More Beautiful QR Codes
Hey IndieHackers!
I was experimenting with artistic QR codes, and I was able to generate really beautiful results sometimes. So I turned it into a tool for everybody. The art is still ai-powered, but maybe I will add a way to upload original design assets and merge the code around them (would you find that interesting?).
Sometimes it takes a couple of generations to make a code that works, but in my opinion, it’s something we can do today that we didn’t have yesterday*. So let’s go :)
P.S.: I've just launched this on ProductHunt today:
https://www.producthunt.com/products/scanwayy
- diffusion codes look different.
Trending on Indie Hackers
The interesting question isn't whether AI can generate more beautiful QR codes—it's whether visual identity becomes important enough that people stop treating QR codes as purely functional. I'd keep validating whether customers come for the novelty, or because branded, recognizable QR codes genuinely improve how people engage with them.
That actually looks really cool, I'm so impressed
Congrats on the PH launch. Honest take on the product and the positioning, since the AI Tools crowd will size this up fast:
AI-artistic QR codes are genuinely cool, but the category has a specific trap you're already brushing against: "sometimes it takes a couple generations to make a code that works." For a QR code, "works" isn't aesthetic, it's binary. It scans or it doesn't, and a beautiful code that fails to scan is worse than an ugly one that works, because the entire job of a QR code is to be scanned. If reliability is a coin flip, that's the thing to solve before anything else, because a designer who gets a gorgeous code that 30% of phones can't read won't come back.
Which points at your actual wedge. The market isn't "make pretty QR codes" (fun, but a novelty people try once). It's "make pretty QR codes that reliably scan every time." Scan-reliability is the hard part and the defensible one. Lead with guaranteed-scannable, not just beautiful. That's what turns a novelty into a tool a business pays for (menus, packaging, posters, where a failed scan costs a real customer).
On your upload-design-assets question: yes, that's the more interesting direction, because it moves you from "novelty generator" to "brand tool." A business wanting their logo/colors woven into a scannable code is a real, recurring, payable use case. The random-art version is a toy people share once. The bring-your-brand version is something a company needs repeatedly.
Who's the buyer you're picturing, individuals making one fun code, or businesses needing branded scannable codes at volume? That decides whether this is a viral novelty or a real tool.