Building Your First AI Agent with LangChain

Introduction

In this tutorial, we'll build a simple AI agent using LangChain that can search the web and answer questions based on the results.

Prerequisites

• Python 3.9+
• OpenAI API key
• Basic Python knowledge

Step 1: Setup

First, install the required packages:

pip install langchain openai serpapi

Step 2: Create the Agent

from langchain.agents import initialize_agent, Tool
from langchain.llms import OpenAI
from langchain.utilities import SerpAPIWrapper

# Initialize components
llm = OpenAI(temperature=0)
search = SerpAPIWrapper()

# Define tools
tools = [
    Tool(
        name="Search",
        func=search.run,
        description="Search the web for information"
    )
]

# Create agent
agent = initialize_agent(
    tools,
    llm,
    agent="zero-shot-react-description",
    verbose=True
)

Step 3: Run Your Agent

result = agent.run("What is the latest news about AI?")
print(result)

How It Works

1. The agent receives a query
2. It decides which tool to use (Search)
3. It executes the search and processes results
4. It formulates a response based on findings

Adding More Tools

You can extend your agent with additional tools:

• Calculator - For mathematical operations
• Wikipedia - For encyclopedic knowledge
• Code Interpreter - For running Python code

Best Practices

1. Be specific with tool descriptions - The agent uses these to decide which tool to use
2. Set appropriate temperature - Lower for factual tasks, higher for creative
3. Add error handling - Tools can fail, plan for it
4. Monitor costs - Each agent step costs API credits

Conclusion

You've built your first AI agent! From here, you can add more tools, improve prompts, and create more sophisticated workflows.