Yep. CEO of our company didn't like the progress on one of our frontends, so he took a weekend to write a whole new one himself, claiming it was "almost done."

Yep. CEO of our company didn't like the progress on one of our frontends, so he took a weekend to write a whole new one himself, claiming it was "almost done." It was, as you may predict, nowhere near done, but still shoved into Git and became the new frontend. In reality he seemed offended that he didn't know much about how we (the frontend team) built it and he didn't like that, so he built the way he thinks he knows, which turned out to just be a hodge podge of code from various tutorials and no singular vision for how the application functions. There are no less than three different paradigms in use in this codebase, and none of them really play well together. The end result is that it took us nearly as long as it took us to develop the first version to bring the second version up to feature parity, but now we have a giant mess of a codebase that was very strictly written to do exactly the things the CEO wrote it to do. Predictably, development slowed to a crawl pretty quickly because we're essentially forced to rewrite every single individual piece that needs touching. I told them this was a bad idea and would only make things take far, far longer than finishing the one we had. The whole reason this was done? The CEO and CTO found some bugs but didn't want to file them on the issue tracker. They decided it was better that the CEO write his own, despite the CEO having exactly zero experience with frontend development. This is just one of a long line of bizarre decisions that I have accurately predicted would only cost us more time and money than is worth it. We're currently looking for new jobs.