Skip to main content

Command Palette

Search for a command to run...

πŸ€– Build Your Own WhatsApp AI Assistant with Twilio + Make.com + OpenAI 🌟

Published
β€’2 min read
πŸ€– Build Your Own WhatsApp AI Assistant with Twilio + Make.com + OpenAI 🌟

Want to respond to WhatsApp messages automatically with AI-generated replies? 🀩 In this blog, I'll walk you through the exact steps I followed using Make.com, OpenAI or Together AI, and Twilio to create an intelligent WhatsApp chatbot – no coding required! πŸš€


πŸ“Œ Tools You’ll Need:


πŸ› οΈ Step 1: Setup Twilio WhatsApp Sandbox

  1. Go to Twilio Console

  2. Navigate to Messaging β†’ Try it out β†’ Send a WhatsApp Message

  3. Register your personal WhatsApp number.

  4. Save the From Number (whatsapp:+14155238886) and Sandbox credentials (Account SID & Auth Token).


🧩 Step 2: Create Scenario in Make.com

  1. Go to Make.com and click Create a new Scenario

  2. Add the first module βž• Webhooks β†’ Custom Webhook

  3. Click on the module β†’ Add a new webhook β†’ Copy the generated URL.


πŸ“² Step 3: Test Webhook Trigger

  1. Open Postman or any webhook testing tool

  2. Send a POST request to your webhook URL with this JSON:

{
  "from": "91970017XXXX",
  "message": "Hi AI"
}

βœ… Once it captures, click OK on Make.com.


πŸ€– Step 4: Add OpenAI or Together AI API Call

  1. Add HTTP β†’ Make a Request module

  2. Use the following setup:

    • Method: POST

    • URL: https://api.together.xyz/v1/chat/completions or OpenAI URL

    • Headers: Authorization: Bearer YOUR_API_KEY

    • Content-Type: application/json

    • Body (Raw JSON):

{
  "model": "mistralai/Mistral-7B-Instruct-v0.1",
  "messages": [
    { "role": "user", "content": "{{1.message}}" }
  ]
}

πŸ’¬ Step 5: Send Reply via Twilio

  1. Add another HTTP β†’ Make a Request module

  2. Set:

    • Method: POST

    • URL: https://api.twilio.com/2010-04-01/Accounts/YOUR_SID/Messages.json

    • Auth: Basic (Username = Account SID, Password = Auth Token)

    • Body Type: x-www-form-urlencoded

    • Fields:

      • To: whatsapp:+91970017xxxx

      • From: whatsapp:+14155238886

      • Body: {{2.choices[0].message.content}}


βœ… Final Result

Now, when someone sends a message to your WhatsApp sandbox number, they will receive a smart AI-generated reply! 🎯


πŸ’‘ Tips

  • Add logging with Google Sheets or Airtable

  • Use Make.com filters to control responses

  • Upgrade to production WhatsApp after testing


🌈 Wrap-up

You've just created an intelligent WhatsApp AI Assistant using no-code tools! πŸ§ πŸ’¬ If you liked this tutorial, follow me for more AI + automation ideas.