Fifteen years of getting things into production.

I started in client servicing and became the person other people escalated to. That turned into a technical analyst role, and then into designing and specifying the internal tooling the company ran on — reporting consoles, scheduling systems, and the scoring logic that decided which accounts got attention first.

It taught me the thing I still lead with: the gap between what a client says they need and what will actually work is usually the entire job.

Seven years running product and operations inside a portfolio of consumer apps, finishing as Deputy General Manager. I took nine products from an idea to a launch — among them a referral platform for financial products, a notification service for people who follow creators, and a browser-extension SaaS tool.

I architected the data platform underneath all of it — sixty million user records, fifty-plus behavioural attributes, and a segmentation engine that evaluated rules in real time rather than against a frozen list.

The result I am proudest of is not technical. I found monetisation levers nobody had shipped, argued for them, built them, and revenue went from ₹1.8 crore to ₹20 crore. Finding the thing that is worth building is a separate skill from building it, and most engineering organisations are short of the first one.

I left to find out how much one person could build with the current generation of tools. The answer so far is three production systems: a Connected TV exchange, a real-time ad exchange carrying two-sided billing, and a platform for creators who want to own their audience. All three are live. You can open them.

Building them alone taught me where projects actually die. Not in the prototype — prototypes are easy now, and everyone has one. They die in the month afterwards, when the demo has to become something with authentication, tenancy, a billing path that reconciles, evaluations that catch a regression, and an answer for what happens at three in the morning. That month is what I sell.

B.Tech in Information Technology, 2007 to 2011.