2026-03-22 07:34:30

Using Codex has its own new pain points:

Architecture migration always drags on—it's a common problem with every AI, but it's especially pronounced with GPT 5.4, with various messy compatibility layers.
Additionally, every time Codex makes a change, it adds tests. The test files end up being larger than the actual implementation! While it looks stable on the surface, it's really tedious and slow.
The result is that refactoring becomes extremely, extremely slow! And each time it requires me to say "continue"—I've already been sitting at my computer for a solid 2 days, and Codex is still working on the same thing!
Below is Claude's sharp critique of Codex:

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

2 Likes