I never failed an exam. I ran out of money.

My mother, Poonam Devi, is a teacher, and the first job I ever had was hers. From class three I was checking her students' copies and totalling their marks in the evenings. It was not framed as work. It was simply what happened in our house.

By the time I was ten I had started my own — teaching the children of tenants in our building on our rooftop for twenty-five rupees a student. By class nine it had become deliberate: two home tuitions at three hundred and fifty rupees each, seven hundred a month, and five hundred of it going straight to my own tutor. I was fourteen, running a small business whose only product was my own education.

I say this for one reason. Everything my mother did, she did while struggling for it, and for me. What looks on paper like early ambition was really her refusal to let circumstances decide what her children could become. I learned the whole of my trade, and most of my stubbornness, at her side.

My Class 10 results came in June. In July I applied to Patna Science College. I missed the cut-off by three marks. Three.

In August I went to a coaching institute in Patna run by a chemistry teacher named Khakhan Mahto — one of the large, well-regarded ones, where everyone preparing for engineering wanted to be. I paid five hundred rupees for the trial classes. That was the money I had. When the next instalment came due I could not pay it, and I could not find anywhere else I could afford either.

So I did not prepare for engineering. Not that year, and as it turned out, not ever. I want to be exact, because it is the foundation of everything since. It was not that I couldn't do the work — I had been financing my own education for two years. The work was simply not available at a price I could reach.

What I decided that August was narrow, and I have never moved off it: I would earn enough that money would never again decide what I was allowed to learn. And having been on the wrong side of it once, I have never been able to stop noticing everyone else it happens to.

I already knew Zero Error Education — I had spent the first months of 2006 in Delhi, sitting in its classroom as a Class 10 student preparing for my board exams. So when, in 2007, my maternal uncle, Mr. Lokesh Kumar Singh — a senior police officer — offered me a way to finish class twelve in Delhi with a part-time job attached, I was not going somewhere unfamiliar. I was going back.

I lived with him for two or three years, and I would not be who I am without them. My education until then had been academic and financial: how to learn things, and how to pay for learning them. What he taught me was neither. How to conduct yourself. What to do when something is wrong in front of you and you are in a position to act. What a person owes to the place they live in, rather than what they can take from it. Almost everything I now do that has no commercial purpose comes directly out of those years.

In 2008 I walked back into Zero Error Education — this time as a teacher. Sixteen years old, doing the only thing I had ever been good at, in the same classrooms where I had sat as a student two years before.

I take on things that have no commercial return, and I have lost things for it. Standing up to something wrong has cost me time, money and relationships more than once, and I expect it will again.

I don't present this as a virtue. It is closer to a temperament, and I learned it from my uncle rather than working it out myself. If something is wrong in front of me and I am in a position to act, the fact that acting is expensive has never seemed like an argument. I intend to keep doing it, and I know what it costs. The rule I follow is plain: when I come across a real problem — something genuinely wrong, not merely inconvenient — I step in and try to solve it, even when the cost of doing so falls on me.

I'll be plain about the thing that sits under everything above, because it's the honest engine of it. At fourteen I was told, in effect, that money decided how far I could go in science. I have never accepted it.

What I am working towards — deliberately, every day — is to study at MIT, and then to work as a scientist at the Large Hadron Collider at CERN: the furthest edge of the science I was once priced out of. My ground for it is chemistry and mathematics — quantum, nuclear and polymer chemistry, and a facility with numbers I've had since I was a boy totalling my mother's mark registers. Whether or not that particular door opens, it is the direction every day's work now points.

Nearer at hand: the doctorate finished and published; the diagnostic part of teaching — the part that has never scaled — built properly into software; the tutoring network growing outward from Delhi; and eighteen years of things I've worked out and never written down, written down at last.