I started in graphic design.
I ended up caring about
why things work.

I started as a graphic designer print, branding, visual communication. I was good at making things look right. But at some point I started noticing that making something look right wasn't the same as making it work right. That gap is where I found UX.

When I joined Lybrate in 2015, it was still early days for digital health in India. I was brought in as a UI designer, but the nature of the product, a platform where people make decisions about their health pushed me into thinking harder about why users were doing what they were doing, not just how the interface should look while they did it.

I stayed at Lybrate for seven years. Part of that was the work building an app used by over 80 lakh people, designing the doctor community platform GoodMD from scratch, launching a video consultation feature when telehealth was still a new concept in India. Part of it was honest comfort, I was a senior designer with real autonomy and a product I genuinely believed in. Eventually Lybrate was acquired by Pristyn Care, which became its own chapter a faster-paced environment where I learned what it means to design for conversion at scale.

I care about how things look. I won't pretend otherwise. Visual craft matters a design that functions perfectly but feels cheap undermines the trust you're trying to build with users. But the visual is always in service of something.

The habit that defines how I work is a simple question I ask when something feels off, why is this working this way, and what would happen if it worked differently? Sometimes the answer is "it's fine, leave it." Sometimes it opens up a better solution. That instinct to question the assumption before accepting it has driven most of the design decisions I'm most proud of.

The hardest version of this is designing without context. When there's no user research, no PM brief, no data just a screen to fill. That's genuinely difficult. I've been in that situation more than I'd like. It taught me that even a small amount of context one support call listened to, one user complaint read carefully is exponentially more useful than none.