Course Content

​

MODULE 1: Introduction to LangChain

• What is LangChain and why it matters

• Real-world applications: chatbots, RAG systems, AI agents

• Key concepts: LLMs, prompts, chains, agents, tools, memory, vector stores

• Environment setup: Python, LangChain installation, API keys (OpenAI, Google Gemini, etc.)

Module 2: LangChain Fundamentals

• Prompt templates: system, user, AI messages

• Chains: LLMChain, sequential chains

• Output parsers: structured outputs, JSON, Pydantic models

• Basic input/output workflows

Module 3: Working with Language Models

• Overview of LLMs: OpenAI, Anthropic Claude, Gemini, local models (LLaMA, GPT4All)

• Embeddings: concepts, creation, comparison

• Using embeddings for similarity search

Module 4: Document Loading & Preprocessing

• Document loaders: PDF, CSV, text, JSON, web scraping

• Document splitting: chunking, RecursiveCharacterTextSplitter

• Data cleaning and normalization

Module 5: Vector Databases

• Importance of vector stores

• Popular options: FAISS, Chroma, Pinecone, Weaviate, Milvus

• Creating vector indexes and performing similarity searches

Module 6: Retrieval-Augmented Generation (RAG)

• Understanding RAG

• Retrieval chains: stuff, map-reduce, refine

• Building RAG chatbots: ingesting documents, embedding creation, vector store integration

• Optimizing RAG: chunking strategies, prompt engineering, source citations

Module 7: Memory in LangChain

• ConversationBufferMemory

• ConversationSummaryMemory

• Entity memory

• Using memory with agents and chatbots