Keeping the backlog funnel full has been nearly a full ltime job for me for about 5 years now. Both pre and post AI. Once you get a team together that knows how to execute keeping the backlog full of well defined and groomed tasks becomes the problem management rarely sees and subsequently doesn't know how to solve.
This.
I recently switched, but at my previous job, the PMs (who would typically do most of this work) got so overworked after AI took hold, that they’ve doubled the PM-to-dev ratios on all (the productive) teams.
I have read about this in other companies as well. Some teams in Google have a PM pr dev, I hear.
Where I work now, sh*t gets done MUCH quicker with AI, and to a higher quality as well.
IMO, if you work with software and can’t get a meaningful boost out of AI, I think you’re doing something wrong.
I think that's why you keep hearing about all these tech stack rewrites.
Many of these companies have no idea what to do to begin with, so improving dev throughput is not really the business problem to begin with (even though frequently blamed).
Trying to fill a backlog to just keep a dev team busy is not what a backlog is for. It would be better for the business to have them rake leaves out front than add to a system something no user needs for the sole purposes of looking busy.
I think the assumption is that there are clearly problems to solve and useful work to do, but that useful work is not always well-defined enough to be ticketed and represented in the backlog as individual parallelizable work items.