A retro audio time capsule

My friend Jeremy and I were kicking ideas back and forth. We came up with an idea for an audio time capsule where recordings are returned to your future self. We built a series of prototypes and finally launched a version to the general public. Now available on both web and iOS, our pilot users have completed nearly 200 recordings in just the first few weeks.

The day we first discussed the idea, I whipped up a prototype. Using the native iOS shortcut functionality, I made two custom shortcuts. One to record an audio message, and a second to return the messages. This initial prototype took about 2 hours to figure out, and we used it for a few months. My partner, Sophie, and I used this first prototype to capture moments in our relationship together. Jeremy used it to celebrate progress in his recover from an injury.

After a few months of using the shortcuts prototype, we decided to build a real version. Because I had been using the original prototype to record moments with friends and with my partner, I wanted to support sharing in this next version of the app. The app was balancing between being an audio journaling app and a minimalist audio social media. Wary of social media, we felt it was important to keep the app design minimalist and nod to the nostalgia of our childhoods.

Having never built a native app before, and with limited React experience, I opted to use React Native to build the app. I found Expo as a solution that offered much of the necessary functionality in easy to use packages. While the front-end work is straightforward for me, the backend work is way outside my comfort zone. Cursor did the heavy lifting there. I found that I could write clear product requirement as if I were writing tickets, and cursor would brrrrrrrrrrrr until it spit out functional code.

We launched the iOS app and had convinced a few dozen friends to give it a shot. There was lots of praise for the concept and design (acknowledging the friendly bias here), but downloading an app was too much friction for some. To succeed as a consumer product, we had to make it easier for folks to try out the concept.

In theory, the Expo app could have worked on web. I had included a bunch of native app functionality — contacts, FaceId, etc. — that would make launching that same codebase on web a bit more difficult. We also acknowledged that the iOS app was perhaps too complicated for a true mvp. Building a web app was an opportunity to take a step back and reconsider the mvp.

Fortunately, Supabase is easy to work with, and front-end web is easy to build. I started a separate Next.js repo to launch the web app. This version only supported single player recording and listening to returned recordings.

We launched in early February, 2025. We've since onboarded close to 100 users that have completed a combined 200 recordings. While these are small numbers, it's something — and I can now say I've built an iOS app!