Find problems to solve with code, and write code to solve those problems. You’re building muscle strength in the ability to rapidly pattern match to potential reference code paths.

https://projecteuler.net/

I'm not a good programmer by any means, but here are some ideas that have helped me along my journey:

* pick a language. It doesn't matter which one. A popular one will have more tools already built for it and a bigger community;

* learn the basic grammar of the language - it's usually trivial, unless it's your first language;

* understand what kind of algorithms and data structures exits and what problems they solve. You don't need to implement them yet, just know they exist;

* find something interesting to work on, if you can. Are you interested in something specifically? Do you want to know how to make a game or how to automate some everyday tasks, for example?

* find something your friends are contributing to or working on, and ask them questions;

* search online whenever a though pops up in your head. Not only for references from a manual, but for best practices and even open-ended questions. If you're not sure if something is correct, search for articles or comments that say it's wrong or that it's dangerous or that it's the best. Don't rely on 1 source of information. If you're using AI, ask it to give you links so you can read what actual people wrote. Don't feel inferior for making 50 search queries in an hour and writing 10 lines of code;

* take time to set up your environment. Look for ways to make it better, to automate a lot of what you do. It may be boring, but it will save you a lot of time down the road;

* try different technologies, languages and tools. Some suck objectively, some will suck for you personally. Don't stick with the first thing;

* look for similar threads elsewhere so you get more suggestions;

* learn how to freely acquire books and papers that are behind a paywall;

* take your time and enjoy the process.