I grew up in a small town in India, without much access to good colleges or competitive environments. For engineering, that meant a tier-2 college — in India, the IITs, IIIT Hyderabad, and NITs are tier 1; state university colleges are the tier below. Mine came with rules that sound made up now: no laptops until third year, no internet in the hostel after 5pm, classes until 4pm.
In my second year I met C programming and loved it. The library had computers, and it was open from 4 to 5. So that became the routine: one hour a day, every day, on a shared library machine, solving problems on SPOJ, CodeChef, and HackerRank. It was a hobby — the way crosswords are a hobby.
Then I visited IIIT Hyderabad's R&D showcase. Students were presenting research projects — real systems, real results — and when I asked how any of it worked, the answer kept coming back the same: programming. Something reframed in my head that day. Programming stopped being puzzle-solving and became a tool that builds things. I asked the students how they'd learned, and they talked about their programming culture, research at the undergraduate level, and a competition I'd never heard of: ACM ICPC, where the top teams from each university compete.
Registering my best friends without telling them
ICPC requires a team of three to even register. Nobody else in my college was doing competitive programming. So I wrote down my two best friends' names and registered us — without telling them.
I competed, cleared the minimum bar, and didn't expect anything. Then an email arrived: selected for the ICPC Asia regionals at Amritapuri. I had no team, and the regionals fell exactly on my college's semester exams. I wrote to the organisers. They let me compete completely alone — one person against three-person teams — with the caveat that a solo competitor couldn't advance to the World Finals no matter what.
I finished in the top 50 teams across South Asia. Around the same time I was a finalist in the National Programming League at NIT Warangal, out of 7,500 participants.
Twenty-five teams in the same jersey
At the regionals I counted 25–30 teams from IIIT Hyderabad alone, all competing for the top spots. That was the environment I didn't have, sitting right there in one column of the scoreboard.
The next year, IIIT Hyderabad announced a lateral entry program: students from other colleges could join directly into the second year. They took about ten students from all of India — and if they didn't find a candidate they liked, they left the seat empty rather than fill it with the next-best person. The process was an SOP, a written exam, and an interview with six professors, each leading a research lab in their subfield of computer science. I walked out of that interview convinced I didn't stand a chance. It's still the toughest interview I've ever sat.
I got in. It was one of the happiest days of my life — not because of the brand, but because for the first time I had the environment: wifi, a computer, and research I could touch as an undergraduate.
What I did with it
I tried everything. Competitive programming. Android apps — Fantastic Chat went on the Play Store, and I built a Bluetooth messenger because this was pre-Jio India, internet was expensive, and I wanted my friends to text for free across a classroom. A compiler good enough that the professor checked the code and waived our final exam. Teaching assistant work, volunteering at Ashakiran — a school for underprivileged children — university best player in badminton, dancing on stage.
I stayed on for a Masters by Research in distributed cryptography, which became published work at ICDCN 2019. That meant two years of walking into the lab every morning to push on a problem that gave nothing back for months — including three months spent proving a problem couldn't be solved at all. It's still the hardest thing I've done, and it recalibrated me permanently: industry problems come with a guarantee that a solution exists. Research doesn't.
Why I tell this story
Not for the scoreboard results. I tell it because everything I've built since — a clothes-rental startup run from my bedroom, secure playback systems at Qualcomm, serverless infrastructure at Microsoft, and now Folia — has been some version of that library hour: limited time, limited access, no one to hand me a team, and a problem I refused to leave alone.