Build an AI Restaurant Reservation Voice Agent with Telnyx

Restaurants lose bookings when callers hit voicemail, wait on hold, or call outside staffed hours. A voice ai reservation agent gives developers a practical way to answer inbound calls, collect booking details, and send a confirmation without building a full contact center.

The canonical code example is in the Telnyx code examples repo:

https://github.com/team-telnyx/telnyx-code-examples/tree/main/ai-restaurant-reservation-voice-agent-python

What This Example Builds

This Python example creates an inbound restaurant reservation agent using Telnyx Voice AI, AI Inference, and SMS/MMS.

The call flow is straightforward:

1. A caller dials a Telnyx phone number.
2. Telnyx sends call events to a Flask webhook.
3. The app answers the call and plays a greeting.
4. The app gathers caller speech.
5. Telnyx AI Inference decides the next assistant response.
6. The app repeats the loop until it has the reservation details.
7. The app saves the reservation and sends an SMS confirmation.

Why This Is A Useful Voice AI Example

This example is small enough to understand in one sitting, but it still shows the core pieces developers need for a real voice ai workflow:

• Voice API webhooks for call lifecycle events
• Call Control commands for answer, speak, and gather
• AI Inference for conversational logic
• SMS confirmation after the voice interaction
• A simple /reservations endpoint for demo visibility

That makes it a good first project for developers who want to understand how voice, ai, and messaging fit together in one application.

Products Used

The example metadata lists:

telnyx_products: [Voice AI, SMS/MMS, AI Inference]

Voice AI handles the phone-call experience. AI Inference interprets what the caller says and generates the host response. SMS/MMS sends the reservation confirmation after the booking is complete.

Architecture

inbound phone call
  -> telnyx voice api webhook
  -> flask app
  -> speak greeting
  -> gather speech
  -> telnyx ai inference
  -> speak response
  -> save reservation
  -> send sms confirmation

The application stores demo reservations in memory. For a production deployment, replace that with a database and add authentication, observability, signature verification, rate limits, and retention controls.

Run The Example

git clone https://github.com/team-telnyx/telnyx-code-examples.git
cd telnyx-code-examples/ai-restaurant-reservation-voice-agent-python
cp .env.example .env
pip install -r requirements.txt
python app.py

Fill in the environment variables:

TELNYX_API_KEY=KEY...
RESTAURANT_NUMBER=+1...
AI_MODEL=moonshotai/Kimi-K2.6
PORT=5000

Expose the local webhook:

ngrok http 5000

Then configure your Telnyx Voice API application webhook URL:

https://<ngrok-id>.ngrok-free.app/webhooks/voice

Assign your Telnyx phone number to that Voice API application and call the number.

Demo Script

Say:

hi, i would like to book a table for two tomorrow at 7 pm.

If the agent asks follow-up questions, provide:

my name is anusha.
my phone number is +1...
yes, that is correct.

After the booking completes, check:

curl http://localhost:5000/reservations

You should see the saved reservation and receive an SMS confirmation.

Production Considerations

Before using this pattern in production, add:

• Persistent reservation storage
• Webhook signature verification
• Input validation for dates, times, and party size
• Duplicate booking protection
• Host escalation or transfer
• Monitoring for webhook and SMS failures
• Data retention and privacy controls