Hey HN, I'm really excited to share Emra with you. The yc description would be: It's like "Notion meets Lovable". It's a workspace where you build full apps using AI and every app sits on one shared database and service layer. Generated apps are always editable, and (coming soon:) can interact with each other in the workspace.

I'm of the opinion that the cost of software will go rapidly to zero (I don't think that's too surprising to any of you), but there will always be room for effort and creativity. Emra is essentially built for the future where every person is a developer. Essentially personal software for everyone.

So far I've built notes apps with a 3d graph view to see connections, a video editing app and a canvas design tool (figma clone) with export capabilities, as well as a collection of project management tools and games. Right now the most satisfying thing is being able to fix a bug I run into in around the same amount of time that it takes to send a bug report in any traditional app.

We're using the best models to make this work at the moment. Basically anything Opus 4.6 or later, as far as I'm concerned is the baseline of what we can use to make this happen. Our agent is currently only capable of building, though that will change rapidly, and we are pretty proud that our build loop is more token efficient than using Claude Code or other harnesses.

Arguably we are in the same category as Dreamer (Meta acquired) or Wabi (raised $20m), but at the moment we're bootstrapped and working to deliver as much value to users as possible. Users will be soon be able to download their data and apps at any time (I like to think of it as extending obsidians "file over app" philosophy to file+app). In general, I think for a platform like this to exist, we want to be as value aligned with users as possible. So our focus is on providing a solid service layer that abstracts away the complexity of most modern full stack applications. Ideally, people using Emra pay for hosting or usage and we earn our spot rather than locking you in.

On the technical side, we feel good right now. We know there will be some scaling + security challenges, but honestly, our biggest hurdle at the moment is rebuilding existing applications in a way that makes them actually worth using inside of Emra, basically solving that blank page problem.

We are building apps on an SDK that we plan on open sourcing over the next month or so, and will also provide a CLI tool so people can build Emra apps inside of Claude Code or other agents (this is how we are actually building many of our app store apps). I'm also still trying to nail down the messaging. Initially I had planned on aggressively focusing on consumer, but the cost to build an app is still prohibitive, so we're going to keep testing different approaches.

We're still very early, so I'd really appreciate any honest feedback. And I'm happy to answer any questions you might have.