Its a tough problem. You have to find something that you want to exist; like an app or a website or a game. For example, try making a GUI for managing SSH keys. You know, like the ones github makes you create in order to clone and push to a repo. Make a visual representation of those keys (stored in the .ssh folder), and tools to add/delete them.

Along the way you’ll find tons of missing things, tools that should exist but don’t. Those are the “real” projects that will really expand your capabilities as a developer.

For example, I was coding in python and wanted to make a function that caches the output because the code was inherently slow.

• but to cache an output we need to know the inputs are the same
• hashes are good for this but lists can’t be hashed with the built-in python hash function
• we can make our own hash, but hashing a list that contains itself is hard
• there is a solution for lists, but then hashing a set that contains itself is a serious problem (MUCH harder than hashing a list)
• turns out hashing a set is the same problem as the graph-coloring problem (graph isomorphism)
• suddenly I have a really deep understanding of recursive data structures all because I wanted to a function that caches its output.

This consternation is definitely common. It’s hard to apply skills to something with no long term impact of benefit. I’ve improved my skills by finding stuff I can help on in the communities I participate in.

It’s natural to be overwhelmed, so deciding on a project does scope what you can learn, but a hard part is architecting the foundation of that project.

Introducing new features to an existing project is a great way to get your feet wet - it has multiple benefits, for one of you do take a position as a developer in the future, you likely won’t be architecting anything initially, primarily improving on existing projects. So participating in OSS projects is a similar mechanism to that - you have to learn their codebase to a degree, you have to learn their style and requirements, etc.

Even if you don’t ultimately contribute, it’s still a learning experience.

To take up a career in programming is to be professionally confused on a day-to-day basis. It’s normal to feel overwhelmed at times, but this is the advice I give younger engineers: accept that confusion, and try to channel it into curiosity instead of letting it fester into stress and anxiety.

Try to replicate software/apps you use everyday. Not to improve them, but to figure out how they work. In addition to learning how they work, you’ll learn the problems the original devs had to solve, and one way to solve them.

Leetcode is a great way to polish your skills. When I was your age, I read programming books and made projects I cared about, it’s turned out very well.

I’ve helped a few others learn programming, practice and working on any project at all always help more than anything.

It’s a lot of work. I’m in a similar boat. I’ve been self teaching myself how to program in various languages and working as a Fullstack developer at a very small startup.

While I’ve learned alot, there’s always more to learn and finding direction is a challenge you’ll hit many times over if you stick with it.

Right now I’m reviewing various Open Source Projects I could contribute to as I know ultimately this will make me a much better developer. There are so many CLI and web development tools I use that I’d like to expand on but will require a marginal jump in my skills. Here are a few examples:

• Add a feature to the rTorrent client that would allow it to torrent over i2p
• Expand the amount of tools available within the SolidJS framework ecosystem to be competitive with React’s ecosystem.
• Create a HTTP server in Rust that has feature parity with Golang’s Caddy
• Add a feature to AntennaPod akin to YouTube Sponsor block that would allow for user contributed timestamps to facilitate auto skipping of ad mentions in podcast episodes.

I also am slowly working out plans to learn enough about Rust and browsers to recreate the terminal based browser Links in Rust.

But these are all just me spitballing. In truth, my skills aren’t quite there yet and the amount of time to get there for each one of these is a lot. So I have to choose wisely and dive in deep on the project’s codebase, probably for at least a year or more to make headway. But this is the mindset I think might help you. Look into the software you use a lot or like the concept of. Figure out how you’d like to improve it or add to it. Reach out to the developers of said software however you can, either make your case for the improvements you want to make or just help out on something they want to do with the software. Learn, listen. Rinse and repeat.

Hope this helps a bit.

If you want to improve significantly, go read someone else’s code and modify it. Try to fix a bug in a program you use, add a feature you want that doesn’t exist already, or even just do something simple for the sake of proving to yourself that you can do it – like compiling it from source and figuring out how to change some small snippet of text in a message box. Even if you don’t succeed, if you put in a serious effort attempting it, you will almost certainly learn a lot from trying.

Edit: changed wording to try to be clearer

Read/Inspect and contribute to FOSS. They’ll be bigger and longer lived than small, personal, and experimental projects.

Study computer science.

Work, preferably in an environment with mentors, and long-/continuously-maintained projects.

Look at alternative approaches and ecosystems. Like .NET (very good docs and guidance), a functional programming language, Rust, or Web.

That being said, you ask about “should”, but I think if it’s useful for personal utilities that’s good enough as well. Depends on your interest, goals, wants, and where you want to go in the future.

For me, managing my clan servers and website, reading online, and contributing to FOSS were my biggest contributors to learning and expertise.

I like to pick a fun project, pick a language I don’t know / wanna learn better, and then just go for it. Don’t be discouraged if somebody’s already made it - nothing says your learning project has to be useful in the real world, tho it’s kinda nice if you think of something that can be. If your project seems intimidatingly hard, remember the programmer workflow of breaking it down into manageable pieces and tackling those. If it doesn’t seem hard enough to teach you anything, I sometimes like to write it without using any external code or libraries (or a minimum of them; if it’s something like a GUI program I’ll use direct vulkan bindings instead of like Qt). This is also one of the few areas I get some use out of LLMs, cuz bullying ChatGPT or a local equivalent into giving me huge and tailored lists of program ideas can be really helpful. Either way my main advice is just to pick something that interests you and have fun with it; things don’t have to be worthwhile to other people to be worthwhile to you.

Multi-Account Containers extension is looking for more contributors

https://lemm.ee/post/43431167

https://github.com/mozilla/multi-account-containers/discussions/2672

Look at MIT and UC Berkeley’s CS curricula and start tackling things that you haven’t covered. They’re both available freely online and you might still be able to find video recordings of Cal’s lectures somewhere (they recorded every class for students who weren’t present or had difficulty understanding in real time until 2015 or so but were hit with an ADA accessibility lawsuit because they weren’t captioned or something.)

I found reading through the rust book was a nice walkthrough of problems one can hit and how that language elected to solve them.

In terms of practice:

• Write a vim config
  • Shell out to python if you’d like
• Learn a bit of elisp and org-mode
• Rewrite all your shell scripts into a python CLI
• Write a pyqt6 GUI for tasks and notes on the exact way you’ve always wanted it to work
• Write an AI tool to auto-format links etc with phi3
  • Very exciting how much these smaller models can do!

Join in on your school clubs and research projects, or start some with friends!

There are many great competitions where previous programming experience would come in handy.

One competition that takes place in the U.S.:

NASA Student Launch

It actually IS rocket science! Student Launch is a 9-month long challenge that tasks student teams from across the U.S. to design, build, test, and launch a high-powered rocket carrying a scientific or engineering payload. It is a hands-on, research-based, engineering activity and culminates each year with a final launch in Huntsville, Alabama home of NASA’s Marshall Space Flight Center. The activity offers multiple challenges reaching a broad audience colleges and universities as well as middle and high school aged students across the nation.^[1]

Culminating Event Dates: April 30 – May 4, 2025

Culminating event location: Huntsville, AL

Eligibility: Open to U.S. Students

Grade Levels: Grades 6-12, College and University

You need to think of things you might want to build. Try building software that nobody ever has, try fixing issues in current software using git. You can find lots of python program in GitHub, waiting for contributions. That’s how I learnt. If you want to learn and use C++, try building software that interacts with the real world with Arduino, try contributing to the Linux kernel or try making simple games using C++ in Godot.

Does your school have a robotics team or something along the lines of computing? That would be a good option. Also if you are still in high school and plan on going to college, you still have plenty of time to learn.