Karo Zieminski

karozieminski.substack.com

Claude Design Review: 48-Hour Builder’s Test + Hero Prompts

I spent 48 hours testing Claude Design on Opus 4.7. The prompts, the credit traps, and why the “Figma is dead” take is louder than it is true.

open.substack.com

ChatGPT 5.2: What It Really Changed, And Why The Internet’s Take Is Mostly Wrong

Why “it feels worse” is the wrong metric for evaluating GPT-5.2.