Somewhat tangential, but HathiTrust was born from what I would consider the "golden age" of technical work coming out of libraries (2002-2010). One of the unintended consequences of the dotcom crash was that compensation falling meant that there were a lot of talented software people working on what interested them rather than what simply paid the most (since the gap was much smaller).
As a result research libraries were well staffed with very technical people all genuinely interested in making software that made the world a better place. MIT's DSpace, LibraryThing, Open ILSs like Evergreen/Koha, and a huge range of quirky/innovative smaller projects that no longer exist all came out of this period.
It ended around 2010 since the GFC fallout started to hit library budgets while tech suddenly started getting really hot. Even if you loved libraries, most library devs where facing pay cuts to stay in libraries versus massive raises and other quality of life improvements for going into tech. Plus startups and tech companies in general at the time felt more inspired.
And now that government funding sources like IMLS, CLIR, NEH, NARA and LoC have been nuked and/or crippled, things are unlikely to get better any time soon, especially for collaborative research projects that have no immediate commercial benefit.
I worked at a university library for a few short years in the 2010s. Reading your comment helped me make sense of some of the experiences I had there. I still try to keep on top of some of the trends, with the vague hope of working in that field again one day.
I'm curious what some of the "quirky/innovative smaller projects that no longer exist" are, if you're inclined to go into some details. Or if you could point to a good resource on this somewhere. A lot of technology projects in the library space seem to reinvent the wheel over and over, so I think such a list is very valuable.