AI for restaurants

AI for restaurants that works on a Friday night.

Restaurants do not need a chatbot demo. They need the booking to land on the right table, the guest's history on the host's screen and a menu that sells the dish. Here is what we build for restaurants and cafés, and what running it taught us.

See TablePilot

30-minute Google Meet with Antoine or Marco, weekday slots. The invite and link come by email.

What we build

Bookings & CRM
TablePilot, and a platform a restaurant group owns
Menus
Vimus video menus behind a QR code
WhatsApp
Booking, waitlist, ordering and shop questions
Voice
Ordering over a WhatsApp call (prototype)
Products

What we build for restaurants

In the field

Restaurant and café projects

  • Restaurants · South AfricaIn production, pilot phase

    A booking and guest platform a restaurant group owns

    A reservation system built for a restaurant group to replace a third-party booking tool: guests book online or on WhatsApp, hosts run the service from the real floor plan, and every guest's history sits in one CRM.

    • Booking widget for guests and a host dashboard with the restaurant's own floor plan
    • Guest CRM with the booking history imported from the previous system
    • Email confirmations, a WhatsApp AI booking assistant and a waitlist
    • A marketing console for guest journeys
    • React
    • API gateway + services + worker
    • PostgreSQL
    • Redis
    • Docker
    • WhatsApp Business Platform
    Read the case study
  • Restaurants · FranceLive

    A bilingual website and booking back office for a brasserie

    A French and English website for a brasserie, with its own booking system: the guest books online, the request lands in the restaurant's reservations inbox, and the team confirms or declines it from a back office — which e-mails the guest either way.

    • Two domains, one per language, from one codebase
    • Booking requests stored, e-mailed to the team and acknowledged to the guest
    • A back office where the restaurant confirms or declines each booking
    • Automated API tests and a one-command rollback
    • HTML
    • Node.js
    • Docker
    • Transactional e-mail
    Read the case study
  • Hospitality · East AfricaSales demo

    QR menus, table ordering and a WhatsApp assistant for cafés

    A hospitality platform shown to café and restaurant groups: every dish with a photo and a video behind a QR code, ordering from the table, loyalty, and a WhatsApp assistant that answers questions about each shop — opening hours, locations, contacts.

    • Branded menus where every dish has a photo and a video
    • Ordering, loyalty and frequently asked questions on WhatsApp, with mobile-money payment simulated in the demo
    • Node.js
    • MySQL
    • PostgreSQL
    • Redis
    • WhatsApp Business Platform
    Read the case study
  • Hospitality · prototypePrototype

    Voice ordering over a WhatsApp call

    A customer calls the business on WhatsApp and orders by voice; an AI voice agent takes the order, then the payment link and the tracking link arrive as WhatsApp messages.

    • Node.js
    • WhatsApp Business Platform
    • ElevenLabs
    • Stripe
    Read the case study
What we learned

What actually works with AI in a restaurant

Five lessons from building and running these products with real hosts, owners and guests.

  1. The floor plan is the product. Hosts think in tables, not lists. The plan must open framed on the tables, show who serves which table, and make assigning a server something you find in one tap during service.
  2. History has to come with you. A guest CRM is only useful on day one if the bookings and guests of the previous system are imported — including the careful work of merging duplicate guests without merging two different people.
  3. The guest has one hand free. Menus and booking forms are used standing or seated, on mobile data. Thumb-sized buttons, forms that scroll to the last field, and a still image while a video loads.
  4. WhatsApp needs lead time. Messages that start a conversation — reminders, confirmations outside the 24-hour window — use templates Meta must approve first. Submit them early.
  5. The owner is not technical. Plain restaurant words, no dead buttons, one obvious way to do each thing. A confusing screen costs a support call during service.
FAQ

Frequently asked questions

  • What can AI do for a restaurant today?

    Take bookings and manage a waitlist on WhatsApp, keep every guest's history in one CRM, turn the menu into short dish videos behind a QR code, translate it into the guest's language, and take orders with payment links.

  • Can AI take restaurant bookings on WhatsApp?

    Yes. We built a WhatsApp AI booking assistant and waitlist for a restaurant group; it writes into the same booking system the hosts use on the floor plan.

  • Do guests need an app for a QR code menu?

    No. With Vimus the QR code opens the menu in the phone's browser, with no app and no login.

  • Can we keep our booking history if we change system?

    Yes. For a restaurant group we imported the booking and guest history of their previous system so the new CRM was useful from the first day.

Running a restaurant or a group?

Tell us how you take bookings and orders today. We will show you what we would change first.