Products Built for Q Station

QAI: ask the property a question.

QAI is an AI agent for the people running a hotel and events venue. Staff ask in plain English; QAI reads the property’s own systems and answers with where each number came from and how fresh it is. Q Station in Manly is the design partner, and its team uses QAI in a pilot today.

Client: Q Station
Status: In pilot
Access: Internal system

Build something like this

An internal system behind access control, so it has no public link. The screens on this page use sample data.

QAI on sample data: asked what is happening this week, it answers with a table of six events, what needs attention, and the source and time of its data. — Mockups on sample data, drawn from the real QAI. Every name and figure in them is made up.

The same QAI answer on a phone. — Mockups on sample data, drawn from the real QAI. Every name and figure in them is made up.

The problem

At a property like Q Station, the answer to “what’s on this week?” is spread across events, rooms, dining and activities, each in its own system. Pulling it together is someone’s weekly job, and the summary is out of date as soon as a booking changes.

What we built

QAI reads the property’s systems and never writes to them. Connectors copy each system’s data on a schedule, so answers come back in seconds. The agent can only use the tools it has been given, and who is signed in decides which tools those are.

Every answer carries a receipt of its source and freshness. Staff rate each one, and the department that owns the data corrects anything wrong.

What it does today

In pilot.

Next. Later phases would connect more of the property’s systems and let QAI propose an action, then wait for a person to approve it before anything happens.

Answers questions about the week’s events and bookings in plain English
Shows the source and freshness of every number it gives
Only reads: it cannot change a booking or send anything
Gives each department the data it needs, decided by who is signed in
Lets staff rate every answer, and the property’s admins review conversations

The stack

What QAI runs on.

Python The connectors that read each of the property’s systems
LangGraph The agent: which tools it may use, and in what order
Claude The model behind the answers
FastAPI The agent service
Next.js The chat and the console
TypeScript The chat, the console and their API

Need something like QAI built?

Tell us what you need. The team that built QAI will tell you plainly what it would take.

Start a conversation