unthinkable
Case study · A US travel services provider

Travellers get a day-by-day plan and budget for their trip in a few clicks, instead of hours of research.

The client wanted travellers to plan a personal trip within their budget without spending hours researching, and without being steered by agencies' commission deals.

United StatesPublic transit & mobility (live tracking, ticketing, ride-sharing)

What changed

2 monthsto develop and launch the MVP (as reported)
<10saverage planning time (as reported)
20kactive users on the platform (as reported)
50kitineraries planned (as reported)
What stood in the way

What stood in the way.

What stood in the way
  • Travellers spent hours researching destinations, building itineraries and keeping to a budget.
  • The client needed a market-fit MVP, launched in less than 2 months, to test its idea with real users.
What changed
  • 2 monthsto develop and launch the MVP (as reported)
  • <10saverage planning time (as reported)
  • 20kactive users on the platform (as reported)
  • 50kitineraries planned (as reported)
What we built

What we built.

  • A one-question-at-a-time questionnaire where travellers click through options instead of typing, then receive an itinerary split by day and budget.
  • Itineraries, activity and hotel suggestions generated with OpenAI's GPT-3.5 and shaped through prompt engineering.
  • Live data from Expedia, Skyscanner, weather APIs, Google Maps and Eventbrite, so suggestions reflect real options and events.
  • A total estimated budget per person covering stay, transport, activities and meals, and itineraries shared by email or link or exported to PDF.
  • Node.js and React on microservices that scale itinerary generation and recommendations separately, with autoscaling, load balancers and a NoSQL database.
  • OpenAI GPT-3.5
  • Node.js
  • React
  • Microservices
  • NoSQL
  • Expedia
  • Skyscanner
  • Google Maps
  • Eventbrite
What it shows

What this work shows about us.

Generated plans, grounded in real options

The model writes the itinerary, while flights, hotels, weather, maps and events come from live travel services.

An MVP in weeks, built to grow

Launched in 2 months on microservices that scale each part on its own during peak seasons.

Want to build a similar AI trip planner? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.

Start your plan ↗
If we built it today

If we built it today, with AI.

What we would add now, moment by moment.

Booking

An agent would go beyond the plan and hold the flights and hotels the traveller picked, then ask them to confirm before anything is paid. The traveller still makes every booking decision.

A delay

An agent would watch the trip for a cancelled flight or bad weather and offer a changed day plan within the same budget, for the traveller to accept.

Ask Unthinkable

Want this for your business?

Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.

Send us a note instead

Required

Thanks. A senior AI engineer will reply to you.
u
Ask UnthinkableClarity first, then the right team
Your conversation

An AI agent. This prototype uses prepared answers and sends nothing.