In my experience, Sololearn: Learn to Code is ur best bet? Honestly, others werent as good as expected, but this ones free or $69/year for Pro. Super bite-sized tho!!

Coming back to this... I totally feel u on the "textbook on a tiny screen" vibe. It's literally the worst when you're trying to learn something complex like neural networks while getting bumped around on a train. Ngl, I’ve been pretty disappointed with a lot of apps that claim to be "interactive" but basically just give you multiple-choice questions. It just doesnt stick, you know? If you want actual hands-on stuff that doesn't suck on mobile, I’d highkey recommend DataCamp: Learn Data Science. Their mobile app is actually built for this—they have a custom coding interface so you don't have to struggle with your phone’s regular keyboard layout. It’s super bite-sized and perfect for a 20-minute commute. Plus, they have a whole AI Fundamentals track that covers the technical side really well. Another one that's honestly great for concepts is Brilliant.org. It’s not "coding" in the code-editor sense, but it’s incredibly interactive. You’re basically solving visual puzzles to understand how AI logic works. Way better than reading a dry textbook tho... I also messed around with Mimo: Learn Coding/Programming for Python basics, which is okay, but it felt a bit too "Duolingo-ish" for deep AI stuff imo. TL;DR: Go for DataCamp: Learn Data Science for mobile-friendly coding, or Brilliant.org for the logic/math side. Good luck with the commute!! peace

👆 this

totally agree with adlerser about that textbook-on-a-phone feel, it kills the experience when youre on a bumpy train. ngl i spent hours looking at the technical specs and interactive features for these platforms and they vary so much between brands. tbh your best bet is just searching for mobile ai learning app tier list on youtube. i saw a really detailed video a few weeks back where they actually show the screen recordings of the coding snippets on different devices. its way better than me trying to describe the ui here. check the reddit threads too, there are tons of people who have already done the heavy lifting comparing the latency and hands on exercises for you.

I went through a phase where I spent an hour each way on a bus trying to practice machine learning without a laptop. Most standard educational apps felt way too passive for my taste. Eventually, shifting my focus toward a cloud-based development environment accessible via a mobile browser proved to be a decent option. It took a bit to configure the interface so the keyboard didn't block the terminal, but the effort paid off. My current setup involves a subscription-based virtual machine that runs notebooks on a remote server. Tweaking layers in a model and hitting run while walking to the station really changed how I internalized the concepts. One downside is that setting everything up manually is a bit of a process, but it works better than anything else I have tried. The main takeaway for me was that mobile learning works best when the environment feels like a playground rather than a quiz. Having a real console available is just much more engaging than clicking through slides.

Ive been messing around with these platforms for years and unfortunately, most of them are pretty disappointing. I had issues with the laggy interfaces and the way they handle code snippets on smaller screens, it just doesnt feel natural yet.

• coding on a phone is a recipe for carpal tunnel
• screen real estate is never enough
• most apps feel like they were designed for tablets only Honestly, my commute is so loud anyway that I can barely think. Last week some teenager was blasting techno from a bluetooth speaker at 7 AM and I ended up just arguing with them about public etiquette instead of finishing my lesson. Why do people think everyone wants to hear their playlist? Its getting ridiculous out there. Anyway lol sorry kinda went off topic there.

🙌