What fifteen years of building actually taught me.

Between 2016 and 2018, three of us built CeleryHQ — a cloud point-of-sale for Indian restaurants, years before the market was ready. One visionary in the US writing the Android app, one fresh graduate selling door to door, and me building everything else. Over a hundred restaurants paid for it. Growth needed a street sales force; three founders who kept their day jobs couldn't fund one.

The product outlived our courage: the last paying customer left in 2026, eight years after we stopped pushing. We didn't build the wrong thing. We under-bet on ourselves — the one mistake I've refused to repeat with my time or anyone else's money.

The visionary from note 01 had first hired me as a freelancer, years before CeleryHQ. In 2021 he called a third time — now founding Simplified, an AI content platform backed by some of Silicon Valley's best-known funds. He didn't run a search for someone to build his platform and his India team. The same founder trusted me twice. Five years later, Simplified spans text, image, video, and audio generation for fifteen million users, on architecture I own end to end.

The fresh graduate who sold CeleryHQ door to door? He's my co-founder at Vellestudio today. One "failed" startup produced two working companies and my two most important professional relationships. The failure wasn't a detour on the way to the story. It was the founding chapter of it.

My first engineering job handed a junior an entire platform for yoga studios — schema, backend, frontend, deployment. I shipped it alone, learning as I went. Every big thing since — the agency, the 15M-user platform, engagements for clients across New Zealand, South Africa, Singapore, Italy, and the US — runs on that same muscle: take responsibility for the whole thing, not your layer of it.

Hired as a firm's first Python engineer; left behind a department of eight and fingerprints on a marketing platform later part of a $4.75 billion acquisition. Joined a US retail giant's Pune floor that had never shipped a single-page app; introduced Angular, trained the team, shipped their first. The best work I've done started before the thing existed — when the honest answer to "what's the architecture?" was a whiteboard and a deadline.

People love asking about the weekend I moved Simplified's entire production from AWS to GCP. It's the least interesting thing I've done — and that's the point. It was boring because the architecture had been platform-agnostic from day one, three years before it needed to be. Clever is a demo. Boring is a discipline.

These days my team ships AI-first — engineers write specs and review, AI writes most of the code, and I hold the architecture straight. I'll design your platform, advise you the way a CTO would, or sit in your sprint and take tickets. Happily. Fifteen years in, most engineers retreat into slide decks. I'd rather stay dangerous at every altitude.

I graduated in computer engineering in 2009 — into the worst hiring market in a generation — and started on a support desk, listening all day to people whose things were broken. Fifteen years later I went back to a classroom on purpose: the Leadership in the Digital Era program at IIM Lucknow, 2024. Fifteen years studying machines; it seemed fair to spend one properly studying humans. Leading people is a craft, and crafts deserve deliberate practice.

Most mornings: badminton, before the standups. After fifteen years of decade-long games, there's real joy in a point that ends in seconds — and the shuttle, like production, does not care about your plans.