Zurich — Engineer — Builder
Engineering mind. Banking career.
Now building with AI.
Fifteen years across UBS, Credit Suisse, and Deutsche Bank. Two engineering degrees and a Booth MBA. Left to build what I actually believe in.
About
I grew up in India taking apart radios and putting them back together wrong. That instinct — understand it by building it — never left. I studied electronics engineering, then embedded systems, because I wanted to know how things worked all the way down to the silicon.
Then I went where the complexity was. Internal audit at UBS in Zurich, then healthcare M&A at Deutsche Bank in New York, then back to Zurich for governance and analytics at Credit Suisse. Fifteen years across three banks, six countries, and every flavour of institutional complexity. Made Director. Learned that the hardest problems aren't technical; they're organisational.
Chicago Booth was the pivot. Not because of the credential, but because it forced me to think about what to build, not just how to build it. Strategy, behavioural economics, entrepreneurship — the toolkit for deciding which problems are worth solving.
In late 2024, I walked away from banking. Not from frustration — from conviction. The tools had finally caught up to the ambition. AI, specifically large language models, had made it possible for a single engineer with domain expertise to build products that previously required a team of fifty. So that's what I'm doing.
I'm Irish, Indian-born, Zurich-based. I've lived on four continents and worked across six countries. I think in systems, I build in code, and I ship before I'm comfortable.
Background
Perspective
The majority of what passes for AI product development is remarkably thin — an API call dressed up with a landing page. The harder, more valuable work is understanding where AI genuinely changes the economics of a problem, not where it adds a chatbot to an existing workflow.
I spent fifteen years inside institutions that routinely spent eight figures on systems their own people refused to use. The failure mode was always the same: optimising for procurement committees rather than end users. I see AI companies making precisely the same mistake, only faster. The advantage now belongs to small teams with domain knowledge and the discipline to ship.
My stack is Claude, Supabase, and whatever moves the product forward. No research lab, no advisory board, no pitch deck. Deployed software, real users, honest feedback. That, in my experience, is where conviction actually comes from.
Off the Clock
Contact
Building something that matters, or simply want to exchange ideas — I'm always interested.