Just do both. Your early code will probably suck. Your early startup ideas will probably be quite flawed. But working on real project will make coding more fun,

Just do both. Your early code will probably suck. Your early startup ideas will probably be quite flawed. But working on real project will make coding more fun, and all your "failed" startups should teach you something for the next one. I've been coding for a long time, and sometimes my code still sucks but it's better than it used to be. I haven't been bootstrapping "startups" for quite as long but I've already made a ton of mistakes, and learned from them. PS. Wouldn't call myself an HN "Elite" by any stretch, but my $0.02 nonetheless.