Meet Zoi: An Assistant for Short-Term Travel

For most people, traveling is a pleasure. A nice vacation, a getaway, a break from daily life. However, for people who travel weekly, the daily hassles of traveling are quick to add up. This past week, I experienced 24 hours in Houston. As I had to take 6 Ubers, find a decent restaurant that was close by and open late, and struggled to find new activities that I could do with my limited free time, I knew I was an opportunity to create a fulfilling and smooth travel experience.

Zoi, as a mobile application or built-in for a device, would create a smooth and fulfilling travel experience by giving recommendations(through notifications) for a specific market of short-term travelers. These recommendations could range from touristy things to do to restaurants to anything important for the user. Short-term travel is defined as from any trip for an hour to five days and can be repetitive. This market could consist of students who travel for students/recruiting events/weekend trips and consultants/executives/people who travel for work. Zoi considers your personal interests, personal data, and time when you travel by creating a relationship with the user.

An example of a notification from Zoi (recommending the optimal time to call an Uber from the airport *based on data that I never check bags and the current traffic in the city)

Some of the current problems I found in my past trip with the current resources were:

  • It was incredibly hard to find a restaurant that fit my personal preferences that was near my hotel and open after 10 pm
  • Many tourist recommendations on Yelp. TripAdvisor, etc. do not fit my personal preferences (ex: I would like to visit an art museum because of my art history class rather than the Astros’ stadium)
  • Other resources cannot provide recommendations based on my personal purpose for travel. For an example, if I am traveling for an case competition, I do not want to try a new food because I’m afraid of getting sick.
  • Many recommendations do not consider the amount of time the user has. I had approximately 2 hours of free time in my trip while most recommendations required more free time.
  • There is a plethora of transportation options which I had no idea which one was best at the time of travel.

Let’s take a trip to New York with Zoi.

Part 1:

A few days before, Zoi would send an upcoming trip agenda notification based on my time availability that he read in my invitation email. Zoi knows that I land in Newark around 9:00 pm on Thursday and am busy with a program on from 9:45–5:00 on Friday.

An agenda that would notify the user at an appropriate time before the trip

This list is generated by the baseline of my trip. Zoi knows that this trip is not for an interview, but still a for a business reason so it’s a high priority. Therefore, by knowing the purpose and time of the trip, Zoi recommends activities based on combining my interests, data and timing. Some examples of what could be generated on my list could be:

  1. Checking which friends will be in town. Zoi knows that this trip is a high priority, but also that my friend Andrei who goes to school in London will be in New York. Due to the purpose of this trip, Zoi wouldn’t recommend usually meeting with friends, but he realizes that this might be my only chance to see Andrei in years.
  2. Zoi looked at my personal interests such as what I have been watching on Netflix. Rather than having to go through the same Yelp-generated list of popular things to do in New York that I might have already done, Zoi looks at my personal interests such as Netflix to recommend activities.
  3. Zoi knows that my New Year’s resolution is to grow closer in my faith so he finds that St. Patrick’s Cathedral is close to my hotel and has a Friday morning mass.
  4. Zoi saw that I had liked Dumpling Galaxy’s Instagram’s photo last week, which is on the way to the airport.
User highlights which recommendations he/she prefers

As the user, I can highlight activities that I want to “set in stone” in green, am interested in in yellow, and not in red. If I highlight dinner with Andrei in green, Zoi would give me the option to send a customized message or it could automatically send a message to him asking to confirm the plan. If Zoi saw that I highlighted the Broadway show in yellow, it might confirm that I had been price-sensitive on similar activities in the past, so it could show me a cheaper ticket for the show.

Part 2:

Zoi does not only serve as creating a personalized itinerary, but also gives real-time recommendations.

A few examples could be:

“Let’s call it a night? You have a big day tomorrow!” *Based on data that I have only been sleeping 3–4 hours a day this week and the purpose of the trip

“Ready to call an Uber? Limit wait time by requesting one now” *Based on data that I never check bags and the current rate of traffic

“You have 15 minutes to spare. Want to go to MOMA?” *Based on data from my current class schedule which has art history and the price of the activity

The format of these recommendations could look something like:

*Based on the assumption from data about the purpose of my trip(so that I don’t want try new foods), the current time, and past responses

With Zoi, users can maximize their time and experience wherever they go. Say goodbye to scanning through Yelp and TripAdvisor lists to find an activity that you haven’t done already and fits your preferences and time frame. The possibilities with Zoi are endless. The way Zoi frames questions can change based on the user’s personality or mood. The connections from data based on current heart rate, past responses, and current news could create highly-personalized recommendations. Time is money. With Zoi, you can make sure you use your time during your short-term travel in the most effective and fun way.




A 15-week course focused on techniques for designing products powered by AI that form new relationships with users. Written by students of the Advanced Design for AI course at the University of Texas in Austin.

Recommended from Medium

Airline Lost My Bags and I Have a Big Mountain to Climb

Road Tripping on King’s Highway in Jordan

“A tourist is an ugly human being”: On Reading Jamaica Kincaid During a Global Pandemic

Beyond Chichen Itza — The Responsible Traveler’s Guide to Yucatán

Using Credit Cards to Get More out of Your Ad Spend — AdLeaks

Lessons from Peru

4 Common Misconceptions About French People and 1 Rather True Stereotype

Budapest — Vuelta a Europa, a European tour. Part I. — Besides the Obvious

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Martha Czernuszenko

Martha Czernuszenko

Data Science for Public Good @ UVA. Responsible AI @ AI Global. KodeWithKlossy. Avid Hiker. Bruncher.

More from Medium

Cognitive Computing in Industry 4.0

Tackling today’s toughest energy challenges at CERAWeek

Universal Orlando is one of the premiere theme parks in the U.S.

Car Rental: what to pay attention to while picking up a vehicle