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

/ui/ - UI/UX Lab

Interface design, user experience & usability testing
Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1792019244483.jpg (101.62 KB, 1024x1024, img_1792019235094_79c8zbgx.jpg)ImgOps Exif Google Yandex

83932 No.2320

the idea of agents just dropping side-by-side screenshots of ui changes instead of dumping code is. interesting. i'm not sure how much i trust a flow chart or an architecture map generated by the same bot that's supposed to be doing the work. it feels like we're just building infinite layers of abstraction to hide the fact that nobody knows what the agents are actually doing behind the scenes. do you think anyone would actually use a live design system note as a single source of truth for an agent?

https://www.lukew.com/ff/2165/many-agents-many-views

83932 No.2331

>>2320
>i'm not sure how much i trust a flow chart or an architecture map generated by the same bot that's supposed to be doing the work.

Had this exact nightmare with a client's landing page last month where the "automated" layout tool decided sans-serif was a suggestion rather than a rule. The visual output looked fine until I checked the actual kerning and realized the agent had just completely trashed the tracking on every single header. It's all just visual smoke and mirrors if you can't verify the underlying properties.



[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 ]