Why I build things.
I'm drawn to the gap between the data a company has and the decisions it actually makes. That gap is where I've spent my career — and it's never really a data problem. It's a trust problem, a clarity problem, a systems problem.
I started at American Express, in enterprise loyalty analytics, working on decisions that moved real money — reshaping a benefit structure, sizing opportunities, shaping how millions of cardholders were treated. What I learned there stuck with me: the hard part isn't finding the number, it's getting an organization to believe it enough to act. A correct answer nobody trusts changes nothing.
Now I lead product and AI work at BrowserStack, where I've built internal AI tools from nothing to something people use every day — a self-serve analytics layer, an agentic workflow that flags deals quietly slipping, and a revenue assistant that has to explain its own answers. Different products, same instinct: put a better decision in front of a real person, and make the system trustworthy enough that they'll use it.
How I work
Close to the engineers and data scientists, not above them. The AI products I'm proud of came from being in the daily detail — arguing about what a signal means, where the model should stay quiet, what "good" even means for a feature whose output isn't a single fixed answer. I write the requirements, but I also sit in the messiness of building.
I'm suspicious of confident answers — from people and from models. The work that lasts is the work that knows when to say "I'm not sure," shows its reasoning, and can be argued with. An AI product that can't be challenged doesn't get trusted, and one that isn't trusted doesn't get used. So I care a lot about the unglamorous parts: evaluations, explainability, the definitions underneath a number.
What motivates me
Watching someone make a better call because of something I built. Not a dashboard they glance at — a moment where the decision was sharper, faster, or more honest because the system did its job. Especially when the problem is unglamorous and internal: the operational, behind-the-scenes work that quietly makes an organization run better. That's the whole game for me.
I also just like building. I learn in public, ship before it's polished, and write to think. This site is part of that — less a résumé, more a record of how I work through hard problems in AI, product, and enterprise systems.
The short version
Product person who actually builds AI systems. Analyst who cares about decisions, not just dashboards. Operator who writes. MBA from IIM Indore, six-plus years across Amex and BrowserStack. If any of that is useful to you, let's talk.