I take a peak every month or so at spend for my company and notice more and more are consumed $1k in tokens a month and it is bewildering to me how. I use llms daily, and see anywhere from $200-$400 tops. This is using the most expensive models, in deep thinking mode. So I'm not a Luddite against the usage of them. I just can't figure how _how_ to burn that much money a month responsibly. I genuinely challenge someon…
I'm on the same page. Do people not analyze the problems themselves? Are they just copy/pasting their entire ticket description into Claude Code and having it iterate until they land on something that works? I don't get it.
If your ticket description is large enough to put a dent in your context window, or otherwise end up as a meaningful share of total token expenditure, then something is very, very wrong with your ticket workflow. Nobody needs a 50-page ticket.