Web development is one of those fields that looks incredibly inviting from the outside. Everyone wants a website. Every business needs one. Every other person wants to learn how to build them for a living. It is creative, it pays well, and from the outside it looks like you just sit at a computer and make cool things all day.

I had exactly that mindset.

I went in with big goals and even bigger confidence. I thought it was going to be fun, straightforward, maybe a little challenging but nothing I could not handle. Turns out I was spectacularly wrong.

CSS broke me. Not metaphorically — I genuinely spent days trying to center a div and align a button correctly. I would tweak one property and three other things would fall apart. I started questioning my life choices somewhere around hour six of trying to make a hover effect work.

But I stuck with it. I did Colt Steele's Web Development Bootcamp — a long, thorough, absolutely massive course that took forever to finish. And I mean forever. Web development has a lot to learn. Anyone who tells you otherwise is lying or selling something.

My first real project was Orythos — a mathematical graph and sequence tool. I was convinced it was going to blow up on the internet. People were going to use it, share it, talk about it. I was going to be famous in the maths education community or something.

It did not blow up. Looking back at it now with slightly more mature eyes — it was honestly a pretty rough site. But it was mine and I was proud of it, and that counts for something.

Around this time I also had a realisation — I do not enjoy frontend development. At all. I am not creative in that visual design sense. I would spend an hour on a button and still hate how it looked. So I pivoted to backend. Node.js, Express, EJS. And something clicked. Building logic, handling data, making systems work — that felt right.

That pivot led to my first real paid project. I built a full admin panel and POS system for a brand called Morpankh Dezine. It had product management, barcode scanning, GST billing, WhatsApp invoice delivery — the works. They used it at an actual exhibition for live checkout. Watching someone scan a product and get an invoice on WhatsApp in real time was one of the best feelings I have had sitting at a computer.

I should also be honest about something. I used a ridiculous amount of ChatGPT while building that project. To understand concepts, debug things, figure out why nothing was working at 11pm. I felt genuinely guilty about it even though I had absolutely no reason to. Using tools to learn faster is not cheating. It is just learning.

But I did not want to be dependent on it forever. So I started making structured notes — Google Docs, one document per topic. Backend auth. Routing. Middleware. Sessions. Everything. Written in my own words, in a way I could actually understand later. Now when I need something I check my notes first. It saves time and it actually sticks better.

Right now I am building the full brand website for Morpankh Dezine. Learning new concepts, adding them to the notes, then using them in the project. It is an insane amount of work. But the notes have made it manageable. And if any developer tells you they remember every single piece of code off the top of their head — that is just a fake macho attitude. Even Zuckerberg did not grow up without copy pasting Stack Overflow. The tool has changed — Stack Overflow became ChatGPT became Claude became whatever comes next — but the idea is exactly the same. You look it up, you understand it, you use it. That is just how this works.

Speaking of AI — I genuinely cannot tell you how this field looks five years from now. Right now I use Claude to understand concepts, talk through logic, and figure out why something is broken. It is a tool, same as the notes, same as Stack Overflow before it. But the pace at which AI is growing is something else entirely. Who knows — maybe in a year or two someone just describes a website and the whole thing is built in thirty seconds. It is honestly not that far fetched.

But we are not there yet. The creative thinking, the feature logic, the decisions that require actually understanding a person's problem — that still needs a human brain. At least for now.

And that is the thing — building a website is not just writing code and making functions that technically work. It is just as much about the user experience, the design, the small elegant touches that make someone actually enjoy using what you built. Any AI or developer can put together something that functions. But what makes something worth remembering are the details — the interactions that feel just right, the UI tweaks that make someone think "oh that is clever", the care that went into every little decision. Those things do not come from a prompt. They come from actually giving a damn about what you are making.

And honestly, beyond all of it — the money, the career, the uncertainty about AI — web development is just something that is very much me. Every time I build something that works, something that actually solves a problem for someone, there is this quiet satisfaction that is hard to describe. I mean I also need money, I am not going to pretend otherwise — I am definitely not a billionaire. But at the end of it all, I think the best thing a person can do is find the thing that makes them feel most like themselves. For me, that thing just happens to involve a lot of CSS and a questionable number of open browser tabs.