Most technical mistakes that a non-technical founder can make will involve one of two things: Scalability, and maintainability
Lacking knowledge in these two things will not hinder an early startup website much, so it's easy to go ahead and build something without expertise in these domains.
The only thing that sucks about these situations is that once your company takes off, the engineers you hire will be left to maintain/fix the mess that the non-technical founder made - which can be quite miserable for the engineers.
Or rebuild it in the background with a second 'team'. I know that people warn about rebuilding your product in the early stage. But in such a case, why not start from scratch one more time? Seems reasonable, if there is a user demand already and the first version didn't take very long.
Lacking knowledge in these two things will not hinder an early startup website much, so it's easy to go ahead and build something without expertise in these domains.
The only thing that sucks about these situations is that once your company takes off, the engineers you hire will be left to maintain/fix the mess that the non-technical founder made - which can be quite miserable for the engineers.