labs@atheron:~/industries/restaurants/ai$ train agent --grounded

Ordering assistant

Orders taken at the rush, allergies left to staff.

At the dinner rush the phone rings while every hand is busy. An assistant working from your own menu takes simple orders and answers the usual questions, and leaves allergies and anything unusual to the people who know the kitchen.

Start a project See the example projects

// what matters here

What is different in this industry.

Why a frontier model here

Menus are public and order volume is moderate, so a frontier model through its API is the quickest start. Card details never enter the conversation.

Your menu, not the internet

Answers come from your menu, hours and delivery areas, so the assistant never invents a dish.

Allergies go to a person

Every allergy or dietary question is passed to staff. The assistant never says a dish is safe.

Orders through your till

Orders are placed through your point of sale, so the kitchen sees them like any other.

// example projects

Priced examples for this service.

Each one is hypothetical, labelled as such, and priced live from our rate card. Open any of them in the estimator and make it yours.

Example projectWeb appAIFrontier model

Menu and ordering assistant

Hypothetical. Not a client, not a result.

// The problem

A busy pizzeria misses phone and chat orders at the dinner rush, and staff answer the same questions about hours, delivery areas and specials all evening.

// What we would build

An assistant on the website and chat that answers from the restaurant's own menu and information, takes orders through the point of sale, and passes every allergy question and anything unusual to staff. It uses a frontier model through its API: menus are public and the order volume is moderate.

// What is in it

Answers from your menu, hours and delivery areas
Orders placed through your point of sale
Allergy questions always passed to staff
Replies in the guest's language
No card details in the conversation

// Stack

Frontier model API
Next.js
PostgreSQL with pgvector
Point of sale interface
DigitalOcean

// estimate

Build: ≈ US$96,300 to US$147,000, delivered within 21 weeksCAD 137,100 to 209,600
Hosting: ≈ US$558 a monthCAD 795 a month
Support: ≈ US$2,500 a monthCAD 3,565 a month

Prices in your currency are estimates from today's Bank of Canada rate. All invoicing is in CAD or USD.

AI route: Frontier model
Model running cost: ≈ US$383 a monthCAD 545 a month

Timeline by milestone

Discovery: 2.9 to 3.4 weeks
Specification and evaluation plan: 0.4 weeks
Design approved: 2.4 to 3.4 weeks
Core features: 1.4 to 1.9 weeks
Full build: 0.9 to 1.4 weeks
Working pilot: 3.9 to 6.4 weeks
Testing and fixes: 0.9 to 1.9 weeks
Launch: 0.9 to 1.4 weeks
Production: 0.9 to 1.4 weeks

Outside our hands, and added to the calendar

Interface access from your point of sale vendor: 1 to 4 weeks

How it is paid

Deposit 20%: CAD 27,420 to 41,920
Discovery 2%: CAD 2,742 to 4,192
Specification and evaluation plan 10.2%: CAD 13,984.20 to 21,379.20
Design approved 2%: CAD 2,742 to 4,192
Core features 6.2%: CAD 8,500.20 to 12,995.20
Full build 4.1%: CAD 5,621.10 to 8,593.60
Working pilot 30.7%: CAD 42,089.70 to 64,347.20
Testing and fixes 2%: CAD 2,742 to 4,192
Launch 2%: CAD 2,742 to 4,192
Production 10.8%: CAD 14,806.80 to 22,636.80
Holdback, 30 days after launch (10%): CAD 13,710 to 20,960

Start from this example

// questions

Questions we get about this.

Can the restaurant assistant answer the phone?

It can work on web chat and text first; phone ordering is possible with a voice service, which we price separately.

What does the ordering assistant say about allergies?

It tells the guest that staff will answer and hands the question over. It never confirms a dish is free of an allergen.

Could the ordering assistant run on an open-source model?

Yes, in your own cloud account. It costs more to run at this volume; we price both.

Back to the industry: Restaurants and food service About the service: AI agents and models

// next

Not quite your project?

Tell us what you have in mind. We will come back to you with a range and the questions that would narrow it. Or book a call and talk it through.

Start a project Book a call