@aryanXmahajan: I hate building frontends with AI. Claude, Codex, and Kimi can generate a dashboard in 30 seconds. Then I spend the nex…

X AI KOLs Following News

Summary

The author expresses frustration with AI tools like Claude, Codex, and Kimi for frontend development, noting that while they quickly generate code, the process requires significant manual adjustment to achieve a polished product.

I hate building frontends with AI. Claude, Codex, and Kimi can generate a dashboard in 30 seconds. Then I spend the next six hours explaining that the main button is invisible, the table breaks on mobile, and every screen looks like the same pile of rounded cards. I've installed the design skills. I've fed it screenshots from Linear and Stripe. I've used the “premium, minimal, Apple-level” prompts. It helps. It never removes the babysitting. AI is incredible at writing frontend code. It is still terrible at knowing when a product feels cheap, confusing, or annoying to use. One-shot AI frontend design is a lie. The first 70% takes 30 seconds. The last 30% ruins your Saturday.
Original Article
View Cached Full Text

Cached at: 08/26/26, 11:13 AM

I hate building frontends with AI.

Claude, Codex, and Kimi can generate a dashboard in 30 seconds. Then I spend the next six hours explaining that the main button is invisible, the table breaks on mobile, and every screen looks like the same pile of rounded cards.

I’ve installed the design skills. I’ve fed it screenshots from Linear and Stripe. I’ve used the “premium, minimal, Apple-level” prompts.

It helps. It never removes the babysitting.

AI is incredible at writing frontend code. It is still terrible at knowing when a product feels cheap, confusing, or annoying to use.

One-shot AI frontend design is a lie.

The first 70% takes 30 seconds.

The last 30% ruins your Saturday.

Similar Articles

On AI Coding and Its Discontents

Lobsters Hottest

Cal Newport reflects on the initial enthusiasm for AI coding tools like Claude Code and the growing disillusionment among developers due to hidden bugs, quality issues, and unsustainable workflows, arguing that outsourcing all code production to AI is not currently viable.