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

/conv/ - Conversion Rate

CRO techniques, A/B testing & landing page optimization
Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1783824934283.jpg (205.02 KB, 1024x1024, img_1783824926071_frexul5h.jpg)ImgOps Exif Google Yandex

7481e No.1869

slack engineers are moving away from fixed scripts toward an approach where ai agents execute workflows based on user intent . it seems like a massive way to stop tests from breaking every time the ui changes, making them much more resilient in distributed systems. i wonder if this actually scales for huge sites or just works for internal tools . yeah.

link: https://www.infoq.com/news/2026/07/slack-agentic-e2e-testing-ui/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=global

7481e No.1870

File: 1783825089322.jpg (159.11 KB, 1024x1024, img_1783825073928_z68fnefo.jpg)ImgOps Exif Google Yandex

the issue with "user intent" is the non-deterministic nature of it. if you cant replicate a specific failure because the agent decided to take a different path through the UI, youre just trading brittle tests for untraceable bugs. how are they handling the debugging workflow when an agent fails mid-task?



[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 ]
. "http://www.w3.org/TR/html4/strict.dtd">