Along The Way

A social app for nomads that matches people by real journeys - anchor points and headings, not fixed locations. Fun, flexible, offline-friendly, with smart matching and built-in trust network.

Comment

Inspiration

I am a software developer by trade and, believe it or not, this is my first ever hackathon! I know the concept well but I've never participated in an actual competition like this before. The Shipyard concept inspired me for three reasons:

  • I think it's about time that I start accepting the fact that my career skills as a software developer are changing. The Shipping Container deals for this hackathon gave me the perfect opportunity to try building a real product fast with the latest tools without having to commit to the expensive subscriptions and see how the industry is changing.
  • I've built mobile applications in the past and tried to integrate with the native app store payments system. It was a pain. I should have used RevenueCat on previous projects and wanted to try it out on a new one.
  • Whilst I do have a product-focussed career path I've always been far more on the technical side. This means I usually don't give my ideas enough focus on the marketing as they need. When I saw the idea of pairing with an influencer I saw an opportunity to solve this problem whilst focussing on what I do best. I got inspired to create something that might actually solve a problem that a real audience actually needs, and that sounded like a nice thing!

What it does

  • The app focusses on making designing and sharing your travel route fun rather than burdensome and then uses a sophisticated matching algorithm under the hood to match you with like-minded people you can meet up with along the way.
  • The app is designed for the practicalities of travellers:
    • Caters for all kinds of travellers: Those who fly around, those that wander, and those that don't even know where they're going!
    • Offline first: App caches data when your offline so you can use it when you're offline.
    • Safety first: Built in social media connections and vouch network allows you to have confidence in your meetups before you accept.

How we built it

I tried to lean heavily on the shipping container deals and make use of the discounts with the vibe coding tools to prototype the app. After starting with some user research and planning (AI-assisted) I built multiple prototypes with Anything, Vibecode, and Rork. These platforms were not capable of deep server-side capabilities however and I quickly exceeded monthly limits tinkering with the UI and design. So I exported the best one out of those tools to code with the resulting application in Claude Code directly. The application is built with React Native as that is what these vibe coding tools use. However, I have also used Supabase for database and auth, Railway for the server deployment, and Expo and RevenueCat to manage purchases and releases. Most of the code was written by an AI and reviewed by me to build the app within the deadline - certainly a learning exercise in what AI is capable of and what it is not.

Challenges we ran into

  • The no-code vibe coding tools burn through their monthly subscription limit very quickly (I went through mine in two days!)
  • I think I had more issues unravelling the no-code tooling from the app code when I wanted to export it than if I'd just built it myself with the tooling that I know. I also usually do app dev in Flutter and all of these tools use React Native. That tooling switch, even with AI's help, was more of a pain than the initial prototyping helped me.

Accomplishments that we're proud of

  • Designing the matching algorithm that works well matching people not just on exact locations or overlaps but distances and headings.
  • Putting in the audience research beforehand to find out what they value and working it into the features of the app. I'm not part of the audience myself for this app so I can't wait to see how they respond and work on the first direct feedback.
  • The aesthetic and feel of the application.
  • I only realised that the hackathon existed in week 2. I've pulled a few all-nighters trying to get the full app released. I'm proud to have submitted it.

What we learned

  • I've learned that AI's very rarely fully complete the task you give them. It's much better to have multiple review stages which are also AI-assisted the find gaps and flaws in any implementation. It takes a bit longer but definitely worth it for the result.
  • I'm not sure I'd have been able to complete the project to the same standard without any software knowledge, but that's difficult to know for sure considering I do!

What's next for Along The Way

  • Release the app to real users and start getting some direct feedback on what they want most!

Built With

Share this project:

Updates