How I Got Into AI (And Why I'm Not Looking Back)
If you're reading this, welcome to my digital corner of the internet. I'm Roshan, a Computer Science graduate who didn't stumble into AI by accident. I got into it because I couldn't resist the challenge. And the chaos.
I wasn't the kind of kid who grew up obsessed with robots or reading academic journals. I was the kind who broke things just to figure out how they worked. Whether it was a toy, a random website, or some Python script I found online, I always wanted to mess with it and see what would happen.
That curiosity turned into something more. Over time, I moved from writing simple scripts and debugging them out of frustration to discovering tools like large language models. And suddenly, I wasn't just fixing code anymore, I was watching models write it for me. That moment changed everything.
I wasn't just interested in building software. I wanted to build systems that could learn, adapt, and sometimes even surprise me.
Why AI?
Not because it's trendy. Not because it looks good on a resume. It's because AI feels like the closest thing we have to machines that can think. Not like humans, but in a way that's good enough to solve problems we haven't fully figured out ourselves.
I've always been drawn to complex, fuzzy problems. The kind where there's no clear right answer. AI is full of that. You tune a model, test a prompt, and sometimes it just clicks. Other times, it doesn't. You go back, change something, try again. That loop of improvement is where I find the most energy.
I enjoy the weirdness, the edge cases, the unexpected outputs. That's what makes this space alive. And that's what keeps me building.
What I'm Working On
Right now, I'm focused on making AI actually useful. Not just fun demos. Real tools.
I'm building LLM-powered systems that solve real-world problems like multi-agent workflows, document Q&A tools, and automation that saves people time. You'll find those projects here on this site.
This blog is where I'll share the process behind them. The cool stuff, the bugs, the ideas that failed, the things that finally worked. It won't be perfect. But it will be real.