Chapter 1: LangChain for RAG

LangChain Overview

LangChain provides building blocks for RAG pipelines:

• Document loaders and splitters
• Embeddings and vector stores
• Retrievers and chains
• LCEL for composability

Basic RAG Chain

from langchain.chat_models import ChatOpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import Chroma
from langchain.chains import RetrievalQA

# Setup
embeddings = OpenAIEmbeddings()
vectorstore = Chroma.from_documents(documents, embeddings)
llm = ChatOpenAI(model="gpt-4")

# Create chain
qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever=vectorstore.as_retriever()
)

# Query
response = qa_chain.invoke("What is RAG?")
print(response["result"])

LCEL (LangChain Expression Language)

Modern, composable approach:

from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough

# Define prompt
prompt = ChatPromptTemplate.from_template("""
Answer the question based on the context:

Context: {context}
Question: {question}

Answer:""")

# Build chain with LCEL
chain = (
    {"context": retriever, "question": RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
)

# Use
response = chain.invoke("What is RAG?")

Chain Types

Type	Description	Use Case
stuff	Put all docs in prompt	Small contexts
map_reduce	Process docs separately, combine	Large datasets
refine	Iteratively refine answer	Complex queries
map_rerank	Score and select best	Multiple answers

Adding Memory

from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain

memory = ConversationBufferMemory(
    memory_key="chat_history",
    return_messages=True
)

chain = ConversationalRetrievalChain.from_llm(
    llm=llm,
    retriever=retriever,
    memory=memory
)

# Maintains conversation context
chain.invoke({"question": "What is RAG?"})
chain.invoke({"question": "What are its benefits?"})  # Remembers context

---

Next Steps

Continue to Chapter 2: LlamaIndex.