> At work we're using an LLM to analyze Windows crash dumps, which turns out to be quite expensive -- several dollars per dump, and you might analyze many every hour

Is that in any way useful?, how so?, are dumps from optimized builds?

I've found that most of the time I don't even need to open a dump because of regular automation providing all thread callstacks in tickets... That and logs are generally enough...

▲

rwmj 2 hours ago | parent [-]

I'm not directly involved in the day to day activity around this, but as I understand it, it's about triaging related crashes together and associating them with prior bugs and suggesting fixes, else escalating.

	▲	mittensc 5 minutes ago \| parent [-]
		thanks! Associating with past bugs is a nice use case Suggesting fixes seems risky, can lead people on wrong paths and waste time or fixing a sympthom instead of proper root case and waste the bug report? Grouping is pretty easy without LLMs, just pattern match callstack symbols Triage, sure, but a human would be able to do that as well in a couple of minutes for free with callstack in tickets