The Engineering of One

A few days ago I spoke at the National Scientists’ Round Table Conference at MIT‑WPU in Pune, under a theme that paired two of the world’s great honours: the Nobel Prize and the Queen Elizabeth Prize for Engineering. My argument was a simple one. The Nobel honours discovery — the new truth about the world. The Queen Elizabeth Prize honours delivery — the engineering that carries an idea out to millions of lives. India has long reached for the first; for a developed India by 2047, I said, we must learn to reach just as hard for the second.

The talk was kindly received. But that night, something kept turning over in my mind — a thought I had not made room for on stage, and which I have not been able to put down since.

Between those two prizes there is a third kind of work. It changes a life — profoundly, completely — and yet it will never reach millions. It cannot. It was never meant to. It is a solution built for one person, one body, one very particular need. And in more than two decades of working alongside persons with severe disability, I have come to believe it is among the most important engineering we ever do — and among the least recognised.

A question about number

Notice what the two prizes have in common. The Nobel asks, in effect, how fundamental? The Queen Elizabeth Prize asks how many? Both are questions about magnitude — of insight, or of reach. And almost everything in our world of research and innovation is built around that second question. Grant committees ask how many will benefit. Journals ask how large the effect. Investors ask how big the market. Even our fashionable word impact is, when you look closely, mostly a word about scale.

Let me tell you about the work that fails every one of those tests, and matters anyway.

A population of one

In my talk I spoke of Mr Suresh, a management graduate whose life was interrupted by a brain‑stem stroke in 1999. He lost the use of his limbs and his voice; for years he could communicate only by blinking his eyes. We built him a brain–computer interface, and thirteen years after his stroke he operated a computer again — moved a cursor, switched on a light, spoke through a synthesiser.

How many people did that system reach? One. Suresh. It was tuned to his signals, his residual abilities, his particular way of being in the world. It could not be boxed and sold. The very next person I met needed something else entirely — a switch under a cheek, a puff of breath, the twitch of one working finger. This is the quiet truth of the disability sector: the population you are designing for is very often a population of one.

Every human being is different, and disability multiplies that difference. A device that gives one child a voice may be useless to the child in the next room. A control system perfect for one young man with cerebral palsy must be rebuilt from scratch for the next. And when a person’s condition changes — as it does — the solution has to change with them. There is no economy of scale here. There is only the patient, unglamorous, deeply human work of building, again and again, for one.

You can see why this work sits in a blind spot. It wins no prize for discovery, because it discovers nothing new — it uses what we already know. It wins no prize for scale, because it reaches no millions. It is hard to publish, hard to fund, hard to fit into any metric. A young engineer who spends a year restoring one person’s independence has, by every conventional measure, produced almost nothing. By the measure that actually matters, they have done something close to a miracle.

Did you restore one person’s world?

I have watched a mother hear her child’s chosen words for the first time. I have watched a man who could only blink send a message to his family with his own thoughts. If, in that moment, you had asked either of them how many people did this reach?, the question would have been meaningless.

So perhaps, alongside the Nobel’s what did you discover? and the Queen Elizabeth Prize’s did it reach millions?, we need to hold a third question — quietly, but with equal honour: did you restore one person’s world? It is not a lesser question. It is only a different one. And our institutions have almost no way to hear it.

Scaling the unscalable

Here is what gives me hope.

For most of my career, the reason “solutions for one” could never be sustained was cost. Every bespoke device demanded months of an engineer’s time, and there are simply not enough engineers on earth to hand‑build a solution for every person who needs one. But that arithmetic is changing. Low‑cost digital fabrication — the 3D printers and maker‑labs we have been spreading across the country — lets a custom part be made in an afternoon. And generative AI is beginning to let us design, adapt and re‑tune a personal solution in hours rather than months.

For the first time, we may be able to do something that once sounded like a contradiction: to scale the unscalable. Not one product for a million people, but a million products, each made for one. Mass customisation, turned toward human dignity. If the twentieth century learned to manufacture the same thing a million times over, perhaps the twenty‑first can learn to manufacture a million different things — each one exactly right for the single person who needs it.

India, of all places, should lead this. We have the need — tens of millions of our fellow citizens live with disability — and we now have the tools. What we lack is only the will to honour the work: to fund it, to teach it in our engineering colleges, and to celebrate the engineer who changes one life as warmly as the one who changes a million.

The motto of the conference where all this began was in Sanskrit — विश्वकल्याणाय विज्ञानम्, “science for the welfare of the world.” I have been turning that phrase over ever since. The welfare of the world is not only a matter of millions. The world, for each of us, comes down in the end to a single life — our own, and the ones we are able to reach. To restore even one of them is not a small thing. It may be the truest science of all.