Skip to content
AtheronLABS

You're visiting from the United States. Prices are shown in US dollars. Not right?

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

AI hotel messaging

Every guest answered in their language, day or night.

Most guest messages are the same questions in different languages. An agent working from your property's own information answers them at any hour, takes simple requests through your systems, and passes everything else to the front desk with the conversation attached.

// what matters here

What is different in this industry.

01

Why a frontier model here

Hotel information is public and guests write in many languages, so a frontier model through its API is the fastest and simplest start. We keep personal details out of the prompts.

02

Your property, not the internet

Answers come from your menus, policies and property guide, so the agent never invents a shuttle you do not run.

03

Requests through your systems

Late checkout, spa times and restaurant tables are booked through your own systems, with rules you set.

04

The front desk stays in charge

Anything unusual goes to a person with the conversation attached, and staff can step in at any point.

// 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

Guest messaging agent

Hypothetical. Not a client, not a result.

// The problem

A resort's front desk answers the same questions about parking, breakfast, spa times and late checkout by phone, email and text, in many languages, around the clock.

// What we would build

A messaging agent on the website, the guest app and text messages that answers from the property's own information, books spa times and late checkouts through the hotel's systems, and hands to the front desk with the conversation attached. It uses a frontier model through its API: hotel information is public, guests write in many languages, and it is the fastest way to start.

// What is in it

  • Answers from your property information, menus and policies
  • Replies in the guest's language, day and night
  • Spa, restaurant and late checkout requests through your systems
  • Handoff to the front desk with the full conversation
  • No card details in the conversation, ever

// Stack

  • Frontier model API
  • Next.js
  • PostgreSQL with pgvector
  • Property management system interface
  • SMS provider

// estimate

Build
≈ US$102,000 to US$155,000, delivered within 21 weeksCAD 144,800 to 221,400
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

Certification or API access from your PMS vendor
1 to 4 weeks

How it is paid

Deposit 20%
CAD 28,960 to 44,280
Discovery 2%
CAD 2,896 to 4,428
Specification and evaluation plan 10.2%
CAD 14,769.60 to 22,582.80
Design approved 2%
CAD 2,896 to 4,428
Core features 6.2%
CAD 8,977.60 to 13,726.80
Full build 4.1%
CAD 5,936.80 to 9,077.40
Working pilot 30.7%
CAD 44,453.60 to 67,969.80
Testing and fixes 2%
CAD 2,896 to 4,428
Launch 2%
CAD 2,896 to 4,428
Production 10.8%
CAD 15,638.40 to 23,911.20
Holdback, 30 days after launch (10%)
CAD 14,480 to 22,140

// questions

Questions we get about this.

Can the guest messaging agent reply by text message?

Yes. It can answer on your website, in the guest app and by text, from one set of property information.

What if the guest agent does not know an answer?

It says so and hands the conversation to the front desk, rather than guessing.

Could the guest agent run on an open-source model?

Yes, in your own cloud account, if you prefer guest messages never reach an outside provider. It costs more to run at low volume; we price both.

// 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.

AI guest messaging agent for hotels and resorts | Atheron Network Labs