Inspiration
We've stolen 10 000 years of peoples time with our games. Now we want to give them their time back, by creating time saving products and services. We've interviewed several business people about their flight habits, and most of them value their time more than anything, and spend between 10-60 minutes booking their travel. With our gaming background, and inspiration from the ease of use with Slack, we are creating AVA, your AI travel agent.
How it works
We believe that a long distance business traveller wants a one-step booking process for traveling, which knows their preferences and saves them time. AVA is a text-based interface layer (SMS) to book flights. Our target customer wants to pay for AVA if it’s time saving, convenient and easy to use.
AVA is an AI travel agent where the target customer is a business traveler that:
- find time more valuable than price
- book travel themselves
- open to a one-step booking process
- not loyal to just one frequent flyer program
More technically: A users sends a SMS with a flight request (i.e. “I want a one-way flight from San Franscisco to Los Angeles tomorrow”) to a number handled by Twilio. Our server gets the SMS body, parse the information using NLP by a service called Wit.ai and sends a request to Amedeus. The best options are sent back to the user with an SMS.
Challenges we ran into
US business travellers seems more loyal to their frequent flyer program than European business travellers. This might be due to the number of different airline companies operating on the same routes in Europe. We don't know if the people we interviewed in the departure hall are fully representative to the target customer group. People want to see their choices when booking flights.
Accomplishments that we are proud of
We went out and interviewed people to test our hypothesis before doing any programming. We created req.spec. and screenshots of the application before doing any programming.
What we learned
Our first hackathon. Learned a lot and will post a retrospective blog post on our Medium account early next week: https://medium.com/@Megacool
What APIs we’ve used from Amadeus
Airport Autocomplete Low-Fare Search API
What's next for AVA
Improve prototype based on feedback from judges Integration of more Amadeus APIs (i.e. Extensive Flight Search API, Flight Inspiration Search API, the Booking API) User test with target customers
Log in or sign up for Devpost to join the conversation.