5. RAG and Memory - A Comprehensive Dive

64. Hands-on ~ Basic RAG Pipeline

The passage explains how to build a simple RAG pipeline with LangChain.

Imports and setup
- Uses common LangChain components like OpenAIEmbeddings, PromptTemplate, RunnablePassthrough, RunnableParallel, and related utilities.
- An embeddings model is already initialized for retrieval.

Defines a create_kb function.
Splits a manually created Document using RecursiveCharacterTextSplitter with:
- chunk_size = 5500
- chunk_overlap = 50
The document includes metadata such as source="blank_chain.md".
Uses split_documents() because the input is a Document object.
Creates a vector store from the chunks using vectorstore_from_documents(…), the embeddings model, and a persistence directory like ./temp.
Returns the vector store.

Calls create_kb() to get the vector store.
Builds a retriever with:
- search_type="similarity"
- k=2
Initializes a chat model with temperature=0.2.
Creates a prompt that instructs the model to answer only from the given context and say “I don’t know” if unsure.

Defines a helper like format_docs(docs) to join retrieved chunks into a single string for the prompt.

Creates a chain with two inputs:
- context: retrieved docs passed through format_docs
- question: passed through unchanged with RunnablePassthrough()
Pipes the inputs through:
1. prompt template
2. LLM
3. StringOutputParser
Produces a final string answer.

Invokes the chain with sample questions like:
- “What is LangChain?”
- “Who created LangChain?”
- “What is LangGraph used for?”
Prints the responses.

It follows the standard RAG flow:

The key benefit is that LangChain’s runnable and pipe syntax makes the entire pipeline modular, readable, and easy to compose.