I remember reading someone’s weeknotes rule somewhere: never skip a week. And never worry about skipping a week. Life happens.
Well, life happened, and here we are a couple weeks later. But honestly? I’m glad I waited until now to write this because the past week brought something worth celebrating.
Zenkai is live
Over the weekend, I finally opened the doors. Zenkai is now available to everyone—no waitlist, no barriers. Just go to zenkai.ee and start analyzing your data.
This sounds simple, but it meant:
- Getting the experience and product to a stage where I wasn’t afraid to share it publicly
- Completely reworking the landing page and adding the main CTA
- Building out a docs page and creating the content from scratch (fortunately I had user feedback with questions)
- Setting up a changelog
- Preparing and sending the email to everyone on the waitlist
That last one felt particularly good. There’s something special about finally being able to say “here it is” instead of “thanks for your interest, we’ll be in touch.”
Learning Loops
Part of launching meant figuring out how to actually send that waitlist email. I’d been putting off learning Loops, but turns out setting it up with a custom domain was pretty straightforward once I sat down with it.
One of those tasks that feels bigger in your head than it actually is.
The Lovable struggle is real
Here’s where I need to be honest about something that’s been eating at me: Lovable is becoming… frustrating.
I remember seeing someone mention that Lovable gets “very chatty” and not understanding what they meant. Now I feel it.
The tool seems to make more errors than it used to, even when I try different approaches—thinking through requests with Claude first, getting advice from a seasoned CTO friend, breaking things down into smaller tasks. Nothing seems to help consistently.
But the real killer is the editing UX. It’s just painfully slow and clunky. Maybe I’m spoiled by tools like Framer’s website builder, but when you’re trying to iterate quickly, every friction point compounds.
Case in point: I couldn’t get the changelog working in Lovable at all. Ended up building it myself with Claude’s help instead, and it was done in a fraction of the time.
The tool crossroads
This is pushing me toward a decision point I’ve been avoiding. Claude Code keeps looking more appealing, but that’s another learning curve to climb. Time is always the constraint, and I need to be thoughtful about these tradeoffs.
What’s becoming clear is that Lovable seems great for certain things—adding new buttons, new pages, extending marketing sites. But maintaining a fully-fledged product there? I’m starting to doubt whether that’s the right fit, or whether I just haven’t figured out the right approach yet.
Community energy
One bright spot: the discussions in the Lovable Shipped community have been awesome to follow. There’s real energy there, people sharing wins and helping each other through challenges
Reminds me why building in public feels right, even when the tools are imperfect.
What’s next
Zenkai is live, the first real users are finding it, and I’m getting actual feedback on real use cases. That’s the milestone that matters most.
The tooling questions can wait. For now, I’m just focused on making sure those first users have a great experience and that the product delivers on its promise to help teams find real signals in their data.