As for engagement, looking at the number of comments alone doesn't tell the whole picture either. If an author replies to every comment, that means that the engagement inbound is half of the comment number. Then how many of those comments are "Nice post boss" or "I found this very interesting" - these should be disregarded as there's a good probability that the person didn't even read the post. Just as there's (subjective) value in post quality, the same can be said for comments and engagement.
You get pretty good responses which shows that people actually read your content and take some time out to write a decent reply. It's not always the case though if you look around. Generic comments I think should just be muted, they aren't adding any value to the chain (but that's just my opinion).
Yeah for sure, a quick glance at the comment section usually helps with that though. For instance there's days I don't even read the full post of certain authors like tarazkp as I know he's not just shitposting or trying to take advantage of autovotes, etc, often times I find myself reading more of the comments rather than the full post. Just as an example as he's someone I've curated and supported with my votes for years manually.
Same can be said about ocd voting, we often get recommendations/nominations on what to curate, it's not necessary that we always re-read everything other curators/community curators have recommended for votes but a quick glance at the comments on the author's post or the comments of the author themselves usually paints the picture quickly and easily. You can tell if they've at least been trying to not just reply to comments on their own posts but attempted to build connections with others, even if they're too new to not receive a lot of engagement themselves, that's usually enough for us. The thing I don't like is when authors don't even try yet somehow get big and consistent votes landing on their content.