Day 7 since launching Swisy 🍌

A week ago I shipped Swisy, a paint-first AI image editor.

I built it because “small edits” kept turning into long prompt sessions—yet the results still weren’t precise. I wanted a workflow that feels more like “editing” than “prompting”: you paint the exact area you want to change (remove/replace objects) and keep the rest intact.

What Swisy is (in one line):
No complex prompts—just brush-edit with Nano Banana.

The other thing I personally dislike is juggling 10 different models just to get one job done. So Swisy keeps it simple with a small curated set:

• Editing: Nano Banana (Flash + Pro)

• Upscaling: Topaz + Crystal

[What happened in the first 7 days]

This week was mostly about getting real usage, not polishing the landing page.

• People don’t want “AI magic.” They want control. The brush workflow is the whole product.

• The biggest drop-off happens at the first edit: users need a “wow” result within 30 seconds.

• “Simple prompt” doesn’t mean “no guidance.” Users still need clear suggestions (e.g., remove vs replace, common use cases).
  

[What I’m working on next]

Distribution, distribution, and more distribution. Have so small number of "users" to tell if this product is worth something to customers. I have a number of features I want to add, but until I confirm if Swisy provides (or doesn't provide) value to customers and why, I don't think I should develop further. Things I'm focusing on is publishing short-form contents on tt, yt, and ig.


Would love your help
If you’ve ever bounced off AI image editors, I’d love your honest take:

• Is the paint workflow intuitive on the first try?

• Where did you get stuck?

• What feature would make this a “daily tool” for you?

Swisy works for me, I hope it works for others as well. Thanks for reading — I’ll share another update soon.

Swisy: https://swisy.app

Discord (support + feature requests): https://discord.gg/djg9YvFmUQ