Why I built mssgs

I didn’t set out to build a chat app. I set out to find one, and kept not finding it.

Where it started

I’ve used Discord almost since it first appeared. For years it was where my friends and I talked, and I met people there I would never have met anywhere else.

Then, in 2023, I was banned by mistake. There was no warning, and the customer service, run entirely by bots, gave me no help at all. Two weeks later my account was reinstated with a clean record and 600 Nitro credits as compensation: it had been a technical bug on Discord’s side.

Still, being locked out for two weeks over someone else’s bug was enough. That was when I started looking for a new place.

Nothing felt finished

Rocket.Chat and the other alternatives I tried either didn’t meet our requirements or simply didn’t feel finished. The open-source projects kept running into the same wall: the core was there, maybe even maintained, but there were no native iOS or Android apps. A React Native app at best, and that was about it.

Chatting with friends, or working with a team, in an app that feels sluggish was never something I was willing to settle for.

More people, more problems

Meanwhile I had built up a following on TikTok, and I was pulling people into my Discord server from all directions. The problems piled up with them. Bots were hard to build, mini-games were impossible to get working, and slash commands were plain broken. It got on my nerves far more than it should have.

The last straw

What finally did it was safety. With TikTok followers pouring into the community, moderating it became genuinely hard. I ran it for fun. I wasn’t earning anything from it; worse, it was costing me money.

Discord didn’t make it any easier. The moderation tools were poor, and once people moved into one-on-one DMs, they had no proper way to protect themselves. Members were doxxed. We banned the people responsible, and they came back, again and again. At some point, keeping our small community of a few thousand people safe had become a full-time job.

I have never felt so powerless.

Starting again, properly this time

That was the moment I picked mssgs up again. I had tried before, but every time, something else took priority. This time I had the funds to hire people and build it together. That was only possible because of how RUAL has grown over the last few years.

Our first TestFlight, a simple chat app, went out in February 2026, and it was a mess. Almost nothing really worked, and the backend logic was nowhere near done. I had no idea how complicated it is to design a system that scales messages across multiple nodes, let alone across geographic regions.

The mssgs sign-in screen in February 2026: a green glass card asking for an email address to send a login code to — Rebuilt from the app’s own git history. In February, the sign-in screen is as far as a build gets without an account. By July the app had a demo mode, which fills it with sample people and messages.

A direct message in July 2026: two friends planning a meetup, some of it as ghost messages — Rebuilt from the app’s own git history. In February, the sign-in screen is as far as a build gets without an account. By July the app had a demo mode, which fills it with sample people and messages.

Where we are now

I’ve learned a lot since then. mssgs has grown from a simple app into a complete, working one, with a team building it every day.

We believe community moderation is a far better way to fight people who behave maliciously. And thanks to RUAL, which runs local AI servers for us, we can scan public stories automatically and flag them for the community to review.

The #general channel of a community, with mentions, reactions and someone typing — mssgs today, version 3.3.0, in demo mode.

The list of direct messages, each friend with a streak and the latest activity — mssgs today, version 3.3.0, in demo mode.

You can try it yourself at mss.gs.