I spent my entire first year refusing to pick a domain. Not on purpose. I just couldn't commit to one no matter how hard I tried. Turns out that was the best thing I could've done, I just didn't know it at the time.
The Overwhelm
I came in knowing Python basics, datatypes, the usual fundamentals. I knew AI/ML ran on Python but I was never actually pulled toward it. Then I got to college, found out how many wings existed, and it just felt like too much. Especially CP. I gave it a shot anyway. Started doing CodeChef contests, slowly got into competitive programming. Didn't do anything extraordinary, my brain just doesn't flip into problem solving mode that fast. What actually got me excited was exploring new dev stuff and technologies. But CP was mandatory, so I stuck with it after a few months of just accepting that.
After that I started doing assignments from every wing, sitting in on classes I had no business being in, honestly. I had zero clue what to pick. Web dev felt fine, we built a bookstore site with React and Spring Boot for a group project, but it didn't excite me, partly because it was already baked into our academics anyway. App dev was worse: first two classes I thought I was going to ace this, then a few classes later I lost the spark completely. That was basically my whole year in two modes. Either I can do everything, or I can't do anything. Meanwhile a bunch of my batchmates had already locked in their domain and were deep into it. I never felt that pull to commit. First year was just exploring, for better or worse.
Web3 Hooked Me First
A few Web3 classes in, I thought this was actually cool. Dove into the basics, did every assignment the wing threw at me. My first Solidity contract was a voting one, writing it out and watching it actually deploy felt like I'd cracked open something most people never bother looking at. Then came a betting contract, and for a while I was convinced this was going to be my domain, full stop. I was still sitting in on ML classes on the side, more out of habit than intent. ML hit me hard early on, the math didn't click, the code didn't click, and I remember genuinely thinking I'd never be able to do this. Did a couple of assignments anyway, badly, then during a break gave CP another shot since contests were running again and I felt guilty for ignoring it.
FOSS Changed Everything
Around March I leaned further into Web3, studying for Hack o Fiesta, our college hackathon. Before that even happened, DevSprint hit, the craziest tech experience of my first year up to that point. Every domain, tight deadlines, a leaderboard refreshing in the background the entire time. It was stressful in a way that somehow made me want more of it. Then came a stretch of tech fests, one after another. Contract Battle, a Web3 event, I finished top 5 there and spent the rest of that period building websites for other dev events on the side. ML didn't get a single hour of my attention during this stretch, there just wasn't room.
FOSS Weekend is the one that actually changed things. I landed top 8, but the ranking barely mattered. What mattered was finally understanding how open source actually works. Before that I had no real idea what an issue even was, how it got assigned to someone, what a PR was supposed to look like. I'd applied to Summer of Bitcoin months earlier and gotten nowhere because I genuinely didn't know how to contribute to anything. FOSS Weekend is where that stopped being true. FOSSology followed soon after, a two round thing, questions first to shortlist people, then teams of two actually fixing issues and reviewing each other's work. My team won Best Reviewer Award, which felt less like a prize and more like proof I'd actually understood the workflow, not just survived it. Between that and FOSS Weekend, open source stopped being "a wing I dabble in" and became the thing I actually wanted to get better at.
Hackofiesta: 36 Hours of Chaos
This was the most intense hackathon I've done. My team picked Web3 as the main domain, the idea was a decentralized food delivery app. Building it in 36 hours turned brutal fast, because none of us actually knew how to integrate an app with Web3. So we pivoted to a React web app instead. Spent 7 to 8 hours on that before realizing that was a bad call too, barely anyone knows the stack, and deploying it would be a nightmare. Plan changed again. By this point we were running on no sleep and genuinely losing motivation, half convinced the project wasn't happening.
Took a quick nap. 10 hours left. Started over. Built a phone UI inside a website and got back to it. A bunch of seniors helped debug and build alongside us, genuinely thankful for that. Then we got MetaMask integrated with the frontend and it actually worked, that moment felt incredible. Then deployment, 2 to 3 hours left, errors everywhere. Fixed them one by one and got it shipped. The seniors who'd helped us were just as relieved, because we'd taken a lot of "this isn't going to work" energy from people around us the whole way through, and we still pulled it off.
Landing on ML
Summer break shifted everything. My interest moved to ML and this time I actually sat with it instead of running from it. A lot of my batchmates had picked ML way earlier and were miles ahead of me, but that didn't hit the way it would've a year earlier, I was used to being behind by then. I built CervixAI that break, a tool for cervical cytology image classification, my first real ML project, and the first thing I built in the domain where I didn't feel completely lost the whole way through. That's when I knew I wanted to keep going. I kept pushing into ML and DL and it hooked me in a way nothing else that year had. I'm still learning a lot of it as I go, but it's the first domain that's actually made me want to stick around and go deeper instead of jumping to the next thing.
If you're a first year reading this panicking about not having "your domain" yet, you don't need one. Explore until something actually holds your attention past the first few classes. Mine took a full year to show up.
Top comments (0)