Air Concierge

Booking detail with trip feed
Trip feed showing offers for travelers whose trip has been cancelled

Inspiration

Imagine you are traveling internationally for work, and you have been told explicitly that you needed to be ready to hit the ground running, by 11 am on Monday.

And then your flight gets cancelled. How would you feel?

A few years back I was traveling to France as part of a consulting engagement, I decided to bring along my wife, so she could relax on the beach.

While we are waiting for the time to board, we get a notification that our flight was cancelled. As we are trying to figure out what to do next, we realized that there are already hundreds of travelers queuing at the airline counter to get re-booked.

It took over one agonizing hour and 45 minutes, to reach to the counter, where we were given a couple of paper vouchers and booked on a flight leaving NEXT DAY.

My one day cushion was gone, and I could not afford any more hiccups.

In the end, we landed with about 45 minutes to spare, but no luggage.

What a nightmare!!

Did I mention that it was a formal meeting and wearing a tie was expected, and I had to show up with what I was wearing.

This entire experience left me unsettled and wanting to find a way to remove friction from the process of recovering from a flight disruption.

And that is why I created Air Concierge! Let's watch the demo

I envisioned a system that could alleviate the stress of rebooking flights, sparing travelers the agony of endless queues and lengthy phone waits. The goal was to eliminate the need for tedious manual interventions and provide swift, automated solutions.

Something so simple that would make you go: "Wait, what!!? is it done?"

Air Concierge is a solution that has the potential of changing how millions of persons travel

What it does

It allows a traveler to book flights and then it provides the traveler with a trip feed where relevant and contextual information is displayed

How we built it

Using event driven architecture the system listens for messages in a Kafka cluster, particularly the "flight_updated" topic, and different microservices react to the message, depending if it was a delay or a cancellation.

Challenges we ran into

Making the user experience simple, underwhelming if you will, so the traveler can go back to whatever they were doing: catching up with work, reading, gym. etc. This is the second time I work with an Openshift environment, so figuring out how to build and deploy pods, why pods were not deploying (likely due to exhausted resources).

Accomplishments that we're proud of

We started only 2 weeks out, and it was a mad dash to have something functional. Through project I was able to take my first steps into python programming, Jupyter notebooks and AI.