The bet is that it will be trivial for them to invest in cleaning up Claude Code whenever they face real competitive pressure to do so. My best guess is that it's a bad bet - I don't think LLM agents have solved any of the fundamental problems that make it hard to convert janky bad code to polished good code. But Claude Code is capable in my experience of producing clean code when appropriately guided, so it's not that there's no choice but jank. They're intentionally underinvesting in code quality right now for the sake of iteration speed.

▲ woeirua 6 hours ago | parent [-]

Have you tried just asking CC to make a codebase more elegant? It’s surprisingly effective up to a point. No reason to think that won’t work better down the road.

▲ SpicyLemonZest 5 hours ago | parent [-]

Down the road AI is smarter than all of us. Today (including one time literally today), my experience is that it’s occasionally helpful at cleaning up its own mess but often tries to change behavior in a way that’s unacceptable for a production project.

	▲	andrekandre 3 hours ago \| parent [-]
		`> often tries to change behavior` yes, and (in my experience) at the same time re-write the unit tests that are supposed to guarantee behavior doesn't change...