The real development is when you use received knowledge to write anything to solve your goals. Of course, you need to know basics, and at least know that "to achieve this, i need to use some fuction. let's find what function can {do _something_} and how it works". After you studied theory, using on practice, is the most efficient way to learn. Of course, sometimes good to have somebody to review code to get rid of bad habits, to avoid overengineering something simple, or to give a valuable advice when you stuck and don't know how to solve current issue.
I'm working as DBA, extensively using JS for my work (for the platform I work it's like VBA for Excel). Coding is not a main part of my work, but important.
I coding for my needs 1-3 times in a week, 2-10 hours in a day (sometimes once you dive in code, time feeling is different from usual life, so "oh, it's evening already? it seems like I woke up a hour ago"). Of course it doesn't mean sitting 10 hours, you can break, do some physical exercixces, have dinner etc, except maybe having conversations about something important, because part of your head is still occupied by current task.

Over time, it depends on how interesting this activity is. If, even when you can write something that works, you still have to force yourself to learn, then the chances of success are not so high. When there is pleasure from a finished and working thing, then success is inevitable.

I am generally a lazy person and I am motivated by the fact that if I have to regularly do some routine work for an hour, then I can spend 3 hours to write a script that does it in 10 seconds. Over time, when my level of knowledge is higher, I will write the same script in half an hour, and its running time will probably be measured in milliseconds. And I can automate something and set it on a schedule and it will work for me.