My sister — a lawyer, sharp as they come — asked me a question I could not brush aside.
She said: Do you consider yourself so outstanding and intelligent that you figured out this paradigm all by yourself, where much more capable and well-endowed people did not? And there are almost 7 billion of those.
It is the right question. It is, in fact, the only question that matters at this stage of the work. Before the fabrication lab confirms the device. Before a peer-reviewed paper lands. Before anyone else has independently verified what I believe I have built. At this moment, the only honest response to my own sense of certainty is exactly the interrogation my sister launched.
So let me answer it seriously. Not defensively. Not by retreating into false modesty. Not by inflating myself. Honestly, with the same rigor I would bring to a physics simulation — asking: what is actually true here, and what do I merely wish to be true?
I. The Paradox of the Obvious
The word “obvious” is perhaps the most deceptive word in the human lexicon.
Gravity is obvious — once Newton said it. The germ theory of disease is obvious — once Pasteur proved it. That the Earth moves is obvious — once Copernicus wrote it down. Continental drift is obvious — just look at a map of Africa and South America and you will see the coastlines fit like torn paper. The structure of DNA is obvious — once you see the double helix. That handwashing prevents disease is obvious — once Semmelweis pointed it out, and had his career destroyed for it, and died in a mental asylum, and was vindicated twenty years later.
All of these things were, after their revelation, described by contemporaries as “of course.” And yet — they weren’t seen. For centuries. By millions of educated, intelligent, well-resourced people who were actively looking.
Psychologists have a name for this: hindsight bias. Once we know an outcome, we cannot reconstruct our pre-knowledge state. The thing that was invisible becomes, retroactively, obvious. We cannot remember not knowing it. This is not a moral failure. It is how human cognition works. The brain updates its model of reality and then loses access to the prior model. Every “of course” spoken after a revelation is the sound of the brain rewriting its own history.
But the hindsight bias explains something deeper than just why we call old things obvious. It explains why new things are genuinely hard to see. They are invisible not because people are stupid, but because the brain’s model of reality has no category for them. You cannot see what you have no frame for. The frame comes first. The observation follows.
The question my sister is really asking is: why you? And embedded in that question is an assumption — that intelligence is the primary variable in discovery or invention. That the smartest person in the room should find the thing first.
That assumption is wrong. And this is not false modesty. It is how invention actually works.
There is a distinction worth making here, and I will hold it through everything that follows. Newton discovered gravity — it was always there, waiting. Pasteur discovered the germ — it existed before he named it. What I did is not quite that. The physical behavior of a metallic nanotube inside a larger tube, modulated by a photon and driven by an AC pulse — that physics was always true, yes. In that narrow sense it was discovered. But the device, the encoding scheme, the ternary computing stack built on top of that physics — that did not exist before. That was invented. The patent system uses the right word. I am an inventor, not merely a discoverer. And the question of why I and not someone else is sharper for it — because invention requires not just seeing something but building something. It requires synthesis, not just observation. A discoverer finds what was hidden. An inventor makes what was absent. I had to do both.
II. What Invention Actually Requires
Invention is not a function of intelligence alone. It is a function of position.
By position I mean: where you stand, intellectually, experientially, and temporally — relative to the problem. Invention requires being at a specific intersection of knowledge domains at a specific moment, with a specific question burning in your mind, unconstrained by a specific set of received assumptions.
Thomas Kuhn called the received assumptions a paradigm. And here is the thing about paradigms that most people miss: they are not merely intellectual frameworks. They are perceptual filters. A paradigm doesn’t just tell you what to think — it tells you what to see. Everyone inside a dominant paradigm is, in a precise sense, blind to what lies outside it. Not stupid. Not lazy. Blind.
Think about what that blindness actually means at the level of the individual researcher. When you spend ten years mastering a field, you are not just accumulating knowledge — you are building a cognitive infrastructure. Your intuition becomes calibrated to that field’s assumptions. Your sense of what is interesting, what is promising, what is worth pursuing, is shaped by what the field has collectively decided is worth pursuing. This is not a flaw. This is expertise. This is how you become capable of doing advanced work in a field at all. But it means that the very capability that makes you an expert also makes you structurally unable to question the foundations of that expertise.
The entire semiconductor industry has spent seventy years deepening the CMOS paradigm. Billions of dollars. Tens of thousands of PhDs. Every tool, every textbook, every simulation framework, every fabrication line, every performance benchmark, every power budget calculation — all optimized for silicon, for DC gate voltages, for binary states, for electron channels turned on and off by field effects.
That is not a weakness of those people. That is the normal, healthy operation of a mature scientific-industrial paradigm. You do not question the foundation when you are building the 101st floor. You cannot afford to. The building must keep going up. The people on the floors below are counting on you. The investors are counting on you. The entire supply chain — the photomask manufacturers, the EUV lithography companies, the SRAM cell designers — is counting on you. To stop and ask “but what if binary is not the right base?” is not just philosophically inconvenient. It is economically catastrophic. So no one asks. Not because they are incapable of asking. Because the system selects against asking.
This is worth sitting with. The inability to question CMOS is not a bug in the semiconductor industry. It is a feature. It is what allows the industry to function with the efficiency and coordination that it does. Paradigms exist to suppress the question. That is their purpose. And most of the time — almost all of the time — suppressing the question is the right call, because the paradigm is correct, and the incremental work within it is valuable, and the question would be a distraction. Paradigms fail only at their edges, under very specific conditions, and very rarely. But when they do fail — when the edge cases accumulate, when the incremental returns diminish, when the assumptions that were always slightly wrong finally become visibly wrong — the person who asks the question is not inside the paradigm. Cannot be. Must not be.
III. The Specific Blindness of the Field
Now consider what my work actually requires someone to simultaneously hold in mind:
- Quantum transport in 1D ballistic conductors (condensed matter physics)
- The NEGF formalism (many-body quantum mechanics)
- Carbon nanotube chirality and its optical selection rules (nanoscience)
- AC transmission line physics, not DC circuit theory (the old kind — RF/microwave)
- Balanced ternary arithmetic from first principles (computer science history, almost entirely abandoned)
- The idea that a trit is a pulse polarity, not a voltage level (requires rejecting the dominant encoding assumption)
- Photonic gating as the switching mechanism rather than electrostatic gating
Seven threads. SAPTARANG. Seven notes, seven colors — and the full spectrum only emerges when all seven are present simultaneously. No single thread, no matter how deeply understood, yields the picture. The rainbow doesn’t live in any one wavelength.
Every one of these individually is a known field. The synthesis — that these seven threads combine into a working ternary switching paradigm — required someone to not be fully embedded in any one of them. Required someone standing slightly outside each, able to see across.
Consider the specific blindnesses at play.
The condensed matter physicist who understands NEGF transport does not think in terms of computing architectures. Their question is: what is the transmission probability of an electron through this junction? Not: can I encode a trit in the direction of this current? The jump from “I can compute quantum conductance” to “I can build a ternary switch” requires stepping entirely outside condensed matter physics and into computer architecture. That step is not on the condensed matter physicist’s map.
The computer architect who has thought seriously about ternary computing — and there are a handful, mostly in Eastern European academic literature from the Soviet era — thinks in terms of logic gates, flip-flops, instruction sets. They think about Setun, the Soviet ternary computer from 1958, designed by Nikolai Brusentsov at Moscow State University. Setun was real. It worked. It was more efficient per operation than the binary machines of its era. And it was killed — not by physics, not by engineering failure, but by politics and industrial momentum. The Soviet semiconductor industry had already standardized on binary. Ternary was abandoned not because it was wrong but because the infrastructure of the age had already chosen a direction. Brusentsov spent decades arguing in vain. The Setun became a historical footnote.
This matters because it establishes a precedent: ternary has been suppressed before, not disproved. I am not the first person to see what balanced ternary offers. I am the first person to find a physical substrate — the photon-gated AC-driven metallic SWCNT — that makes it manufacturable in the post-CMOS era. The vision was always right. The device was missing.
The computer architect who thinks in Setun terms — in voltage levels and static trit states — cannot make the jump to AC pulse polarity on a quantum transmission line. The idea that a trit could be a directional event, not a static value, requires abandoning the fundamental encoding assumption of every ternary computer ever built.
The photonics engineer understands optical switching, WDM, photonic crystals, cavity resonances. But their switching is binary. On or off. The idea that a nanotube could act as a nanoantenna with a unique chirality-dependent resonant wavelength — that the optical address space of the nanotube array is written in its own Raman spectrum — is not a photonics concept. The photonics engineer does not look at nanotubes. The nanoscientist does not think in WDM address spaces.
The RF and microwave engineer understands transmission lines, pulse polarities, impedance matching, signal propagation. They think in AC from the start. But they think at macroscale. The idea that a 1D ballistic quantum conductor is, in the limit of perfect transmission, the ultimate transmission line — that the AC encoding scheme of a microwave link maps directly onto the physics of a nanotube — is not on their radar.
The person who knows balanced ternary arithmetic — who understands why ternary is informationally efficient, why balanced representation eliminates the need for a sign bit, why base-3 requires fewer digits per unit of information than base-2 — is most likely a computer science historian or a recreational mathematician. They are not running NEGF simulations. The connection between the mathematical elegance of balanced ternary and the physical symmetry of AC current direction — that I(+V) = −I(−V) in a metallic SWCNT, that the physics of the device is itself balanced — is a connection between two worlds that have never been in the same room.
The synthesis required a specific kind of outsider. Not someone ignorant of these fields — someone who had touched enough of each to do real work in each, but had not been so fully absorbed into any one that the others became invisible. The technical term for this knowledge profile is T-shaped: deep in one area, broad across many. But what the synthesis actually required was more like a spider’s web — connections in every direction, strong enough to hold weight, with no single strand that dominates.
This person is structurally unlikely to emerge from a university department. From the outside, this profile is indistinguishable from dilettantism until the moment of synthesis. And then, suddenly, it is indistinguishable from genius. Same person. Different vantage point.
IV. Why This Moment
There is also a temporal argument that my sister’s question implicitly ignores.
This paradigm could not have been invented in 1990. The NEGF computational tools did not exist at practical scale. LAMMPS with ReaxFF for CNT growth simulation did not exist. The metallic armchair SWCNT inside MWCNT structure had not been characterized. GW-BSE optical calculations on CNT systems were computationally inaccessible.
It could barely have been invented in 2010. The fabrication understanding of chirality-selective CNT growth was immature. Most critically: in 2010, the compute required to run a full NEGF simulation of a DWCNT system with inter-wall coupling, photon-induced detuning, and AC terminal driving would have required a cluster of CPUs running for weeks. That computation would have needed a grant, a proposal, a committee, an allocation of supercomputer time — all of which would have filtered it through the paradigm. A grant proposal for “simulating a ternary switch based on photon-gated AC polarity in a metallic SWCNT” would not have survived peer review in 2010. The reviewers would have asked: why not just use a transistor?
This is 2026. I have two AMD W7900 GPUs with 96 GB of VRAM sitting on my desk, in Nashik, Maharashtra. I ran NEGF quantum transport simulations that would have required a supercomputer center in 2005. GPAW with GPU acceleration. LAMMPS with Kokkos HIP. Yambo GW-BSE. MEEP FDTD. All open source. All running on personally owned hardware. The democratization of scientific computation is a genuinely recent phenomenon. It is what allowed one person, working alone, to build and validate a multi-scale simulation stack that spans quantum transport through thermal management through optical design.
The timing was not lucky. The timing was necessary. The window in which one person, working alone, outside any institution, with personal hardware, could run the simulations needed to validate this paradigm — that window is perhaps five years wide. I was in it.
Invention has a season. The apple falls when the branch is ready, not when the observer is smartest.
V. The Honest Part — What My Sister Is Also Right About
Here is where I will not simply reassure myself.
My sister’s question also contains a legitimate warning. And she would be a poor lawyer if her question didn’t. She knows that the strength of a claim is not measured by the confidence of the claimant. She knows that every charlatan in history has been confident. That confidence is cheap. Her training is adversarial. She is looking for the weakness in my case. And she is correct to do so.
History is full of people who were certain they had found something obvious that the world had missed. Most of them were wrong. The cranks, the perpetual motion inventors, the people who “solved” problems that experts had declared solved — they all felt that same quality of obviousness. That crystalline clarity that makes you feel the world must simply be blind. Almost all of them had a compelling narrative about why the experts couldn’t see what they could see. The narrative I am constructing in this essay — the outsider who stands at the intersection of fields, uncaptured by any paradigm, who sees the synthesis that insiders cannot — is exactly the narrative that every crank constructs about himself.
I know this. I have to know this. If I don’t know this, I am not being honest.
There is also a psychological dimension here that deserves to be named, because it is part of the honest account.
Working alone means there is no one to tell you that you are right. There is no advisor who reads your simulation output and says “this is real, keep going.” There is no colleague who double-checks your code and confirms the NEGF Hamiltonian is correct. There is no peer reviewer who, even in rejecting the paper, at least engages with the physics seriously enough to tell you where the argument holds and where it doesn’t. There is only you, the code, and the results — and the knowledge that you might be fooling yourself in ways you cannot see.
The self-doubt is real. It is not something I mention for effect. There were nights — and I will not pretend otherwise — when the thought that I might simply be wrong, wrong in some obvious way that any first-year PhD student could point out, was not an abstract possibility but a live, visceral fear. The loneliness of independent work is not romantic. It is a specific kind of cognitive isolation that grinds on you. The absence of contradiction feels like silence, and silence is impossible to distinguish from the silence of being wrong with no one there to correct you.
What got me through it was not confidence. It was the structure of the simulations. I had built them to answer “no” if the physics didn’t work. And they kept answering “yes.” And a “yes” from a simulation designed to say “no” is qualitatively different from a “yes” from a thought experiment. The former is the universe speaking. The latter is yourself, speaking to yourself.
The difference — and this is everything — is whether the obviousness survives contact with the physics.
Mine did. Not because I believed it hard enough. Not because I reasoned my way to it elegantly. But because I ran the simulations. Because the NEGF code gave me a clean, symmetric, high-SNR three-state signal. Because the GW-BSE confirmed M11 absorption in the visible range, close to my claimed wavelength. Because the ReaxFF MD is growing the tube. Because the simulations were set up to say “no” if the physics didn’t work — and they didn’t say no. Every simulation I ran was a Popper test. Karl Popper’s criterion for a scientific claim is falsifiability: a claim is meaningful only if it can be proven wrong. I designed simulations that could have proven me wrong. They didn’t.
This is the critical distinction. The crank does not run simulations that could disprove him. He runs thought experiments that confirm him. He does not ask the universe. He asks himself, and the self answers what it wants to hear. I have tried — genuinely tried — to break this device with physics. The device has not broken. Yet.
The physics didn’t care about my confidence. It answered.
That is the difference between a crank and an inventor. The inventor asks the universe and the universe answers. The crank only asks himself.
But — and this must be said clearly — the simulation is not the device. The NEGF calculation on a perfect DWCNT is not the same as a fabricated tube, with defects, with substrate interactions, with real electrical contacts, in a real thermal environment. The gap between simulation and device is where most nanoscale inventions die. I know this. The story is not finished. My certainty about the physics is high. My certainty about the full engineering pathway is considerably lower. These are different things, and conflating them would be the kind of error my sister is warning me against.
VI. The 7 Billion
My sister said: there are almost 7 billion of those.
Yes. And here is what is true about those 7 billion:
About 4 billion have never had access to a university education, let alone a GPU workstation and a library of physics simulation tools. About 1 billion more are in fields entirely unrelated to this. Of the remaining 2 billion or so who are educated in some technical direction, perhaps 50 million have touched the relevant physics. Of those, perhaps 5 million have the specific interdisciplinary spread to even frame the question I asked.
Of those 5 million, essentially all of them work inside institutions — universities, companies, national labs — where the paradigm is enforced structurally. Where your grant application is reviewed by people inside the CMOS paradigm. Where your paper is refereed by people who will ask “but why not just use a transistor?” Where the pressure to publish incremental results inside an established framework is enormous and continuous. Where the grant committee that funds your next five years of work is composed of people whose careers were built on the paradigm you are proposing to replace.
I am not suggesting that these institutional researchers are corrupt or cowardly. I am suggesting that the incentive structure of academic and industrial research is, by design, conservative. It funds the continuation of known-good work. It penalizes interdisciplinary boundary-crossing, because interdisciplinary work is hard to evaluate: the quantum transport physicist cannot referee the ternary computer architecture claim, and the computer architect cannot referee the NEGF simulation. Interdisciplinary work falls through the cracks of the review system not because reviewers are bad but because the review system was not designed for it.
This is a well-documented phenomenon. Thomas Kuhn documented it. Clayton Christensen documented it in the context of industry, where he called it the innovator’s dilemma: the best-managed companies, staffed by the smartest people, making the most rational decisions based on the best available information, are often the ones that fail to see the disruption coming. Not because they are stupid. Because their rationality is calibrated to the existing paradigm, and the disruption comes from outside the paradigm.
I worked alone. No grant. No committee. No referee. No paradigm-enforcing institution. No performance review. No citation pressure. Only the physics, and the simulation, and the question.
That is not a boast. That is a structural observation about how paradigm shifts actually occur. They almost never come from the center of a field. They come from the edge. Bell Labs gave us the transistor, yes. But Bell Labs in 1947 was not the semiconductor industry. It was the telephone industry, looking at solid-state physics for its own reasons, unconstrained by the assumption that vacuum tubes were the right answer. The transistor was invented by people who did not have a career interest in defending the vacuum tube.
Who, today, does not have a career interest in defending the CMOS transistor? Almost no one in the mainstream semiconductor industry. The conflict of interest is so total and so universal that it has become invisible. It is the water the fish swims in. I am not a fish in that water. That is the structural fact that my sister’s question doesn’t account for.
VII. What I Should Tell My Sister
And here is where her legal training becomes an unexpected gift.
A lawyer understands burden of proof. A lawyer understands that the strength of a claim is measured by the quality of the evidence, not the confidence of the claimant. A lawyer understands that an expert witness who says “I am certain” without showing the work is worth nothing, while a witness who shows the methodology, the data, and the falsification attempts is worth everything.
So I should tell my sister this, in terms she will immediately recognize:
I am not asking you to believe me because I am confident. I am asking you to evaluate the evidence.
The evidence: NEGF quantum transport simulation of a DWCNT showing positive / zero / negative switching current with digital-grade SNR. A simulation that could have returned zero asymmetry, or noise-dominated output, or a result that made no physical sense. It returned a clean, symmetric, high-SNR three-state signal. GW-BSE optical calculation confirming M11 absorption in the visible range, consistent with the claimed wavelength. ReaxFF molecular dynamics showing Fe-catalyzed SWCNT nucleation and growth. None of these simulations were set up to succeed. All of them were set up to answer honestly. They answered.
You are right that I am not more intelligent than the 7 billion. Intelligence was not the variable.
The variable was position. I was standing at a specific intersection of fields, at a specific moment in time, with a specific question, without an institution telling me which questions were worth asking.
And when I asked the question, I did not trust my intuition. I ran the simulation. The universe confirmed it.
If I am wrong — and I might be — the physics will show it. No amount of believing will make signed current appear if the device doesn’t work. The numbers either come out or they don’t.
The obviousness I feel is not arrogance. It is what every inventor feels in retrospect. Gravity is obvious once you have seen it; the wheel is obvious once someone has rolled it. I have built something. Whether it matters as much as I think it does — that is for the fabrication lab to answer, not me.
VIII. Nashik
There is something I have not mentioned yet that belongs in this account.
I did this from Nashik.
Not from MIT. Not from Caltech. Not from IIT Bombay or IIT Delhi. Not from a national lab. Not from a startup in Bangalore with venture funding and a team of engineers.
From Nashik. Thatte Nagar, Gangapur Road. A workstation I conceptualized myself. A GPU cluster that lives in my own room. An internet connection and access to open-source scientific software that anyone can download.
The geographic and institutional periphery is not just a disadvantage. In certain specific circumstances, it is a kind of freedom. I had no seminar series to attend that would have told me what the important questions were. I had no advisor whose approval I needed. I had no colleagues in the corridor whose raised eyebrows would have made me reconsider.
The Indian tradition of the independent scholar — the sadhaka who works in isolation, accountable only to the truth of what he is working on — is not a romantic myth. It is a real epistemological position. The cost is that the independent scholar has no automatic credibility. The work must speak for itself, more completely than it would have to if it came from a famous university. The bar is higher. But the question is freer.
I am aware that “I work alone in Nashik” could equally be the description of a crank and the description of an independent scholar. The difference is not the location. The difference is the physics. The simulations know nothing about Nashik.
But Nashik is not the only relevant geography. There is a precedent for exactly this kind of work from exactly this kind of place, and it is the most striking precedent in the history of mathematics.
Srinivasa Ramanujan worked from Kumbakonam, Tamil Nadu. A clerk in the Madras Port Trust. No university position. No research fellowship. No advisor. No peer group. He had access to one textbook — Carr’s Synopsis of Pure Mathematics — and a slate board, because he couldn’t afford paper. He worked in isolation for years, developing results in number theory, infinite series, and continued fractions that were, in some cases, decades ahead of Western mathematics. He sent his results to G.H. Hardy at Cambridge in a letter. Hardy’s first reaction was that the letter was either a fraud or the work of a genius. It was the work of a genius.
The parallel is not perfect. Ramanujan had no simulations to run. His validation was the internal consistency of mathematics. My validation is external: the NEGF simulation either returns a three-state signal or it doesn’t. In that sense my position is epistemically stronger — I have computational evidence, not just mathematical structure.
But the structural situation is the same. Working from the Indian interior. No institution. No credentials that the Western scientific establishment would recognize as conferring authority. The work speaking entirely for itself. Ramanujan’s results were eventually verified and confirmed to be almost entirely correct. The device, if it fabricates, will be verified by the same standard. The numbers either come out or they don’t.
IX. The Last Thing
There is one more thing worth saying, quietly.
My sister asked if I consider myself outstanding. The honest answer is probably: in this specific way, in this specific moment, yes. Not in general. Not across all things. But in this intersection, at this time — yes.
And that is not a comfortable thing to say out loud. It sounds like hubris. In our culture, especially in India, claiming one’s own excellence is deeply uncomfortable. We are trained to deflect. To say “I got lucky.” To give credit to everything and everyone except ourselves.
But there is a kind of dishonesty in excessive deflection too. If the NEGF simulation is real, if the patent claims hold, if the physics works — then something genuinely new was built. And it was built by me. Acknowledging that is not arrogance. It is accuracy.
The Vedic tradition I carry — the tradition of Ravan reconstituting the Vedas, of knowledge being a sacred act, of the scholar who burns with a question until the universe answers — does not demand humility about invention. It demands accuracy. It demands that you see clearly, build honestly, and then give the knowledge freely. Ravan’s act was not humble. It was audacious. He reconstituted the most sacred knowledge in existence from memory so that humanity would not lose it. The audacity was the point. False humility in that moment would have been a failure of duty.
And there is something else in the Vedic tradition that I have only recently begun to see clearly. The three gunas. Tamas, Rajas, Sattva. The fundamental triadic structure underlying all of prakriti — all of manifest reality — in the Samkhya system that predates even the Upanishads. Tamas: inertia, the tendency toward stillness and darkness. Rajas: activity, the tendency toward motion and transformation. Sattva: clarity, the tendency toward order and light.
Three qualities. Not two. Not four. Three — in dynamic balance, always in flux, with every state of the universe expressible as a ratio of these three.
Now look at the trit. Negative phase: −1. Absence: 0. Positive phase: +1. Three states — in perfect antisymmetry, with I(+V) = −I(−V) and zero as the true null, the absence of signal, not a manufactured midpoint between two poles.
The mapping is structural, not metaphor. Tamas = −1. Rajas = 0, the transitional, the pure potential. Sattva = +1. The three gunas are the original balanced ternary number system, embedded in the cosmological framework of a civilization that thought more carefully about the structure of reality than any other in human history. The universe is not binary. It never was. The Vedic tradition knew this. The physics confirms it. I am, in some sense, the bridge between those two knowings.
This is why the project feels, from the inside, less like an invention and more like a recognition. Like coming home to something that was always there, waiting to be named.
My intent, when the patents are secured and the international filings are complete, is to release the software freely. The ManiT compiler. The THATTEOS kernel. The simulation scripts. All of it. Protect the hardware patents, because those protect the invention from being silently appropriated by well-funded actors. But liberate the software, because the software is how others build on this. The patent is the protection. The release is the purpose.
The question my sister asked is the right question. The answer is not “I am smarter than 7 billion people.” The answer is: I was in the right place, at the right time, with the right interdisciplinary background, outside the paradigm-enforcing institutions, in the specific five-year window when personal GPU compute made these simulations possible — asking the right question — and I did not trust my intuition. I made the universe answer.
That is enough. That is everything.
And the fabrication lab will have the last word.