Overview

LangChain is one of the most popular LLM application development frameworks, providing a complete toolchain including model integration, prompt management, chain composition, and agent orchestration. This guide walks you through installing LangChain and building your first conversational chain from scratch.

Prerequisites

• Python 3.9+
• pip package manager
• API Key from OpenAI or another LLM provider

Core Content

Step 1: Install LangChain

# Install LangChain core packages
pip install langchain langchain-openai

# Or using conda
conda install -c conda-forge langchain

Verify installation:

import langchain
print(langchain.__version__)  # Output: 0.3.x

Step 2: Configure API Key

import os
from dotenv import load_dotenv

load_dotenv()
os.environ["OPENAI_API_KEY"] = "your-api-key-here"

Store the API key in a .env file to avoid hardcoding:

# .env
OPENAI_API_KEY=sk-...

Step 3: Create Your First ChatModel

from langchain_openai import ChatOpenAI

# Initialize ChatGPT model
llm = ChatOpenAI(
    model="gpt-4o-mini",
    temperature=0.7,
    max_tokens=1000
)

# Direct invocation
response = llm.invoke("Hello, briefly introduce LangChain")
print(response.content)

Step 4: Use PromptTemplate

from langchain_core.prompts import ChatPromptTemplate

# Define prompt template
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a professional {role}, answer concisely"),
    ("human", "{question}")
])

# Format and invoke
chain = prompt | llm
response = chain.invoke({
    "role": "Python engineer",
    "question": "What is a decorator?"
})
print(response.content)

Step 5: Build a Complete Chain

from langchain_core.output_parsers import StrOutputParser

# Add output parser to return plain string
chain = prompt | llm | StrOutputParser()

result = chain.invoke({
    "role": "code reviewer",
    "question": "How to write maintainable code?"
})
print(result)  # Returns string directly
print(type(result))  # <class 'str'>

Complete Code Example

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

os.environ["OPENAI_API_KEY"] = "your-api-key"

# 1. Initialize model
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0.7)

# 2. Define Prompt
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a professional tech assistant, be concise and accurate"),
    ("human", "{input}")
])

# 3. Create Chain (LCEL pipe syntax)
chain = prompt | llm | StrOutputParser()

# 4. Execute Chain
response = chain.invoke({"input": "Explain what RAG is"})
print(response)

# 5. Batch processing
responses = chain.batch([
    {"input": "What is LangChain?"},
    {"input": "What is LangGraph?"},
])
for r in responses:
    print(r[:50])

Verification

try:
    result = chain.invoke({"input": "hello"})
    assert isinstance(result, str)
    assert len(result) > 0
    print("✅ Chain working correctly")
except Exception as e:
    print(f"❌ Error: {e}")

Common Issues

Q: Dependency conflicts during installation?
Use a virtual environment: python -m venv venv && source venv/bin/activate

Q: AuthenticationError when calling?
Check that OPENAI_API_KEY is set correctly and has remaining credits.

Q: How to switch to Anthropic Claude?
Install langchain-anthropic and replace ChatOpenAI with ChatAnthropic.

References

• LangChain Official Docs
• LangChain GitHub Repository
• LangChain Quick Start Tutorial
• LangChain-OpenAI Integration
• ChatPromptTemplate API Reference