Why I rebuilt Screenity two months into my new indie life

When a storm knocks you offline, it forces you to rethink everything.

Issue #38

Hey! I’m Alyssa X, a serial entrepreneur who’s built 12+ products solo. This email is part of my maker journey, which you’re subscribed to. Feel free to forward it to others if you find it interesting! You can support me through GitHub Sponsors, I’d really appreciate it! ❤️

Hey all,

It’s been about two months since I quit my job and jumped into indie life. I’m still figuring it all out, but here’s where things are so far.

---

Wind, rain, and wireframes

I thought it’d be a good idea to get away from London’s grey and moody weather for a bit. Soak in the sun, work by the sea, clear my head, get inspired.

So we booked a little stay in Cascais. After all, Portugal in spring just sounds so idyllic - sunshine, fresh air, nature…

Instead, the weather somehow followed us there. But it was WAY worse. It was a full-blown Atlantic storm, with violent winds, trees torn from the ground, crushed cars, downed power lines, torrential rain somehow coming from every direction possible… Not exactly the peaceful seaside retreat I had in mind 😭

To make matters worse, the Airbnb was freezing cold, with a tiny portable heater that did absolutely nothing. The Wi-Fi barely worked, if at all. I found myself hotspotting from my phone, desperately trying to get a signal.

Still, I made the most of it. With the internet down and tools refusing to load, I kind of leaned into the offline mindset: sketching ideas on paper, mapping out the product roadmap, and doing the kind of higher-level thinking I usually avoid in favour of just coding the next thing. Turns out, being forced into analog mode wasn’t the worst thing after all 😅

On the final day (of course) the sun came back. Blue skies, perfect weather, no wind. I got in one really beautiful run along the coast, just in time to head to the airport.

Bye bye timeline

The storm and offline chaos forced me to step back and rethink what really mattered: clarity. In my last issue, I shared that I’d been letting people in the editor and gathering early feedback. And what I quickly realized: no one really understood the timeline 🙃

It worked, but it was multi-track, overly complex, and kind of intimidating. I liked the original storytelling vibe, like slides in PowerPoint, scenes you could easily rearrange. But it was just too confusing. The scene list and the timeline felt like two sources of truth that overlapped badly.

So, I scrapped it all and started from scratch.

Now, the timeline contains the scenes, one after the other, in simple, draggable, modular blocks. You can still trim, reorder, and do everything you could do before. But now it’s just in one view, with no ambiguity.

It’s about time I shared a proper sneak peek, especially after keeping things under wraps for so long. So here it is: the new, simplified timeline and editor - cleaner, more intuitive, and way more “Screenity-like”. Plus, a first (quick) look at the brand new public-facing player experience as well.

One refactor after the other

Even after reworking the timeline, I couldn’t shake the feeling that there was more to fix. That chaotic trip back in March didn’t bring the calm reset I was hoping for, but it did force me to zoom out, this time way beyond the timeline. With no Wi-Fi and nothing else to do, I finally stepped back and looked at the bigger picture: where Screenity fits, what makes it different, and what’s holding it back.

And naturally, that led to… even more refactors. Oops 😬

I began with something Screenity already does well: privacy. While I had already designed it with that in mind, I realized I wanted to go even further. Fully GDPR-compliant, EU-hosted, and as independent from third-party services as possible. And not just for privacy’s sake, but for control, performance, and cost. Which, honestly, matters a lot when you’re building a video tool.

The first thing to go was Supabase. Great for getting started, but not great for what I was building. Screenity’s data is deeply nested with scenes, keyframes, effects… and trying to fit that into Postgres was a constant headache. Plus, Supabase kind of locks you in to the whole stack (auth, storage, functions, analytics…).

So I replaced it with MongoDB, local IndexedDB, and patch-based syncing, so now it’s offline-friendly and far more flexible. It was a bit of a nightmare to architect, especially with undo/redo and all sorts of edge cases, but it was worth it.

Then I tackled rendering. I was using AWS Lambda + Remotion, but it didn’t feel right with the privacy angle. So I built my own distributed renderer. Now videos are queued up with Redis and BullMQ, and processed on Hetzner VPSs. It’s faster, cheaper, and I know exactly where every video goes.

While I was at it, I went all in and overhauled the rest of the stack too.

• Swapped third-party transcription APIs for Whisper.cpp, which now runs locally and still gives accurate word-by-word captions for almost nothing.

• Moved my entire app stack - frontend, backend, logging, analytics… to Coolify, fully self-hosted in Hetzner.

• Even dropped Cloudflare for BunnyCDN, which is EU-based and friendlier.

A bit extreme? Maybe. But in a market flooded with overhyped AI slop and endless copycat tools, I’d rather build something intentional, aligned with my values, even if it takes longer.

This is fine (mostly)

The full-time indie hacker lifestyle has been… a lot.

Some days, it feels amazing. I wake up with a plan, go for a run, get into the zone, and actually make progress. Other days, I spiral, stressing about not having income yet, wondering if I’m moving fast enough, or what I’ll do if this doesn’t work out.

But I keep going. I’ve been staying accountable with weekly goals and checklists in Notion, with just enough structure to keep moving in the right direction.

Just a few weeks ago, my previous company, where I’d been designing since early on, announced it got acquired by none other than Figma. Seeing it was a bit surreal, maybe a little bittersweet to see it wrap up, given the years I spent working on it. But still, I know I made the right choice for myself. Now I get to build my own ideas, on my own terms, in my own messy, chaotic way. Even if it means fixing a one-frame flicker no one would notice, rewriting transcripts for better punctuation, and testing the full recorder-to-publish flow for the hundredth time 😅

Anyway, I’m optimistic and staying on track. I have a launch date in mind (but I’m not going to jinx it by sharing it!). I can’t wait for you to try it.

See you,

Alyssa X