• Building Agentic AI applications from scratch is difficult.
• Various frameworks exist to simplify this process, such as CrewAI, Microsoft Autogen, and LlamaIndex.
• We will focus on LangGraph, developed by the LangChain team, which is considered one of the top frameworks for building Agentic AI applications.

Goals of the Lecture

• Provide a deep understanding of why LangGraph exists and the problems LangChain cannot solve(so that we required LangGraph).
• Offer a technical overview of LangGraph.
• Compare and contrast LangChain vs. LangGraph.
• Enable the learner to determine whether to use LangChain or LangGraph for a given application.

3. Prerequisites

• A basic understanding of LangChain is required, including its functionalities and basic coding.
• Specifically, it's recommended to watch introductory videos on "Introduction to LangChain" and "LangChain Components”

LangChain Recap

• Definition: LangChain is an open-source library designed to simplify the process of building LLM-based applications.
• It facilitates integrating Large Language Models (LLMs) into various software applications, such as chatbots or browser plugins.
• How it works: LangChain provides modular building blocks taht let you create sophisticated LLM-based workflows with ease.

◦ Key Components/Building Blocks: