Beniamin Mincu

“Had I had been present at the creation, I would have given some useful hints for the better ordering of the universe.”

  • Library of Alexandria
  • Sherlock Memories
  • Technius
  • About

On October 9, 1903, The New York Times ran an editorial titled “Flying Machines Which Do Not Fly.” It argued that a practical flying machine might take one million to ten million years to develop. Just 69 days later, on December 17, 1903, the Wright brothers achieved the first controlled, sustained powered flight at Kitty Hawk.

Notes On The Future, From the Past

August 25, 2026 by Beniamin Mincu

Epistemic status: this is an autobiographical note. People reading this will not find any announcements between the lines. Instead, the careful reader may discover a few clues about the road to Supernova, and what may follow.

I

I am in this small-ish white room. Alert. Patient. Curious. Been looking forward to this conversation. One by one, a few guys enter and take a seat. A few chairs have been pulled together. Nothing remarkable about the room itself, nothing formal. Just an older building. A plain office room. With a desk or two. There’s something in the people's eyes. A spark. Curiosity maybe. A little anxiety too, from what I can tell. The kind that comes when you know you’re about to step into untried waters. Maybe you have a sense you’re about to be tested, but you’re not entirely sure how. Now everyone is here. Over the previous days, I’ve taken some of the hardest and, to me, most fascinating problems in computer science, broken them down, and given one to each of them as a personal challenge. 

The problems matter. As strange as it may sound, I want a new and better solution for each of the problems. I want that solution in an impossible amount of time. But today the problems are only partly the point. The challenge is a filter. If you look at one of these problems and feel daunted by it, if your first instinct is to retreat from its size or complexity, there probably isn’t much I can do to help you. But if you look at the same problem and its difficulty pulls you closer; if it begins to feel less like a wall and more like a puzzle; if, after staring at it long enough, you start to sense a dim, intuitive path toward a solution that comes to you before you can even explain it, then we have something. Then we can really play. I look around the room and take the moment in. I smile. And, for a split second, wonder what may come of it.

As it turns out, that was the first Elrond core meeting.

II

Blink for a second, and two years have passed. It’s Friday. Nine in the evening. The team is cranking hard. The room is full of energy. The last few weeks have finally begun to pay off after a long and arduous stretch of work. A wild 30x improvement in core performance, just in time for the early testnet release. A testnet, dear reader, is an important step-change. A whitepaper formalizes an idea. A prototype asks whether it can work at all. A testnet is where the idea finally has to survive contact with reality. And run. It is the closest thing yet to a real, working product.

I stop and think out loud, as if to make sure this is all real. Thirty times, I think. Almost unbelievable. Oh, how broken that prototype was. Beyond a certain threshold, it was awful. It failed us almost every time. Written in a language nobody liked. Sorry, Java readers. But that broken prototype had done one thing remarkably well. It demonstrated, for the first time, that you could architect transaction flow in a completely parallelized way, without what’s technically called “double spending.”

That was the interesting part. In financial systems, preventing the same value from being spent twice is one of the fundamental constraints. The hard part is preserving one coherent financial truth while many transactions happen at once. If you could do that in parallel, without losing correctness, the ceiling would change completely.

If this new architecture held, transaction capacity could scale to levels no other network in the world had previously reached.

My mind wanders. Zooms out. And for a moment, the whole stream of reality rearranges itself. A breakthrough. A real, genuine breakthrough. But not a finished product, yet. More like discovering that there is, in fact, a path through the mountain. The path itself is riddled with problems. You solve one and earn the right to face the next. You keep climbing until the thing works, or you never emerge on the other side.

And as we climb, something interesting begins to happen. The higher we get, the more people can see us. No, actually, the more people want to see us. Larger and larger investors start reaching out. They want a piece of this. A piece of what we’ve been building. Then a friend sends me a note. Binance is reviewing us. Binance, at the time, is the largest exchange in the world. And it is known for one thing above all: a pure execution machine. Its products beat the competition, day after day. The culture is 9/9/7. Nine in the morning to nine at night, seven days a week. Moving, experimenting, shipping. New products keep appearing on every front. The review itself is a funny thing. We don’t really know what it’s for. Maybe Binance investments. Maybe Binance Launchpad. Maybe something else. What we do know is that they are checking everything. Every post we’ve published. Every feature we’ve shipped. The robustness of the architecture. The soundness of the assumptions beneath it. The go-to-market strategy. Everything we have done up until that point is suddenly on the table. And the people on the other side are good. Really good. Their questions have a savvy, razor-sharp quality that is almost uncanny in this space. They know where things tend to break. They know where people tend to bullshit. And they keep digging. With a good nose for where the value sits.

We speak almost every day. Sometimes mornings. Sometimes evenings. Sometimes in the middle of the night. And slowly, through these conversations, I begin to sense something. They get it. They actually get it. They understand what we are trying to build. They see the hard parts. They appreciate what has already been solved. There is a certain respect in the air. Another month passes. Then I receive a private invitation. I am to meet someone. No one I ask seems to know exactly who he is. I’m only told that he is very sharp, and high up inside Binance. By coincidence, I am in the US. New York, to be precise, for the Consensus conference. That is where we meet. The guy has this intense focus in his eyes. He listens very attentively. Probes. Asks a few questions, then follows the answers where they lead. The meeting ends, and I walk away realizing I cannot fully read it. It should be positive. I think it’s positive. Is it positive? No idea.

The Binance review has now been running for more than a month. Every day brings another question. Another piece of feedback. Another idea. And hanging above all of this, an unstated fact becomes more and more obvious. Thousands of world-class projects want to get into Binance Launchpad. One actually gets picked. Brought inside, accelerated, and eventually put in front of the world on the Launchpad platform. The intensity of those weeks is unlike anything I have felt before. Every piece starts moving faster, all at once. The technology. The team. The investors. Binance. Something that had existed for years as an idea, then a prototype, then a stubborn little group solving impossible problems in a room, is suddenly gathering momentum. The acceleration is dizzying. And somewhere inside all these questions and late-night calls, I can see why they keep coming back. They see an edge in us. Something familiar. Maybe, for once, they see a little of their own edge in someone else.

III

It’s July 2020. 2am. I am awake. I am also completely dead. Pain at levels I didn’t know I could take. Body finished. Mind racing. We’ve had a full day in Bucharest. Not the best of days though. We spent most of it on a filming stage. After years of wondering whether we would ever actually reach launch day, we have somehow decided that reaching it isn’t enough. We’re also going to film a live session, on a proper stage, with the core team, while simultaneously setting the actual network in motion. The filming part now sucks. Big time. It’s awkward. Slow. And painful. There is too little time for the filming crew to acclimatize to us, and us to them. Too little time to grasp each other’s rhythm. Too little time to calibrate the stage, the conversations, the cameras, the live show. Take, say, a week or two and this could probably be pretty decent. Maybe even a bit fun. But try to compress it all into twenty-four to forty-eight hours and it becomes torture. 

There is a strange paradox in moments like these. Under enough pressure, time seems to compress, attention narrows to a single point, and reservoirs of energy you did not know you possessed suddenly become available. This is when you learn what flow really means.

It’s evening. Things are not falling into place. Too many things. Too many voices. Each trying to drive things in a certain direction. It’s clear to me that today is lost. So I might as well reset, regroup, and plan the next steps. I leave the stage and come home. Well, not home home. Some random apartment we’re staying in for a few nights. I’m supposed to sleep. Instead, my mind keeps racing. There is still so much to do. And so little time left to do it. But the truly big part is actually done. Sort of. The network is all in place. Our technical team has been at it for months. Days, nights, weekends. Ironing out bugs. Refining performance. Patiently, meticulously getting the network ready.

Now, the telling of the story is the thing that needs to click. The two videos we’re working on are close, but still unfinished. Both need to be released tomorrow. Several people are working on them, sending new iterations every hour or two. Visuals almost done, now on to the right soundtrack. Meanwhile, the technical team is preparing the final launch sequence. One last round of checks before the network goes live. The product team is testing the apps. Marketing and community are preparing messages, coordinating people, making sure everything moves together. It’s 4am. I should probably go to sleep. A few hours from now it will be morning. I fall asleep two hours. A reset is good. Even when that reset is brief. 

Then I’m awake again. 6am. Ten hours left. Ten hours until go-live. Now the day becomes arithmetic. Every hour matters. Every minute matters. There is no room left for wasted motion. I’m back on the stage, and something has changed now. This time, we go in and make the rules. Our rhythm. Our energy. Our way of doing it. We stop trying to fit the stage and make the stage fit us. And then, slowly, it begins to flow. Then, I look at the clock. One hour left. Somehow, the entire day is gone. We’re almost ready. Almost. A few final uploads are about to come in, and we can begin the genesis live stream.

The tension is at its peak. But there’s also some new relaxation kicking in. We’re going to make it. We’re going to push this thing further than anyone has before, and set some records today. Backstage, the team is sitting behind screens. Checking servers. Checking scripts. Checking the final sequence. Checking that the message for the genesis block is there, exactly as it should be.

Years of work, compressed into a handful of minutes. All those nights. All those arguments. All those problems that looked impossible until, somehow, they weren’t anymore. All the things that broke. All the things we rebuilt.

Now, there’s nothing left between us and this genesis moment.

Ready or not, we are going live. 

Interlude

I’ve sometimes wondered why someone would do something like this. Why someone, in their right mind, would voluntarily go through the painstaking process of building something new in this world of bits and atoms.

It’s an odd thing, when you think about it. You see something that doesn’t yet exist. Then you press on, giving it years of your life. But why?

Maybe it’s pure curiosity. Maybe it’s an insatiable need to discover, to learn. Maybe it’s the urge to make something with your own hands. To take some small corner of the world that refuses to yield, wrestle with it long enough, understand it deeply enough, until eventually it does.

Maybe, in some sense, it is simply the desire to master some small part of reality.

I don’t really know.

What I do know is that the phenomenon is wild. Bewildering. Unreasonable.

Something inside a person begins to stir. A whisper at first. Easy to miss. Except it keeps returning. It pulls you closer. It bothers you. Refuses to leave you alone.

And if you listen to it long enough, you may eventually find yourself at two in the morning, completely exhausted, body aching, mind racing, years deep into some impossible thing you once decided had to exist.

And yet, somehow, despite all the unreasonableness, there is nowhere else you would rather be.

Maybe this is how it feels.

The mysterious pull toward something you were meant to do.

IV

So what did we actually achieve through all these years?

It took us a surprisingly long time to understand that the network was only one layer of the problem.

First came people.

More than 2.5 million of them came through xPortal. Together, they made more than 2 billion transactions, moving over 10 billion dollars. We integrated onramps, offramps, cards. We became almost obsessive about security, building things like Guardians, an onchain form of two-factor protection and one of the first systems of its kind.

But the point was never simply to build another wallet. It was to take a technology that felt foreign and complicated and make it disappear into the ordinary flow of someone’s life.

Then came a very different problem.

Merchants. Payments. The real economy.

More than 16,000 merchants. Roughly 1 million transactions. Around 1.5 billion dollars processed. Getting there pulled us into a world that, from the outside, can look almost impossibly boring. Licenses. Compliance. Banking relationships. Regulation. Audits. Risk controls. From the inside, it is anything but that.

We acquired and maintained licensed entities across Romania, Portugal, Liechtenstein. Became a principal member of Visa and Mastercard. Built onramps and offramps. Cards. And stablecoin rails.

An enormous amount of machinery. Almost all of it invisible. You hardly notice that machinery when it works. You notice it very suddenly when regulation changes, a banking relationship disappears, a door closes, and a business discovers that the product it spent years building can no longer legally operate.

Over time, our view of regulation changed. It stopped looking like paperwork around the product. It became part of the product. A protocol of another kind: the machinery that lets a new financial system cross into the existing economy without breaking at the boundary. That became xMoney.

And underneath both of these, almost from the beginning, we were building something stranger.

A programmable financial balance sheet. People call it a blockchain. Technically, of course, it is one. But that description has always felt incomplete to me. It works more like a bank from the future.

Traditional finance stores money inside institutions and moves it between them through messages. A programmable balance sheet begins to collapse that distinction. You still use the regulatory license stack to connect to the real economy, but the asset, the settlement system, and the execution environment start to inhabit the same substrate. Money can sit there as a programmable object. Software can act on it. Settlement just happens.

The system does not close on Friday evening and reopen Monday morning. It does not sleep. There is no small committee deciding which applications may be built on top of it. Anyone can write a program, put capital into it, and let that program run.

Around 9.2 million accounts have used it. More than 622 million transactions have crossed it. More than 10,796 smart contracts have been deployed on top of it. Over time, more than 50 billion dollars in value have flowed across the network. That became MultiversX.

And then there are all the other things. Of which some worked spectacularly, for a time.

Maiar DEX, later xExchange, went from tens of thousands of users to hundreds of thousands. At its peak, billions of dollars moved through it in a single day. For a brief and remarkable period, it stood among the largest platforms of its kind in the world.

Then the world moved. Users moved. Liquidity moved. Markets moved. Something that had looked unstoppable became a fraction of what it had been. There are several stories like this. Products that caught fire and then faded. Ideas that were right for one moment and wrong for the next. Things we were certain mattered, until reality developed other plans.

The world does this. It keeps moving while you build. Sometimes faster than you can.

And there is a certain humility that comes from living through enough of these cycles. You discover that building something difficult is not enough. Being right is not enough. Sometimes even being early is not enough.

Seven years also does something else. It lets reality edit your architecture.

Markets remove some assumptions. Users remove others. Regulation removes a few more. Products fail. Others survive. A handful begin to compound. What remains is rarely what you would have drawn on day one. But perhaps that is precisely the point.

Because after enough time, you begin to distinguish the things that were merely clever from the things that are actually necessary. And if you have been paying attention, what remains may be much closer to the truth.

Look around today and that lesson becomes difficult to miss.

Companies with brilliant teams, extraordinary investors, beautiful products, enormous ambitions have disappeared. That sounds dramatic until you spend enough time around startups. Then you realize that death is not the exceptional outcome.

It is the default one. Survival is the anomaly.

I sometimes think about the companies that launched beside us. Competitors we studied carefully. Teams we respected, feared, learned from.

Many of them are gone. And there is something sobering about that.

We are still here.

Scarred? Certainly. Mistakes? Many. Failures? More than I would have liked.

But also something we could not possibly have had when we started. Seven years of reality pressing against us.

Seven years of discovering which assumptions survive. Which products people actually need. Which technologies matter. Which advantages compound.

And after all of that, somehow, we are still standing. Still building.

Only now carrying something we did not have when we walked into that first small white room: a few products people genuinely use; a few capabilities extraordinarily difficult to reproduce; and a much clearer understanding of the problem we had actually been trying to solve all along.

V

So where are we today?

Seven years ago, we walked into a small white room with a handful of problems that looked almost impossible.

Build a high-bandwidth, low-latency financial system. Then give anyone, anywhere easy access to it. That was the mission of Elrond.

At the time, we thought the hard part was technological. Build the network. Make it fast enough. Scale it far enough. Keep it secure. Get the machinery to work.

All of that was difficult. But it turned out to be only one part of the problem.

Seven years of building, learning, adapting, and occasionally getting things very right did something useful. Reality edited the architecture. And slowly, three layers kept surviving.

The first is the network.

With Supernova, MultiversX becomes a financial system capable of processing transactions in fractions of a second. Open. Programmable. Global. Always on. A place where money can move and software can act on it.

There is an important threshold below one second of latency. Above it, people wait for the financial system. Below it, the machinery begins to disappear. Money starts behaving less like a transaction you initiate and more like state that software can continuously act upon.

The second is the bridge into the real economy.

Licenses. Banking relationships. Cards. Stablecoins. Onramps. Offramps. Compliance. IBAN accounts. The dense, mostly invisible machinery required to connect an open financial network to the enormous regulated economy already moving around us. That is xMoney.

The third is the interface.

The thing people actually hold in their hands. Simple. Secure. Delightful. A way for a person or a business to hold money, receive it, send it, exchange it, spend it, invest it, and interact with new financial products without needing to understand the complexity underneath. That is xPortal.

A programmable balance sheet. Regulated rails into the real economy. An interface anyone can use. Perhaps this is the most surprising thing of all. We were never really building a blockchain. We were building a new way for money to move.

Across borders. Across markets. Across systems that, until now, barely knew how to speak to one another.

Seven years ago, in that small white room, the question was whether we could make the machinery work at all.

Much of it works now. Which leaves us with a far more interesting question.

What becomes possible when the machinery disappears, and money begins to move like information?

August 25, 2026 /Beniamin Mincu
  • Newer
  • Older

© Beniamin Mincu