I've worked on two different startups that built products until our runway ended—over a year of work in both cases—and then pulled the plug before getting our first customer.

Maybe that doesn't count?

I've also worked on a multi-year product for a large company in Redmond that I was really proud of when building it. It doesn't matter what the product was supposed to do, just imagine it sounding kind of sexy and being very complex and brainy. It was for scientists. When I told someone who wasn't dumb but was outside the software industry about what I'd built, he looked confused and said "well, why wouldn't you just use [a free piece of software that is well understood and runs on every device] instead?" and I had no good answer, because he was right.