I Built an Online Image Toolkit in 3 Days with AI — Here’s What I Learned

Ricardo Rose — Sun, 08 Nov 2026 15:13:25 +0000

I’ve always found image processing tools frustrating. Either they’re bloated with features I never use, require installing desktop software, or hide basic functions behind a subscription.

So I built XNOV (https://xnov.io) — a lightweight, browser-based image toolkit. It compresses, resizes, crops, removes backgrounds, and even changes ID photo background colors. All online, no install, no account required.

The twist: I built it in 3 days using AI assistance.

Why I built it
The core problem is simple. Most people need to do one or two quick things to an image — compress it to upload somewhere, crop it for a profile, or change the background on a document photo. Yet the tools available often feel like they were designed for professional designers, not everyday users.

I wanted something that loads fast, does the job, and gets out of the way.

What’s inside
XNOV currently handles:

Image compression — reduce file size without visible quality loss

Resize & crop — standard but necessary

Background removal — extract subjects from images

ID photo color replacement — change the background to white, blue, or red for official documents

The ID photo feature is actually the one I use most personally. There’s a surprising lack of simple tools for this specific task.

The 3-day build process (and what AI did)
I won’t pretend I hand-wrote every line. I used AI to scaffold the project, generate UI components, and iterate on the interface. Most of my time went into:

Prompting and refining — getting the AI to produce usable code, then fixing the parts it got wrong

UI polish — making the tool feel responsive and not like a developer side project

Deployment — bought a domain on Cloudflare and shipped it

The lesson: AI can dramatically speed up the “first draft” of a product. But the last 20% — the details that make it feel finished — still requires human judgment. The gap between “it works” and “it’s pleasant to use” is where most of the time goes.

What I’d love feedback on
This is my first public launch, and I’m genuinely curious:

Is the UI clear? Can you figure out how to use a feature without instructions?

Are there missing features you’d expect from a tool like this?

Performance — does it feel fast enough on your connection?

You can try it here: https://xnov.io

No signup, no paywall. Just drop an image and see what happens.

I’ll be in the comments if you have questions or brutal feedback. Both are welcome.

DEV Community: Ricardo Rose

I Built an Online Image Toolkit in 3 Days with AI — Here’s What I Learned