If you ever want to see what it's like for a mathematician to not hand wave anything away, look at excerpts from Bertrand & Russel in principia mathematics (no, not the newton book).
It takes 362 pages (depending on edition) to get to 1+1=2
https://archive.org/details/principia-mathematica_202307/pag...
Of course, just like real mathematicians, in our everyday work we stand on the shoulders of giants, reuse prior foundational work (I've yet to personally write a bootloader, os, or language+compiler, and include 3p libraries), and then hope that any bugs in our proofs are caught during peer review. Like in math, sometimes peer review for our code ends up being a rubber stamp, or our code/proofs aren't that elegant, or they work for the domain we're using them in but there's latent bugs/logic errors which may cause inconsistencies or require a restriction of domain to properly work (ex code only works with ASCII, or your theorm only works for compact sets).
And of course, the similarities aren't a coincidence
https://en.m.wikipedia.org/wiki/Curry%E2%80%93Howard_corresp...