LangChain Basics: Build AI Apps from Scratch

Python basics

Variables and data types:

# string
name = "Xiaoming"

# number
age = 25
height = 1.75

# list
skills = ["Python", "AI", "LangChain"]

# dict
person = {"name": "Xiaoming", "age": 25}

Function definition:

def greet(name):
    return f"Hello, {name}!"

print(greet("Xiaoming"))  # Output: Hello, Xiaoming!

Install third‑party packages with pip:

pip install package_name

LangChain overview

Core concepts

Easily connect various large models (GPT, Claude, Wenxin Yiyan, etc.)

Build intelligent chatbots

Allow AI to access private data

Create automated workflows

Why use LangChain

Answer questions based on internal company documents

Automatically analyze customer feedback

Generate personalized marketing copy

Quick start

Install LangChain

pip install langchain langchain-openai

First LangChain program

from langchain_openai import ChatOpenAI
from langchain_core.messages import HumanMessage

# Initialize model (requires OpenAI API key)
llm = ChatOpenAI(model="gpt-3.5-turbo", api_key="YOUR_API_KEY")

# Send a message
response = llm.invoke([HumanMessage(content="Introduce yourself in one sentence")])
print(response.content)

I am an AI assistant developed by OpenAI, designed to help users answer questions, complete tasks, and provide information.

Domestic model example

# Example with Zhipu AI
from langchain_community.chat_models import ChatZhipuAI

llm = ChatZhipuAI(model="glm-4", api_key="YOUR_ZHIPU_API_KEY")

Core components

Prompt templates

from langchain_core.prompts import ChatPromptTemplate

template = """
You are a {role} expert. Please answer the following question in a {tone} tone:
{question}
"""

prompt = ChatPromptTemplate.from_template(template)

formatted = prompt.format(
    role="Python",
    tone="humorous",
    question="What is a variable?"
)

Chains

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

chain = prompt | llm | StrOutputParser()

result = chain.invoke({
    "role": "history",
    "tone": "vivid and fun",
    "question": "Who was Qin Shi Huang?"
})
print(result)

Retrieval‑augmented generation

from langchain_community.document_loaders import TextLoader
from langchain_text_splitters import CharacterTextSplitter
from langchain_community.embeddings import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS

loader = TextLoader("company_manual.txt")
documents = loader.load()

text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
docs = text_splitter.split_documents(documents)

embeddings = OpenAIEmbeddings()
vectorstore = FAISS.from_documents(docs, embeddings)

retriever = vectorstore.as_retriever()

from langchain.chains import create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain

chain = create_stuff_documents_chain(llm, retriever)

Practical case: intelligent customer‑service bot

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

system_prompt = """
You are an intelligent customer‑service assistant for an e‑commerce company.
Please answer user questions in a friendly, professional manner.
If you don’t know the answer, be honest and suggest contacting human support.
"""

prompt = ChatPromptTemplate.from_messages([
    ("system", system_prompt),
    ("human", "{input}")
])

llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0.7)

chain = prompt | llm | StrOutputParser()

while True:
    user_input = input("👤 You: ")
    if user_input.lower() in ["exit", "quit", "bye"]:
        print("🤖 Bot: Thank you for contacting us, have a great shopping experience!")
        break
    response = chain.invoke({"input": user_input})
    print(f"🤖 Bot: {response}")

References

https://python.langchain.com.cn/

https://python.langchain.com/