One of the best authorities on that concept is Djikstra - they actually formally proved a few full algorithms and also provided a lot of guidance on how to prove others. It's not something you'll see... well, probably ever... but it's a very interesting approach to program design. Dijkstra imagined a world where all programmers would be writing formal proofs of correctness before authoring a single line of code... that's a neat world, I'd love to see it... but it's also incredibly burdensome so our tech explosion would likely be happening a lot slower.

To clarify, are they actually doing research or just CS degree havers? Most of us folks with CS degrees don't work in a research capacity (even if we're working with interesting business problems).