this post was submitted on 17 Jul 2026
-13 points (25.9% liked)

Programming

27783 readers
244 users here now

Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!

Cross posting is strongly encouraged in the instance. If you feel your post or another person's post makes sense in another community cross post into it.

Hope you enjoy the instance!

Rules

Rules

  • Follow the programming.dev instance rules
  • Keep content related to programming in some way
  • If you're posting long videos try to add in some form of tldr for those who don't want to watch videos

Wormhole

Follow the wormhole through a path of communities !webdev@programming.dev



founded 3 years ago
MODERATORS
 

Few weeks ago I wrote about a side project that I'm using to learn, this is my reviewing what I missed on the project, the main issues after a few dozen games in the platform with friends. And what I learned in the process.

All comments, feedback is appreciated.

you are viewing a single comment's thread
view the rest of the comments
[–] tracyspcy@lemmy.ml 3 points 4 days ago (13 children)

Cannot find answer to the question from title in article. So what went wrong ?

[–] hendrik@palaver.p3x.de 4 points 4 days ago* (last edited 4 days ago) (12 children)

Reads like they started experimenting with an AI-driven workflow. Got some prototypes and early progress fast. And now the first draft kinda works, but it breaks left and right and they're stuck in a loop to fix things, another thing breaks and they're not getting anywhere.

Difficult to say what went wrong judging by what they wrote. That kind of information is missing in their rant. Could be a wrong choice of AI model, Agent framework, wrong workflow... Or AI in general. I also tend to get caught up in loops, yelling at the AI for half a day to fix things, write more strict tests so it can see itself what went wrong, and fix whatever it just broke in the process. Or yell at it for giving up and refactoring the entire project because it can't figure out what went wrong and why.

[–] CrypticCoffee@lemmy.ml 5 points 4 days ago (1 children)

Sounds about right. Fits in with the 80:20 rule. Does the first 80 quicker than the 20% of time it usually takes but does the next 20% which takes 80% of the time even slower.

Effectively, it optimises the quicker bit so you suffer long term and overall waste more time while screwing the planet. It's only fine for prototyping and even then, would that be the best base to build off? Is it modular and maintainable? Does it have low coupling and high cohesion? Or is it just trash?

[–] Shin@piefed.social 2 points 4 days ago

Heavily coupled code with a lot of duplication of code. These are the main points of the refactor.

There is some lack of "smart" use of the JS, like the positional arguments instead of objects (for quick fix/change), no separation of concerns, there is something lightly there, but lacks the gross thing.

load more comments (10 replies)
load more comments (10 replies)