This is the coolest demonstration of TLA+ I've seen to date. It takes you through specification, implementation, model-checking with TLC, visualization and animation, checking an inductive invariant, deductive proofs mechanically checked by TLAPS, refinement mapping and even model-based trace checking.
OT: I don't know if this level of rigor is employed by any major libraries. I know AWS has used it a great deal for their systems though notably recently[1]