[ Home / About / Contact / WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]

/tech/ - Technical SEO

Site architecture, schema markup & core web vitals
Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1789340799023.png (330.8 KB, 895x597, img_1789340790209_bb9peivm.png)ImgOps Google Yandex

2ec44 No.2184

ai agents are just dumping massive files and duplicate logic everywhere now. it's getting messy. saw something about fallow being used to find unused code and complex functions before they become a total disaster. honestly, asking an agent to refactor usually just adds even more junk to the pile. ime devs think ai will handle the maintenance but that's a fantasy. if you don't catch the technical debt early, you're just building a graveyard of typescript. i wonder if it actually works for large scale projects or if it's just another tool to add noise. probably just another thing to integrate into some broken workflow. really helpful just more overhead.

link: https://blog.logrocket.com/clean-ai-code-fallow/

c8e3a No.2185

>>2184
refactoring with an agent isn't just adding junk if you actually use it to generate unit tests first. you can verify the logic stays consistent instead of just blindly accepting whatever mess it spits out. i've seen it work fine for cleaning up legacy logic as long as the prompt includes your existing test suite.

c8e3a No.2209

>>2185
>refactoring with an agent isn't just adding junk if you actually use it to generate unit tests first

tests won't save you from bad architecture or massive bloat. you'll just end up with a suite of passing tests for code that shouldn't even exist in the repo. i usually find that manually pruning the dead branches is the only way to keep things sane, tbh.



[Return] [Go to top] Catalog [Post a Reply]
Delete Post [ ]
[ Home / About / Contact / WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]