Escaping the tutorial trap and starting my ML training arc

You ever start watching an anime or a series on episode 12 and just sit there wondering wtf is going on? 😂

Yeah, that’s exactly what it would feel like if I just dropped my Kaggle code today without an intro. So, let’s start at Episode 1.

Hey, I'm Vishnu. I’m currently a 1st-year B.Tech CSE student, but I'm playing the long game. My ultimate endgame is pursuing my Master’s in AI and Machine Learning at a top research lab in Japan.

Recently, I had this massive realization: the clean, custom datasets they give you in basic ML tutorials are basically a trap. They teach you the syntax, but they don't teach you how to survive. The real world is a messy warzone of missing values, typos, and useless noise.

I realized I need to treat Data Science less like a calculator and more like detective work.

I’ll be using Dev.to as my captain’s log. I'll be posting my Kaggle wins (actually just hit a 78% accuracy on my first real Medical Insurance dataset today!), the bugs that fry my brain, and probably a lot of anime analogies to explain complex ML math (obviously I use this to teach myself lol)

If you are also on the tech grind or just want to watch a disciplined newbie, try to decode the AI world from scratch and make it to Japan, stick around.

The real work starts now. Let's get it. 💻🔥⚡