> The problems that I see most junior devs struggling with are at a much higher level. How do you organize large scale programs? How do you make sure that different parts interact flawlessly? A book about design patterns could help here, but without experience you won't appreciate what the patterns are good for.

I couldn't disagree more. Algorithms are the base level that you build on. Anything non-trivial you will have to understand what is available to you and create something new. Take something common, doing a substring match on a list of items. If you don't have algorithms knowledge I'd bet your solution is going to adversely affect the final program.

Good choices of data structures, and solid understanding of algorithms are the units of composition in creating software. IMO

> But by far the most difficult part to wrap your head around is concurrency. How do you make sure your assumptions all hold up when multiple threads / processes / users are all doing stuff in parallel? That's really hard to understand, and you need a combination of theory and experience to master this.

Or use Rust and have the compiler check it for you, so you don't have to be a wizard.