People always ask how I got into programming. The honest answer involves a Supreme drop, a resale spreadsheet, and a lot of F5 refreshes.
When I was in my early teens, the sneaker resale market was my whole world. Supreme drops, Off-White x Nike collabs, BAPE — I lived for it. I'd set alarms at 4am, have multiple tabs open, and write down sizes and prices in a notebook like I was running a small hedge fund. In a way, I was.
The part that nobody tells you about sneaker reselling is how data-driven it actually is. You learn to read demand signals — which colorways sell, which ones sit, how hype cycles work, when to hold and when to flip. I was doing all of this manually, and I started to wonder: what if I could automate some of it?
That question led me to Python. My first script was a simple price tracker. It was ugly, it broke constantly, and I was genuinely proud of it. That feeling — of building something that actually does something — got me completely hooked.
Now I build ML systems that do exactly what I was doing manually as a teenager: reading signals, scoring data, making decisions. The tools are different, but the mindset is the same. I think that origin story is part of why I care so much about the practical side of AI — it was never academic for me. It was always about solving a real problem.