
adventures.run
Jun 4, 2026It's not like it hasn't been done before. I could list off at least five different highly popular products that let you drop some pins on a map and calculate distance, elevation, etc. But none of them worked exactly the way I wanted. I either hated the UX, it didn't quite have the features I wanted, or it had way too many and got in my way.
So why not create an app for myself to plan routes exactly like I want, then see if anyone else could benefit too?
The idea
The idea is straightforward: I want to be able to draw a route on a map, following roads and trails, and add waypoints so I can gauge how far it is between important places. Then I want to be able to export it so I can import the route into my watch and follow along. Bonus points for being able to share it with friends to get feedback and comments, and maybe even plan an adventure together.
What I ended up with is the first version of adventures.run.

How it's built
The stack is super simple, which was part of the goal. It will scale just fine by scaling the VM vertically -- at least for a few hundred users. If it ever gets bigger than that I have plans, but being honest this is just a hobby project and I'm totally fine if it stays like this.
Stack
The application is built entirely with Svelte (specifically SvelteKit 2 and Svelte 5). It runs both the APIs and the frontend. The database is SQLite, which keeps things incredibly simple for both development and hosting.
There are several libraries used for the auth and mapping, including BetterAuth, MapLibre GL, and Photon. Strict typing was a must for this application, so everything is TypeScript.
Infrastructure
Fly.io has been my latest obsession, and it's running several apps for me now. This is the only one I have always-on, and it's still incredibly affordable for a small VM. It will take a whole lot of actual users before I need to bump up the specs, which will be an easy one-click vertical scaling situation.
What's next
At the moment, the project is invite-only. I'm testing it with myself and a few friends to try to find any major issues. I'll open up invites to anyone who wants to try it for a few months, then I'll open it up to the public and maybe even do a half-hearted attempt at some advertising or organic search work.
If it looks like something you'd want to try, head over and request an invite!