I'm not sure where I say we "have some expectation that they [candidates] should be upset and disappointed if they don't make the cut". I would assume they'd be disappointed they didn't get an offer, wasted application time, and didn't get an impactful role... for easily avoidable reasons.
Agreed that people not a match for a job shouldn't get the job offer nor should they take if the offerer have messed up. This is all match making... Which is two-sided.
I'm lucky enough to have reached a point personally & professionally where I can highly value where each hour goes & doesn't go. A lot of time goes into a job, so the idea of applying for a bunch of 2-5yr (or longer) journey candidates, and the possibility to do the best professional work of my life to date.. and not doing my homework on the options just doesn't make sense. For some people it does and for many legit reasons, and for them, a leetcode interview for a FAANG style job probably makes more sense. Just that kind of approach is a harder sell for making a good match at a startup at the more formative years.
I'm implying it somewhat, since if you're "looking for ownership, interest in our customers/mission/long-term" and getting involved in what you see as "career-defining creations", then that's a relationship with a job that requires a candidate to buy into what you're doing to the point where they'd be upset to be rejected from the opportunity. I'm not trying to say you're deluding them or anything, but that the level of commitment you're asking for out of candidates before they've even started with you is such where it's a harder rejection than just "it's on to another job."
What I'm trying to say is that expecting the candidates to do the work to sell themselves on your vision and importance and viability feels like both a large burden on them and something that's setting even those dedicated enough to do it up for likely heartbreak.
Like you say, this is two-sided. Loading up the free version of your offering is more work than most interviewers are putting into a candidate. I could see it being very fair and very interesting to have a section of an interview where I sat with one of the company's developers or product folks and played around with the product to learn and talk about it - that seems like it would be a good signal both ways.
Otherwise, it just feels like you're asking candidates to put in far more time and personal effort deciding to care about them than you're likely putting in to learn and care about them outside of the interviews. That's always the case with interviews, but this additional layer just feels like an unnecessarily interviewee-unfriendly level on top. If it's working for you, then that's good for you, but consider the sorts of candidates that could be a good fit but don't have the time after work to spend even more time becoming invested in your company and a vision.
Agreed that people not a match for a job shouldn't get the job offer nor should they take if the offerer have messed up. This is all match making... Which is two-sided.
I'm lucky enough to have reached a point personally & professionally where I can highly value where each hour goes & doesn't go. A lot of time goes into a job, so the idea of applying for a bunch of 2-5yr (or longer) journey candidates, and the possibility to do the best professional work of my life to date.. and not doing my homework on the options just doesn't make sense. For some people it does and for many legit reasons, and for them, a leetcode interview for a FAANG style job probably makes more sense. Just that kind of approach is a harder sell for making a good match at a startup at the more formative years.