Context switching between more than 2 threads of work is untenable if you want to really review code in depth. And with AI, you need to go through everything with a fine toothed comb

▲

shiroyasha 2 days ago | parent | next [-]

The same work as any senior software engineer reviewing his teams work, imho.

	▲	lazide 2 days ago \| parent \| next [-]
		Eh, you quickly learn who you can trust and who needs the super skeptical detailed look with humans. With LLM’s you have you to super skeptical of everything.
	▲	aisizbzbzh 2 days ago \| parent \| prev [-]
		[dead]

▲

CuriouslyC 2 days ago | parent | prev | next [-]

Counterpoint, you need to develop more robust automated systems so you don't have to go through everything with a fine toothed comb.

▲

Aeolun 2 days ago | parent | prev [-]

Yeah, I find I need to interrupt Claude at least once every two turns to prevent it from going off into the wrong rabbit hole.