Curious About How Things Work
I'm a final-year BEI student at Pulchowk Campus, and honestly, I got into AI and ML because I kept wondering how machines could "learn" anything at all. That curiosity turned into late nights debugging neural networks and a genuine love for figuring out the math behind it.
What I enjoy most is taking something that exists only as equations on a page and turning it into something that actually works — whether that's catching a fake AI-generated voice or getting a small robot to find its way across a room on its own.
Still Figuring It Out, And I Like That
Most of my work lives somewhere between databases, reinforcement learning, and front-end code — three things I never thought I'd be juggling at once. I picked up web development almost by accident while building tools for my own projects, and it stuck.
I spend a lot of my free time just trying things out — a new RL algorithm, a hackathon idea, a half-finished side project. I don't always know where it'll go, but that's kind of the point.
Building Things That Actually Get Used
I care less about writing "clever" code and more about writing code that holds up — neural networks that actually generalize, databases that don't fall apart under real data, interfaces that don't confuse people.
Every project teaches me something I didn't know I needed to learn. That's usually the part that keeps me coming back, even when the bug count is higher than I'd like.
Teaching Robots to Navigate on Their Own
One of my favorite projects so far has been training a TurtleBot3 to navigate using reinforcement learning — watching an agent go from bumping into walls to actually finding its way is oddly satisfying. I worked through ROS2 pipelines and PyBullet simulations to get there.
I'm still learning a lot of this as I go — comparing model architectures, fixing things that break in simulation but not in theory, and slowly getting more comfortable with the messiness of real engineering.
