Building a WhatsApp AI Assistant Using Node.js

Text automation is a vital counterpart to voice AI. In this tutorial, we will write a Node.js microservice that connects the WhatsApp Cloud API with OpenAI's GPT-4o to create an automated, responsive chat assistant.

Prerequisites

Before starting, ensure you have:

• A Facebook Developer account with the WhatsApp product configured.
• Node.js installed locally.
• An OpenAI API key.

Project Setup

Create a new directory and install the necessary dependencies:

mkdir whatsapp-ai-assistant
cd whatsapp-ai-assistant
npm init -y
npm install express dotenv openai body-parser

Creating the Webhook

WhatsApp relies on webhooks to deliver incoming messages to your backend. Let's create an Express application to handle WhatsApp's verification request and receive message payloads.

const express = require('express');
const bodyParser = require('body-parser');
require('dotenv').config();

const app = express();
app.use(bodyParser.json());

// Verification endpoint for Meta
app.get('/webhook', (req, res) => {
  const mode = req.query['hub.mode'];
  const token = req.query['hub.verify_token'];
  const challenge = req.query['hub.challenge'];

  if (mode && token === process.env.VERIFY_TOKEN) {
    res.status(200).send(challenge);
  } else {
    res.sendStatus(403);
  }
});

// Receiver endpoint for messages
app.post('/webhook', async (req, res) => {
  const body = req.body;
  if (body.object === 'whatsapp_business_account') {
    if (body.entry && body.entry[0].changes && body.entry[0].changes[0].value.messages) {
      const msg = body.entry[0].changes[0].value.messages[0];
      const from = msg.from; // User's phone number
      const text = msg.text.body; // User's message

      console.log(`Received message from ${from}: ${text}`);
      
      // Process with OpenAI and respond
      await handleIncomingMessage(from, text);
    }
    res.sendStatus(200);
  } else {
    res.sendStatus(404);
  }
});

app.listen(process.env.PORT || 3000, () => console.log('Webhook is listening...'));

Connecting OpenAI

Now, let's process the user's message using OpenAI and trigger a reply back using the WhatsApp Cloud API:

const { OpenAI } = require('openai');
const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

async function handleIncomingMessage(phoneNumber, userText) {
  // 1. Generate reply from GPT-4o
  const completion = await openai.chat.completions.create({
    model: "gpt-4o",
    messages: [
      { role: "system", content: "You are a helpful customer assistant for Smart Voice AI." },
      { role: "user", content: userText }
    ],
  });

  const replyText = completion.choices[0].message.content;

  // 2. Call WhatsApp Send API
  const url = `https://graph.facebook.com/v18.0/${process.env.PHONE_NUMBER_ID}/messages`;
  
  await fetch(url, {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Authorization': `Bearer ${process.env.WHATSAPP_TOKEN}`
    },
    body: JSON.stringify({
      messaging_product: "whatsapp",
      to: phoneNumber,
      type: "text",
      text: { body: replyText }
    })
  });
}

Testing Your Webhook

To test locally, run node index.js and expose your port using a tunnel utility like ngrok:

ngrok http 3000

Copy the generated HTTPS forwarding URL, set it in your Meta developer dashboard, and send a WhatsApp message to your test number. Your AI assistant will reply instantly!