01
Systems thinking.
The model of how something works, to me, is more interesting than what it looks like. Get the model right and the screens mostly design themselves.
More about me
I'm - a designer, engineer, & traveller.
Born and raised in Delhi, my intro to design was messing around in Photoshop on my dad's computer when I was 12. Eventually, I found myself drawn to interaction design during my CS degree.
For the last 3 years, I've been working with startups at every stage, seed through Series D, in small, focused teams that ship fast. Lately, I've been shipping some myself.
Today, most of my work sits at the intersection of AI and enterprise software.
Currently, I'm rabbit-holing on agent memory patterns, bringing all sorts of ideas to life in Claude Code, and planning my next holiday.
How I work
01
The model of how something works, to me, is more interesting than what it looks like. Get the model right and the screens mostly design themselves.
02
I work across Figma and the repo on the same day. Designs in Figma are a guess about how a thing will feel. I like to build prototypes (& sometime the whole ship) in code to test.
03
Empty states, error handling, the small moment between clicking and seeing something happen. This is where people actually live.
04
A small product that does one thing extremely well beats a big product that does a lot of things almost.
05
Between a good AI demo and a product people use every day sits a huge amount of design. That distance is the most interesting problem right now, and it's where I want to be.
How I work
01
The model of how something works, to me, is more interesting than what it looks like. Get the model right and the screens mostly design themselves.
02
I work across Figma and the repo on the same day. Designs in Figma are a guess about how a thing will feel. I like to build prototypes (& sometime the whole ship) in code to test.
03
Empty states, error handling, the small moment between clicking and seeing something happen. This is where people actually live.
04
A small product that does one thing extremely well beats a big product that does a lot of things almost.
05
Between a good AI demo and a product people use every day sits a huge amount of design. That distance is the most interesting problem right now, and it's where I want to be.
Beyond design