Travlr

flow chart of how we proceeded
rf-forrest severity model

SoftwareHack

Travlr came from the idea that flying could be a lot less stressful if travelers had better access to real-time updates and more direct communication with flight attendants. We know the frustration of waiting for information that often feels like it’s out of reach, so we set out to create a tool that would streamline communication and make travel smoother. By combining up-to-date flight info with an AI chatbot for instant help, we wanted to improve the overall experience for passengers.

Travlr helps users make their flight experience easier by providing a few key features. First, it allows passengers to directly communicate with flight attendants, so they can get assistance without the wait. The app also delivers real-time flight updates, keeping travelers in the loop with everything from gate changes to delays. Plus, there’s an AI-powered chatbot built into the app that answers any questions users might have about the flight, the airport, or anything else related to their journey. How we built it

We built Travlr with a modern tech stack to keep things smooth and efficient. On the frontend, we used React with TypeScript, which helped us create a dynamic and easy-to-navigate web app. For the backend we used SQL, Flask, and Amazon RDS. The chatbot, which really ties the whole experience together, runs on GPT-4, making it a smart and conversational assistant that can handle a wide range of questions.

One of the biggest challenges was that some of us were a bit out of practice with key technologies like Tailwind and React. Since the last time we worked with them, some things had changed, and we had to spend extra time getting back up to speed. This slowed us down a bit, especially when it came to using Tailwind’s utility classes and the new features in React. We went from using react to react native to flutter to back to react. It felt like we were relearning some things, and that definitely caused some frustration. But in the end, we got the hang of it and were able to move forward.

Despite the hurdles, we’re really proud of what we accomplished. The flight data integration came together nicely, and it’s something we think will really make a difference for travelers. We’re also happy with the flight attendant communication feature – it’s simple but super useful. And, of course, the AI chatbot turned out to be a huge win. It’s capable of answering a wide range of questions, and it’s fast, which is exactly what we wanted to make the experience feel seamless.

Looking ahead, we’ve got some exciting ideas for Travlr. We want to add more AI features, like language translation, so international travelers can get the help they need no matter where they’re from. We’re also looking to enhance the frontend more so we can give our customers a more efficient web app. A mobile version of Travlr also seems more ideal!

https://drive.google.com/file/d/1Ngm99gq4TzfviQfVfjaAAW9PjSeTMCko/view