The thing that finally made me want to learn to code wasn't a course, a YouTube video, or a good explanation of variables. It was a meal planning app.
I built it four days ago. Three hours with Claude Code, start to finish. It connects to an AI, retrieves recipes based on what I want to cook that week, assembles a meal plan, and exports the whole thing — plan, grocery list, everything — directly into Notion. Nothing clinical about it, no grand vision. Just a tool that maps to the way I actually think about food and planning, built around my specific workflows in a way that nothing off the shelf has ever managed to do.
When it worked for the first time, I didn't have a word for the feeling. Magic is the closest I've come — not because it was mysterious, but because I'd had a vision in my head and then it existed in the world, and the gap between those two states felt shorter than it had any right to be. I knew I didn't fully understand what I'd built. I also knew, for the first time, that I wanted to.
That's the inversion that changed everything for me.
Every resource I'd encountered before assumed that understanding came first and building came second. Work through the fundamentals. Earn the right to make something real. I'd tried that path more than once and hit the same wall every time — not a comprehension wall, but a relevance wall. I could follow the lessons. I just couldn't find a reason to keep going. The gap between "I understand this concept" and "I can build something that actually matters" felt enormous, and nothing in the curriculum helped me cross it.
What changed with Claude Code wasn't that the concepts became easier. It was that I had something real to build, and I built it, and now I have a genuine reason to understand the mechanisms underneath. I'm at the beginning of that foundation now — JavaScript, React, Git, the underpinnings of the tools I've been using without fully grasping. I've barely started. But I'm learning backward from things I've already made, trying to understand what I was actually doing, and that's a different kind of motivation than any tutorial has ever given me.
If AI is doing the heavy lifting, are you actually learning anything? Is this just a more sophisticated crutch?
I keep coming back to my PT clinicals. The thing that made everything from school finally stick wasn't the lectures — it was treating real patients, in real situations, where something was actually at stake. Nobody asked whether hands-on learning was a crutch. The doing was the learning. The classroom concepts came alive because there was finally context for them.
I don't think what I'm doing now is different in any way that matters. A library isn't a crutch. A mentor isn't a crutch. We learn by watching people do things we want to do — that's how motivation has always worked. The fact that my partner in this is an AI agent doesn't change the underlying dynamic. I'm building real things, developing real taste about what works and what doesn't, and generating genuine curiosity about the mechanisms underneath. That's not a shortcut. That's just learning.
I'm not trying to become a developer. I'm trying to become a better partner in building things — someone who understands enough of what's happening under the hood to ask better questions, catch more mistakes, and have a clearer sense of what's possible. The foundation I'm building isn't an end in itself. It's so the next thing I build is better than the last one.
I still have a long way to go. I couldn't rebuild the sous app from scratch right now without help. But I understand it well enough to improve it, explain it, and know exactly what I want to build next. For the first time, that feels like enough.