How to Intermediate · 3 min read

How to build ecommerce chatbot with AI

Quick answer

Build an ecommerce chatbot by integrating a large language model like gpt-4o with your product database using retrieval-augmented generation (RAG). Use OpenAI API for chat completions and vector search tools like FAISS to enable product recommendations and customer support.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0
pip install langchain>=0.2.0
pip install faiss-cpu

Setup

Install required Python packages and set your OPENAI_API_KEY environment variable.

Use pip install openai langchain faiss-cpu to install dependencies.
Export your API key in your shell: export OPENAI_API_KEY='your_api_key'.

bash

pip install openai langchain faiss-cpu

output

Collecting openai
Collecting langchain
Collecting faiss-cpu
Successfully installed openai langchain faiss-cpu

Step by step

This example shows a simple ecommerce chatbot that answers user queries by searching a product catalog using vector embeddings and then generating responses with gpt-4o.

python

import os
from openai import OpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.chains import RetrievalQA
from langchain.llms import ChatOpenAI

# Initialize OpenAI client
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Sample product catalog
products = [
    {"id": "p1", "name": "Wireless Mouse", "description": "Ergonomic wireless mouse with USB receiver."},
    {"id": "p2", "name": "Mechanical Keyboard", "description": "RGB backlit mechanical keyboard with blue switches."},
    {"id": "p3", "name": "Noise Cancelling Headphones", "description": "Over-ear headphones with active noise cancellation."}
]

# Prepare documents for vector store
texts = [p["name"] + ": " + p["description"] for p in products]

# Create embeddings
embeddings = OpenAIEmbeddings(model="text-embedding-3-small", client=client)

# Build FAISS vector store
vectorstore = FAISS.from_texts(texts, embeddings)

# Create a retriever
retriever = vectorstore.as_retriever(search_type="similarity", search_kwargs={"k": 2})

# Setup LLM for answer generation
llm = ChatOpenAI(model_name="gpt-4o", temperature=0)

# Build RetrievalQA chain
qa = RetrievalQA.from_chain_type(llm=llm, retriever=retriever)

# Example user query
query = "Do you have wireless accessories for laptops?"

# Get answer
answer = qa.run(query)
print("User query:", query)
print("Chatbot answer:", answer)

output

User query: Do you have wireless accessories for laptops?
Chatbot answer: Yes, we offer a Wireless Mouse which is an ergonomic wireless mouse with a USB receiver, perfect for laptop use.

Common variations

You can enhance your ecommerce chatbot by:

Using async calls with the OpenAI SDK for better performance.
Streaming responses for real-time user experience.
Switching to other models like claude-3-5-haiku-20241022 or gemini-2.0-flash for different response styles.
Integrating with databases or APIs for live inventory and order tracking.

python

import asyncio
from openai import OpenAI

async def async_chat():
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    stream = await client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": "What wireless accessories do you have?"}],
        stream=True
    )
    async for chunk in stream:
        print(chunk.choices[0].delta.content or "", end="", flush=True)

asyncio.run(async_chat())

output

Yes, we have wireless accessories including an ergonomic wireless mouse with USB receiver and Bluetooth headphones.

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY is set correctly in your environment.
If vector search returns irrelevant results, increase k in search_kwargs or improve your product descriptions.
For slow responses, consider using smaller models like gpt-4o-mini or enable streaming.

✅

Key Takeaways

Use vector embeddings and FAISS to enable product search in your chatbot.
Integrate OpenAI's gpt-4o for natural language response generation.
Leverage LangChain for chaining retrieval and generation easily.
Use streaming and async calls to improve user experience and performance.
Ensure environment variables and API keys are correctly configured to avoid errors.

Verified 2026-04 · gpt-4o, gpt-4o-mini, claude-3-5-haiku-20241022, gemini-2.0-flash, text-embedding-3-small

Verify ↗