I tell people to build a small crud application that you can use to manage some data you have some interest in. I've build a similar non-marketable applications, usually a todo/project management application or something that imports a bunch of stock market data probably 30 times now, I actually use the project itself to learn new languages and frameworks because it hits most of the areas I will have to learn to be productive and I can pickup the language semantics along the way. When you're done you have code that is going to be doing things you will actually encounter in the wild instead of contrived and isolated examples from text books.
My favorite part is once I'm done I feel comfortable enough with reading through repos and some books over how to be more efficient/idiomatic and then going back through my little project and refactoring it.
With all that said I have a graveyard of projects in languages and technology that is probably not used anywhere commercially and the project itself is languishing if it was completed at all - but for example, I have a half working todo app in racket and despite being a junk app in a non-commercial language, I had to learn lisp and it at least subtly has influenced how I approach problems. Same thing with Rust which I actually use quite a bit personally and commercially, but when I originally learned it I was 100% golang professionally and I noticed my golang functions would resemble rust functions if I squinted and I think the complexity in my golang went way down.
I guess my advice is Just Start Building Projects (that you may never even show anyone or post on github, much less go marketing it.) and you will benefit from it massively. If you are able to learn and benefit from it _and_ it becomes popular or a commercial success than great! But who cares, on the internet no one knows you aren't a dog.
My favorite part is once I'm done I feel comfortable enough with reading through repos and some books over how to be more efficient/idiomatic and then going back through my little project and refactoring it.
With all that said I have a graveyard of projects in languages and technology that is probably not used anywhere commercially and the project itself is languishing if it was completed at all - but for example, I have a half working todo app in racket and despite being a junk app in a non-commercial language, I had to learn lisp and it at least subtly has influenced how I approach problems. Same thing with Rust which I actually use quite a bit personally and commercially, but when I originally learned it I was 100% golang professionally and I noticed my golang functions would resemble rust functions if I squinted and I think the complexity in my golang went way down.
I guess my advice is Just Start Building Projects (that you may never even show anyone or post on github, much less go marketing it.) and you will benefit from it massively. If you are able to learn and benefit from it _and_ it becomes popular or a commercial success than great! But who cares, on the internet no one knows you aren't a dog.