Hospitality context
AI Guest Assistant
When available, Hestix can use stay and property context, operational information, business Knowledge and other data available at runtime.
AI & Knowledge
AvailableThe WhatsApp assistant uses approved information about your property. For enabled accounts, it can check availability on Octorate before replying.
With or without a booking
The available context changes how Hestix assists a conversation, not whether the conversation can happen in the first place.
You do not need a booking to talk to Hestix.
Hospitality context
When available, Hestix can use stay and property context, operational information, business Knowledge and other data available at runtime.
Generic context
A conversation can exist without a recognised booking or guest and still use business Knowledge. It is not a general-purpose chatbot.
AI Guest Assistant
For enabled tenants, the WhatsApp assistant uses property Knowledge and can invoke Hestix tools to check live stay availability on Octorate on demand, distinguishing alternative units from multi-unit requests.
Controlled enablement
Contextual answers about units, amenities and known operating details without inventing missing information.
AI interprets the request, while Hestix validates arguments and calls read-only PMS APIs when enabled.
Uncertain results, unverified quotes and out-of-policy requests remain explicit, with staff handoff where needed.
A live availability check is not a booking or a final price guarantee. Auto-reply, live checks and quotes depend on tenant policy and current evidence. A public website assistant widget remains on the roadmap.
Knowledge Manager
Knowledge Manager helps organise and manage operational information owned by the business, so it can be reused as context across conversations and workflows.
AI does not have to remember everything by itself. Hestix provides controlled context and Knowledge.
Information remains connected to the operation that manages it.
Approved Knowledge can support multiple conversations and workflows.
Knowledge remains governed information, not implicit model memory.
The model
Hestix orchestrates available inputs, applies relevant context and rules, assists the response and keeps the path that can update Knowledge separate.
AI does not directly change Knowledge: a useful signal can enter a supervised process before it becomes reusable.
Real product proof
The Candidate Inbox shows Knowledge candidates separately from approved Knowledge and provides a dedicated review process.
Conversation Learning
Conversation Learning is a supervised learning mechanism. Selection criteria can be applied to useful signals: not every question or human answer automatically creates a candidate.
Hestix can learn from conversations without treating every conversation as truth.
Conversation Learning can be enabled and configured for each business.
Human in the loop
Control does not happen after publishing: it is part of the path separating a promising signal from reusable Knowledge.
AI and conversations can surface useful information.
A candidate is not approved Knowledge yet.
Human review controls what becomes reusable.
Cross-platform AI
AI and Knowledge can support guest operations, Knowledge management, data interpretation and workflows. They belong to the Hestix operational layer, not a separate box.
Hestix calculates the numbers. AI helps you understand them.
Business rules and verifiable calculations remain separate from generative interpretation. AI does not replace the rules or invent missing data.
Explore the Hestix platformBusiness context
Hestix uses each business’s context to keep information and configuration relevant. Some AI and Conversation Learning capabilities may require configuration or activation for each business.
Frequently asked questions
No. Hestix can use hospitality context when it exists; Generic AI / Knowledge can also work without a recognised booking.
No. Conversations can produce signals or Knowledge candidates, but the process is supervised and a candidate is not approved Knowledge.
No. It can be enabled and configured for each business.
No. Business rules and verifiable calculations remain separate from generative interpretation.
AI & Knowledge in practice
The demo starts with the context, operational information and workflows you want to make more useful and controllable.