You aren’t accounting for managerial politics. A product manager won’t gamble on a large project to lower operating cost, when their bonus is based on customer acquisition metrics.
The use of ai agents allowed them to shrink the problem down to the point where it was small enough to fit in their free time and not interrupt their assigned work.
I get that it's fun and there's personal satisfaction in it, but it just reinforces to management that they don't need to care about allocating resources to optimisation, the problem will just take care of itself for free.
Should it be this way? No. Is it this way in practice? Unfortunately often.
This also explains this blog post
For the managers, it's about a bonus. For engineers it's the existential question of future hirability: every future employer will love the candidate with experience in operating a $500k/a cluster. They guy who wrote a library that got linked into a service... Yeah, that's the kind they already have, not interested, move along.