The AI feature seems to just generate random icons unrelating to the input.
i'd be embarrassed if I made such a thing, and there's no way i'd show it off to someone else. this feels like someone took a "learn how to use AI in 10 days" and released the first project they worked on. i read "teach yourself PHP in 10 days" in 1999 over a weekend, and was banging out a website with a database in no time. i'd never ever ever show the code written then to anyone, but at least it did what was advertised on the box.
Loads of people on HN could likely build something 'better' by some metric, but of those, how many would still have the nerve to post it in public? I'm not sure I would. Although I've got something I'm planning to post to show HN sometime soon, my default mode is telling me not to post - someone might post something negative, and I'll be embarrassed!
FWIW, the OP isn't sharing the code, and you wouldn't have needed to show your original 1999 PHP code; just share the built service with folks. That's all most folks care about - the utility/value.
That said, there's nothing (yet?) compelling about this service that I've not seen done before. But it might be new/novel to some.
right, but that's why i followed up with "at least it did what was advertised on the box". this product is so bad at making "logos" that it teeters on the verge of not doing what's advertised on the box
Most developers, including myself, fall into this trap. The product doesn't have to be sophisticated, innovative or pretty... as long as it solves a problem.
- I don't test my code
- I don't use git branches
- I don't change my tech stack
That's phrased like a brag, but I'm not sure what's so cool about making 17 failed businesses in 2 years. Working through failure can lead to success, yes, but simply failing many times over does not mean you'll automatically succeed at one point.