LangChain

JS + Python

Database & backend

Build LLM-powered apps — chains, agents, RAG pipelines.

What it does

LangChain is a framework for building apps that use large language models (LLMs) like GPT or Claude. It handles the plumbing for chatbots, document Q&A, agents that use tools, and retrieval-augmented generation (RAG — when you want the AI to answer based on your own documents).

When to use it

Use this when building a chatbot, document Q&A tool, or AI agent that needs to use multiple tools or data sources.

Real example

You want to build a customer support chatbot that answers questions based on your PDF documentation. Prompt: 'Use LangChain to load your docs PDFs with PyPDFLoader, split them into chunks, embed them with OpenAIEmbeddings, store in a Chroma vector store, then create a RetrievalQA chain with ChatOpenAI to answer user questions based on the docs.'

Good to know

Can be overwhelming to start — it does a lot. For simple LLM calls, just use the OpenAI or Anthropic SDK directly. LangChain shines for complex chains and agents.