Build a voice assistant backend for restaurant calls

Set up an n8n backend so a voice agent can check tables, handle bookings, take orders, and log callback requests.

Done with n8n· Intermediate

What they did
They built an n8n workflow to sit behind an ElevenLabs conversational voice agent. The system used five webhook endpoints for live calls: checking table openings, creating or changing reservations, taking pickup or delivery orders, and asking for a human callback. Reservation records were stored in Google Sheets, and staff alerts were sent through Gmail. A similar pattern was also used for Vapi travel and home-service booking agents.
How it went
The project included importable JSON files and a description of the workflows, but no operational results were reported. There were no figures for call volume, accuracy, or usage, and the workflows were presented as examples to start from.
Tips
Test with fake data before connecting real bookings or customer details. Lock down the public webhook URLs and review privacy and service costs before putting it into production.

Try it yourself with n8n

Create an n8n workflow for [restaurant name] that serves as the backend for a voice agent from [ElevenLabs or Vapi]. It should accept webhook requests to check table availability in [Google Sheets, Google Calendar, or another booking source], make, update, or cancel reservations, record pickup or delivery orders, and log callback requests. After each step, show me the planned action and wait for my approval before any confirmation is sent, any booking is changed, any message goes out, or any other irreversible action happens.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on github.com ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.