All projects

Agentic AI · Map-first travel & journey planning

Agentic AI for
map-first travel.

Project Wikistay In development

A journey you design on the map itself — stays, food, rest stops, tours and health services as layers you plan across, instead of ten tabs and a spreadsheet.

WhyBusiness modelBusiness analysisArchitectureThe agents

Why we built it

The reason this project exists.

Travel planning happens on a map in your head, then gets flattened into lists, bookmarks and screenshots that lose all their geography. Wikistay makes the map the interface: you see the services along a route, drag stops into the order you actually want, and the plan stays spatial — because that is how a trip is really shaped.

Every BIXSO project starts the same way — with a real business problem worth solving, not a technology looking for a use. Then comes the model, the analysis, and the agents.

01 · Business model

It starts with the business.

The map is free to use because the value compounds in the place data and the connections between businesses along a route.

  • Travellers plan free — the map, the layers and the journey planner cost nothing to use.
  • Businesses along a journey are the paying side, because a route is a queue of nearby demand.
  • Place data is owned, not rented — the moat is the dataset, not the map tiles.

02 · Business analysis — the DNA

Then the rigour.

We separated what a map engine should do from what a planner should do — most travel tools blur the two and end up with a map that cannot hold a plan.

The businessMap-first travel & journey planning
01 Processes
  • Browse service layers on the map
  • Build a multi-day journey
  • Reorder or add stops directly on the map
  • Connect with businesses along the route
02 Data
  • Places across five service layers
  • Journey days, stops and ordering
  • Routes between stops
  • Business profiles along a corridor
03 Rules
  • Demo data is always labelled as demo
  • Location is never persisted without consent
  • Owner-only edits on a journey
  • Map tiles carry no per-view licence cost
04 Decisions
  • Which stops make the journey
  • What order the days run in
  • Which businesses to connect with
  • When a suggestion is worth taking
A modelprecise enough for agents to run

fig. — business analysis: decoding Wikistay into processes, data, rules & decisions

The rules that let an agent act without going wrong

  • A suggestion always shows its reason — the traveller decides, the engine explains.
  • Demo content is labelled as demo; nothing unverified is presented as verified.

02½ · The architecture

The shape of the agentic AI.

Pattern Map-as-canvas planning The map is the working surface; AI proposes with reasons and never rearranges your trip behind your back.

Most travel AI hands you a finished itinerary you did not shape and cannot argue with. Wikistay inverts that: the map is a canvas you draw on, and the engine works alongside you. Suggestions are generated deterministically from the journey state — what is nearby, what is still missing, what fits the days you have — and every one of them arrives with the reason attached, so a traveller can disagree in one glance. Journey AI sits at the seam for the genuinely open-ended questions, called through the Brain Gateway like every other BIXSO product.

fig. — map-as-canvas planning, the pattern behind Wikistay

01

The map is the interface

Stops are dragged, added and reordered on the map itself — the plan never leaves the geography it belongs to.

02

Reasons, not verdicts

Every suggestion carries the reason it was made, so the traveller stays the decision-maker.

03

Deterministic first, AI at the seam

Routing and suggestions run on rules that cost nothing and never hallucinate; the model is reserved for open-ended questions.

04

Own the data, rent nothing

An open tile stack with no key and no per-view fee, so the place dataset — not the map — is the asset that compounds.

03 · The agents at work

It ships as agentic AI — verified.

A planning companion that reads the journey as it stands and proposes the next stop — with its reasoning visible.

A01agent
Journey suggestions

Reads the current journey — stops, days, gaps — and proposes what to add next, with the reason shown.

A02agent
Journey AI

Handles the open-ended travel questions the rules engine cannot answer, called through the BIXSO Brain Gateway.

What the product does

  • 2D/3D map across five service layers
  • Multi-day journeys drawn on the map
  • Drag-and-drop stops and custom places
  • Business connect along a route

Guardrail. Suggestions always show their reason, demo data is labelled as demo, and location is never stored without consent.

In development

In development — a 2D/3D map interface across five service layers (stay, eat, rest, tour, health), multi-day journeys drawn directly on the map with drag-and-drop stops, and a suggestion engine that shows its reasons. Built on an open map stack with no per-view licensing, so the place data itself is the asset.

Frequently asked

Questions about Wikistay.

What is Wikistay?

A journey you design on the map itself — stays, food, rest stops, tours and health services as layers you plan across, instead of ten tabs and a spreadsheet.

Is Wikistay available now?

In development. In development — a 2D/3D map interface across five service layers (stay, eat, rest, tour, health), multi-day journeys drawn directly on the map with drag-and-drop stops, and a suggestion engine that shows its reasons. Built on an open map stack with no per-view licensing, so the place data itself is the asset.

How does Wikistay use agentic AI?

A planning companion that reads the journey as it stands and proposes the next stop — with its reasoning visible.

What keeps Wikistay safe and accurate?

Suggestions always show their reason, demo data is labelled as demo, and location is never stored without consent.

What it means for you

Want this — grounded in your business?

Wikistay is the BIXSO method, shipped: a business model, analysed with rigour, running as governed agentic AI. We start with the questions, not the tech.

Book a free consult