baldur

He means Mutants Genesis (Bandai's mobile CCG), Phantom's been posting about it here for a few days. Phantom, worth editing the title to "Mutants Genesis, Latam players are angry" and naming the game in the first line, most people on Lemmy only see this one post.

You may want to search for software development good practices and security. Just create it simply with AI might potentially have a lot of security breaches and code that can be replaced for better performance. It is not a good idea leave all the work to a robot.

I am still exploring PeerTube, as a user, but I am not sure, the UI of certain websites does not give me the confidence, maybe because I am developer and I am really picky with every detail but I am not sure. Maybe I am wrong.

Interesting!! I will take a look, but are they safe? I mean I will not get any surprises later lol

This sounds interesting.\ Is it a paid platform or how it works?

Thanks, noted. Operator details will be added once the business registration is done.

I'm in the same position, I mean, I worked really hard to get MVP done, and I'm just a full stack developer, with 15 years of experience in the industry, and they say it's vibecoded because the UI looks well done haha. That's amazing, but in certain way I see it funny to be honest. It feels more like people those haters tried, failed and they have nothing better but attack XD.

I told one of them, ok, try to create one and see how it goes. An AI cannot do what it has been done here that's why it is so ridiculous when they say it is vibecoded.

Tailwind it’s a utility first framework with no pre built components, I think you are the one that does not know what are you saying, you build whatever you want and can fully customize the theme that’s literally the point of it, and it’s in the official docs.

On the AI and closed source speculation, the auth, signatures, permissions, input handling is standard and went through an OWASP review before launch I said it before. I said I used AI for unit tests, integration tests, performance work, and even to attack my own app. I did not say I let it write the product. After 15+ years writing software, most of us already know AI still ignores SOLID, duplicates existing functions, and introduces security issues if you hand it the whole job. But you’re free to dislike the design or the decision to keep the anti-abuse parts private. The transparency page is public if you want the actual policies and review status instead of the theory.

Again, if you don't like the UI or think this is all robots, you can always join X, Reddit, or any other platform that steals your data and helps big corporations, or you can develop your own and invite me, but for that you need to be developer first. I thought people were different in the Fediverse.

Good question, however the security model doesn't depend on the code being secret, and it shouldn't. The auth, signatures, permissions and input handling are standard and went through an OWASP review before launch.

What's closed is the anti-abuse layer, spam, rate-limit tuning, the signals for detecting ban evasion and coordinated accounts. Those work partly because spammers don't know the exact numbers. Lemmy's and Mastodon's are public and they get played for exactly that reason. Hope it helps :)