Generative AI development with LangChain & LangGraph

Description:

In today's AI-driven landscape, companies increasingly rely on Large Language Models (LLMs) to build intelligent applications for automation, customer interaction, data analysis, and more. LangChain and LangGraph provide powerful tools to streamline the development of such applications. LangChain is an open-source framework that simplifies LLM application development by offering modular, user-friendly components. It allows seamless integration with models like OpenAI's GPT, Hugging Face models, Google Bard, and PaLM, among others. With LangChain, you can create solutions for text summarization, question answering, structured data analysis, and more.

LangGraph extends LangChain’ s capabilities by enabling complex, multi-step workflows with AI agents, supporting stateful applications, decision-making, and adaptive reasoning. Companies use LangGraph to build scalable, intelligent systems that require structured execution flows, such as chatbots, autonomous agents, and Retrieval-Augmented Generation (RAG) pipelines.

This hands-on course covers LangChain fundamentals, data connections, vector databases, memory, chains, and agent-based workflows. You’ll also explore LangGraph’ s role in building AI-powered applications with structured execution logic. By the end of this course, you'll be equipped to create cutting-edge AI applications that leverage LLMs effectively.

Duration: 3 Days

Course Code: BDT406

Learning Objectives:

After this course, you will be able to:

• Understand the fundamentals of LangChain and its components
• Work with model inputs and outputs using structured prompts
• Connect LLMs to external data sources using document loaders, text embeddings, and vector stores
• Build robust applications using LangChain chains to link multiple LLM calls
• Leverage memory to maintain conversational context across interactions
• Develop AI agents using LangChain components to automate complex tasks
• Explore LangGraph and its role in building structured AI workflows
• Implement Reflection and Reflexion agents for iterative learning
• Build advanced Retrieval-Augmented Generation (RAG) applications
• Use persistence techniques in LangGraph to maintain state and human-in-the-loop workflows
• Set up and utilize the LangGraph ecosystem, including LangGraph Cloud API and IDE
•  

• Must have some python programming experience.

• Software engineers, backend developers, full-stack developers, and business analysts interested in building Generative AI applications with LangChain and LangGraph—no prior machine learning experience required.

Course Outline:

1. LangChain Fundamentals

What is LangChain?

Installing LangChain and setting up environment

Creating and using prompts

Hands-on: Writing a basic prompt with LangChain

Model Inputs and Outputs

Understanding model inputs and outputs

Connecting LangChain with an LLM

Using prompt templates (zero-shot, one-shot, few-shot learning)

Hands-on: Experimenting with model inputs and outputs

Data Connections and Vector Stores

Connecting LLMs to external data

Using document loaders and transformers

Working with text embeddings and vector databases (FAISS, ChromaDB)

Hands-on: Implementing a vector database for retrieval

Using Chains for Application Logic

Understanding LLM chains

Building sequential and router-based chains

Hands-on: Developing a LangChain pipeline

Memory in LangChain

Introduction to memory and its use cases

Types of memory and memory buffers

Hands-on: Implementing chat memory for contextual conversations

Building AI Agents with LangChain

Introduction to AI Agents and their applications

Understanding ReAct framework in LangChain

Exploring agent tools and executions

Hands-on: Building an agent-based application

Introduction to LangGraph

What is LangGraph?

Why use LangGraph with AI workflows?

Flow engineering and LangGraph components

Building a Reflection Agent

What is a Reflection Agent?

Creating a reflection chain and defining LangGraph graphs

Understanding State Graphs

Hands-on: Implementing a Reflection Agent

Developing a Reflexion Agent

Understanding Reflexion agent for iterative improvement

Build Actor, Revisor agents and integrate message graph

Integrating search tools (Tavily, others)

Hands-on: Creating a Reflexion Agent with LangGraph

Advanced RAG (Retrieval-Augmented Generation)

Constructing an advanced RAG workflow

Using nodes and states for decision making

Leveraging predefined LangChain prompt templates

Implementing Self-RAG for reducing hallucinations

Working with Adaptive RAG for dynamic retrieval

Hands-on: Build an advanced RAG system

Persistence & State Management in LangGraph

Using state persistence in LangGraph

Implementing memory-saving techniques and human-in-loop interventions

Storing State using SQLite saver

Hands-on: Using persistence in LangGraph application

LangGraph Ecosystem & Cloud Deployment

Setting up LangGraph Studio (IDE)

LangGraph Cloud API (set up local environment)

Using LangGraph Cloud API: Threads, Runs, Assistants

Hands-on: Install ecosystem on local machine (Docker required)