Vihaan Vaghela

I was born in Bharuch, Gujarat, in October 2011, and moved to the United States at four — first Piscataway, New Jersey, then Richmond, Virginia, and Short Pump Elementary. Seven years. I came back to India in fifth grade, into the ICSE curriculum, and later moved to Mumbai, where I live now and am preparing for the IB.

Seven years is long enough to be shaped by a place and short enough to return and find yourself slightly foreign in both directions. American classrooms taught me to argue in public, to say the half-formed thing and let it be corrected. That is a habit, not a virtue, and it does not read the same everywhere. Erin Meyer's The Culture Map was the first book that gave me vocabulary for it — that directness is not a fixed quantity but a setting, and that people who appear to disagree are often running the same argument at different volumes. I read it as a description of my own living room.

The practical residue is that I write things down. When you have moved between two places that assume different defaults, you stop trusting that context is shared.

On 14 September 2020 I opened Scratch. I was eight. I made games, animations and platformers, and I drew every sprite by hand — badly, and with great conviction. One of them was called Attack of Excalius. It had a boss fight I could not balance and a title screen I rewrote more times than the game.

Then Blockly, then an mBot Neo, and a discovery worth more than the robot: I did not care that it moved. I cared that the logic was right. The motors were an output device for the part I actually liked. Robotics was not it. Software was.

The next part was accidental. My father worked remotely, and I overheard the vocabulary of professional engineering before I had any of the concepts — GenAI, DevOps, Jira, RAG, enterprise systems. Words first, meanings later, over about a year. It is a strange order to learn in and I would not recommend designing a curriculum around it, but it left me unintimidated by the jargon, which turns out to be most of the barrier.

Then Copilot, then Claude. I want to be precise about what changed, because the honest version is less dramatic than either the enthusiasts or the critics claim: it removed most of the typing and left all of the thinking. The bottleneck moved from "can I express this" to "do I know what I want". That second question is harder, and it does not have a keyboard shortcut.

When I was about ten, in New York City, I watched a traffic light hold a red for an empty road. The question was not sophisticated: why is this thing static? It has a sensor budget, a road full of information, and it is running a timetable written before anyone arrived. I did not do anything with the question. I was ten.

In October 2025, driving back toward Ahmedabad after a trip through Rajasthan, we were in a minor accident. Nobody was badly hurt. I want to be careful here, because the tidy version of this story is a lie: I cannot tell you a signal caused it, and I have no evidence that a better one would have prevented it. What it did was return me to a question I had left alone for four years, with a great deal more urgency attached to it.

That became EXTREMIS: adaptive signal timing, emergency vehicle prioritisation, routing, prediction. Then it kept growing — forecasting, several models with different jobs, a policy layer, decision engines, explainability, and documentation that eventually outweighed the code. At some point the name stopped describing the thing. It was also, I discovered, thoroughly spoken for by Iron Man.

So: VECTOR — Variable Environment Control Through Observation Response. Traffic is the first application, not the subject. The subject is what it takes for a system to observe an environment, decide something about it, and remain accountable for the decision afterwards.

VECTOR stopped being a project somewhere along the way and became a method. The rule that did it: if I could not explain why a component existed, I did not understand it well enough to keep it. Writing the documentation is not what I do after the thinking. It is where the thinking happens, and it is why the archive is the largest part of this site.

I should say what that actually looks like, because "documentation-first" makes it sound tidy and it is not. For six weeks the system had a p95 latency around 100 milliseconds and I could not find out why. I read the same profile every night. I went to school on four hours of sleep and thought about queue depth during maths. When I finally found it — Sentinel, the anomaly detector, quietly eating ninety-seven percent of the runtime, the one component I had never suspected because I had written it first and trusted it longest — I did not feel clever. I felt stupid about the six weeks. And then I felt the thing that keeps me doing this, which is not triumph but quiet: a problem that had been in the room for a month and a half simply was not there any more. p95 went to 1.124 milliseconds. I sat in the dark and looked at a number and felt completely, ridiculously happy.

My classmates think the amount I write about my own code is faintly unhinged. They are not wrong. I once spent an evening on a document arguing for a component and concluded it should not exist, which means the entire evening produced nothing except the decision not to waste a week — and I would call that a good evening. I started writing because I thought universities would like it. I keep writing because I like it, which is the part I did not see coming.

A misconception worth addressing directly: people sometimes read me as arrogant. The mechanism is that I disappear. When I am working I am entirely inside the problem — the architecture, the documentation, the plan — and I go quiet in a way that is easy to mistake for indifference. The people closest to me know the difference. I have stopped trying to argue anyone out of the first impression; consistency handles it eventually, or it does not.

What I am actually after is best described as sovereignty. Not in any grand sense — just that I would rather originate a thing than execute someone else's version of it. Lead it, own the decisions, own the failures without redistributing them. Agency is the part of the work I am unwilling to trade away.

A few things that follow from that, and that show up everywhere on this site:

• Understanding, not recall. I do not measure myself by how much syntax I have memorised. I measure myself by whether I can explain every line. If I can, I understand it. If I cannot, I have not finished.
• A system should be legible. If nobody can explain why it did what it did, its performance is not the interesting number.
• Authority is not a feature. Anything that can act on your behalf should be constrained by something you wrote down, and you should be able to change your mind at any point.
• Write it down. A decision whose reasoning was never recorded gets re-litigated every six months by someone with less context, usually me.

I cycle at national level, hold a first-degree black belt in Taekwondo, run trails, and play badminton with no ambition whatsoever, which is the point of badminton.

I climb. Johnston Canyon is still the one I measure others against. Mountains and software have the same shape, and it is not the shape people usually reach for — it is not about summits. You take a step, then another, and eventually you arrive somewhere with a view. Then you keep going, because the viewpoint was never the destination, only the place where the work becomes briefly visible. Endurance sports teach patience by making impatience useless.

I read less than I should and re-read more than I admit. The Culture Map is above. Mark Manson's The Subtle Art of Not Giving a F*ck did something narrower but useful: it made me audit what I was actually spending attention on, at an age when the answer was mostly other people's opinions.

I have one friend I would call a brother. I am careful about trust and I do not distribute it widely, which I think is the correct setting even when it is lonely.

People who arrive here through the essays tend to expect someone more serious than they get. The writing is careful because writing is where I am careful. In person I am overdressed — I own more blazers than a fourteen-year-old has occasions for, and I will wear one to a physics exhibition, a family lunch, or a Tuesday — and I cannot sit still, which everyone around me has commented on and I continue to deny. I skip rope between debugging sessions. I take stairs two at a time for no reason. I have a sarcastic streak that does not survive translation into prose, so it mostly stays off this website, and the people who know me offline will tell you the archive is the quiet version of me.

I write for the record. Not of the things that worked — of the thinking, including the parts that were wrong. Someone reading an essay from 2026 beside one from 2036 should be able to see the difference immediately, and the only way that works is if I leave the early writing exactly where it is. It stays not because it is good. It stays because it is not, and because deleting it would quietly rewrite the history of how I got here.

This is not a portfolio and it is not a personal brand. It is an archive of growth, kept in public, built to outlast whatever framework it currently runs on.

So here is the ambition, and it is smaller and stranger than the one people expect a fourteen-year-old to have. I am not trying to arrive somewhere by a particular age. I want this archive to still be here in thirty years — three decades of entries, sitting in order, so that someone can read the whole thing end to end and watch a person change their mind in public a few hundred times. That is the actual project. Everything else I build is something to write about.

The newsletter is the reliable way to follow along; email is the reliable way to reach me.

This is chapter one. I will see you in the next entry.