain and LangGraph, leveraging practical examples to transform your theoretical knowledge into actionable skills.