Our author asks: "but how can you actually prove that your design is error-free?"
The answer is it is difficult but possible. It requires the use of technologies people are generally not familiarized with, such as proof assistants -- prominent examples are Coq, Isabelle/HOL, and ACL2.
There exist, at this point, actual formally verified software artifacts of quite substantial complexity -- the seL4 microkernel, the CompCert C compiler, the Quark browser, and a number of others.
Sadly, very few people outside of academia are familiar with the existence of formal verification tools for software, and the documentation for such systems is (to say the least) non-transparent. The best tool at the moment, Coq, practically requires that its users learn quite a bit about Martin-Löf type theory in order to work effectively in it.
That said, such tools are clearly the way to deal with such things going forward.
The Other Tesla David Blaza5 comments I find myself going to Kickstarter and Indiegogo on a regular basis these days because they have become real innovation marketplaces. As far as I'm concerned, this is where a lot of cool ...