Don't have anything to add here, but wanted to plug Nocturnes, a collection of short stories. The one other fiction he's written that's not listed here, I believe. It's very good, like everything he's written. He's a remarkably consistent author. Rewarding to read all of, and very manageable as there are only the 6 or 8 (don't have the list in front of me and not going to look it up) books.

Liked your point about order mattering, because he does have a pattern. Once you've read one of Never Let Me Go / Artist of the Floating World / Remains of the Day, you know what to start looking for in the other two. His other books are patterned a bit differently. Buried Giant is probably the most obvious (except for the end, which I still haven't quite worked out). A character states the titular point. But I like that the novel makes the obvious macro-point secondary to the micro-level experience of the main characters.

I don't know, but at least I know that someone else doesn't know either. I absolutely hate when people try to cram everything into a genre (I like post-apocalyptic urban fantasy! And being a predictable consumer!), and seemingly that's the only general way to answer that question.

The problem is that it's a bad question to ask. You don't read kinds of books. You read authors. So the correct question is "what authors do you like". That question I can answer. Or, you can ask what they're reading at present. Also a good question.

I don't think any interesting person likes genres in the abstract; they like specific authors. Sometimes that leads to genres, sometimes it doesn't. Of course, the follow-up tends to be "what do you like about them?" Or, "what are their books about?" Which can be damn hard questions. Good novels aren't easily summarized, and pithy descriptions of them tend to sound stupid. And most people probably don't know the authors you like, so it makes for a confusing conversation. Talking about books with random people is hard.

Trying to work backwards into a general set of rules about what makes a book a book you like is very difficult, and makes for a pretty personal conversation. Probably an interesting one with the right person.

The specific claim you're reacting to is ... puzzling. I'm not sure that there's a fiction book in existence that doesn't have the author's voice and opinion in it. I suppose ghost-written books in the style of another author? But I suspect that's not what is meant.

Here's a stab at what this criticism actually means: the author is making it very clear what they think and narrowing the potential meaning of their work in the reader's mind. Very much a stylistic choice. Some writers prefer to leave their work open-ended and don't worry about interpretation. Some writers actively encourage readers to interpret in unusual ways. Others have very definite designs embedded in their work; some interpretations are objectively wrong.

A lot of people don't like the latter kind of author. I'm not sure why. I suppose because some readers get frustrated if they have ideas about a book, then find that their ideas were wrong and that they misunderstood the book. There's also a belief floating around that books never have incorrect interpretations, which makes no sense to me. But people believe it.

As for a fiction book with so much of the author's voice as to be obnoxious at times? Ayn Rand, obviously. I actually like her work a lot, but she sometimes forgets that her characters are characters and starts writing monologues in her own voice, instead of the characters' voices. At times she does this with characters who are ostensibly having a conversation, but the two sides of the conversation bleed together and you can't tell who's speaking anymore because their voices lose all differentiation.

Eh, if I'm reading between the lines correctly, this experience is an artifact of their business model. It's a ludicrous hiring process in part because they need to claim to hire the top-3% for marketing purposes, which means they need to reject a ton of people, which means they need to come up with some way to screen out a lot of applicants quickly. They found one.

It's also a consultancy, so they really, really care about speed. They don't care about coding quality, getting rid of pandas, etc. They care about producing something they can bill for as quickly as possible, so you can move on and produce something else they can bill for as quickly as possible.

Re: top-3%. Joel Spolsky's article on this kind of bullshit metric is great. Short version: a ton of companies can credibly claim to hire the top-3% of applicants, because people apply to a lot of jobs, and because the worst people apply for a whole lot of jobs. The same people are in the denominator for all of the companies hiring the top-N%.

Are they hiring the top-3% of people? Of course not. There's no objective metric for that, and we know where the top-3% of data scientists work (roughly speaking) and it's not company X. Company X is just rejecting a large number of people relative to the number they hire. Universities do this too. They'll deliberately encourage huge numbers of people to apply that they know will be rejected solely to push down their acceptance rate and make them appear to be more competitive, which they hope will convince people that they're high-quality. Complete red-herring though. The quality of the people hired isn't related to N hired / N applied. It's about the applicant pool and the selection process. But it's a nice tricky metric for a consulting firm to throw around.