Who Is Jack Levin? Google Employee #21, ImageShack Founder and the Man Behind XEN and X1

Google’s 21st employee wired 2,000 servers in a week, built the image host that carried early Twitter, mined Bitcoin at a dollar, gave away XEN and now runs X1. He sat down with Owl of Atena for X1 Report.

He brought deodorant

Owl of Atena had asked Jack Levin to bring a couple of objects to the interview, anything at all, and not to think about it. She expected biohacking gear. He held up a stick of deodorant and a bottle of Maui Babe Browning Lotion.

“Smelling good is important, especially in Hawaii,” he said. The lotion has a coffee base. “It makes you smell good too. Like coffee.”

Why those two things? “You told me not to think about it, so there’s no why. I just saw them and I’m like, okay, I’ll bring them.”

It is a small moment and it tells you most of what you need to know. Levin does not perform. He has a resume that most founders in crypto would put on a billboard, and he spent the first five minutes of a career interview talking about tanning lotion, then about a peptide container he bought because he liked the shape of it. “I don’t use it, actually. I just have it because I like the container.”

Then he said the line that runs under everything else he has done: “I do technical experiments, I do biological experiments on myself. That’s kind of like what I do.”

This is a profile of an experimenter. Over ninety minutes he covered Google, ImageShack, Bitcoin, XEN and X1, and returned again and again to a single idea: the technology is never the hard part.

From St. Petersburg to Silicon Valley

Jack Levin was born in 1974 in St. Petersburg, Russia, and emigrated to the United States with his family in 1990. He got his first computer at 16 and taught himself by downloading and reading documents from Usenet groups.

He studied computer science at the University of Missouri, starting at the St. Louis campus and finishing his degree at Columbia around 1997. Then he went west.

His first Silicon Valley interview, at Silicon Graphics, went badly. By his own account he asked for too high a salary too soon. He landed instead at Chromatic Research, a company building media acceleration chips for DVD and MPEG video, where he worked as a systems engineer from 1997 to 1998. A year of independent consulting followed.

In that period he built what he describes as the first open source embedded computer running Linux with a firewall and a VPN. That box is what put him in front of two Stanford graduate students with a search engine.

Google #21

Levin met Sergey Brin and Larry Page in 1999, when Google was a handful of people, and joined that summer as employee number 21. On his first day, he has said, Page handed him cables and told him to plug them in.

His first real assignment was bigger.

“They delivered the servers and they said, Jack, this is your first task. Go to the data center. The servers are waiting. Turn them on, configure them, set them up, because next week we’re turning on a big customer called Netscape. And we need capacity, because we’re out of capacity.”

There were 2,000 machines. After a few days of connecting them by hand he noticed something worse than the workload.

“I realized that there is no design of any kind. Nobody designed anything. It’s just a bunch of servers.”

He had never designed a large network. “Think about it. Nobody ever gets 2,000 servers in one day. You get maybe two, three, four, five.” So he learned network design on the spot and wrote an automated system to bring the machines up, because doing each one by hand was impossible. Within a week it was live.

“I was very proud because everything worked, and I was surprised that it worked, actually. I was a young guy who had no idea how to work for a large infrastructure company. But Google was not a large company.”

He went on to design Google’s early network and its first data centers, and to handle the infrastructure side of early partnerships with Netscape, Yahoo, AOL and others. He calls the period “assembling an airplane while it was flying.” One of the racks he wired later ended up in a computer history museum.

The day he knew

For its first two years, Levin says, Google did not know how to make money. He thought the company might fail, or be bought. Yahoo came with an offer he remembers as $2 billion, and Google said no.

“Larry Page was always a crazy guy who thought hundreds of billions of dollars, trillions, and nobody believed him. But they said, well, he’s convinced, so let’s see what happens.”

Then someone at Google, not a founder, found a rival search engine that returned only paid links dressed up as results. The idea: combine that with real search, and show sponsored links that are relevant to what the person is looking for. Revenue arrived within three to six months. Levin was one of the people watching the numbers.

“I saw revenue almost up to $7 million per day coming in. And I thought, it’s so much money, there is no way this company is going to be unsuccessful.”

What Google taught him

Asked to explain why Google worked, Levin goes back to its first name, BackRub. A page ranks because other pages link to it, and a link from a site with standing counts for more.

“If whitehouse.gov is linking to your stupid website, it doesn’t matter what you put on your website, you will show up at the top.”

Google, he says, was “tapping into the human intent to link websites together.” Importance was measured by social acceptance. He has applied that reading of the web to everything since: systems win when they capture what people already want to do.

Mid-interview he tested it live, typing “largest SVM fork” into Google. Up came his own X post from February 23, 2025, when X1 had 230 validators. “Very cool,” he said, and moved on.

The second lesson came from Page, his mentor: take on the harder project, because almost nobody else will, so you will have little competition.

The third was about giving things away. Levin was there when Gmail launched with far more free storage than anyone else offered.

“All the other competitors were in shock. They said there’s no way we can compete. This is crazy.”

He left in 2005, after about six years. Google had passed 8,000 employees and he no longer felt he had an impact.

ImageShack: an $80 hosting bill

Levin started ImageShack around late 2003 with his brother, as a side project while still at Google. The first hosting bill was about $80. He has said he never put meaningful personal capital into it beyond that.

It was profitable almost immediately, because he built its storage on the clustering approach he had learned at Google, and because the company got very good, very early, at Google’s new ad products. It stayed lean, often under 20 employees, while serving billions of images and tens of millions of monthly visitors. At its peak it ranked 11th on Alexa’s list of the world’s websites.

ImageShack became a major image host for MySpace. Then, in 2009, it launched yfrog, which became one of the main ways early Twitter users shared photos and video, before Twitter had media of its own. Levin knew Twitter co-founder Evan Williams from his Google days. Sequoia invested around 2008.

Then Twitter built its own image hosting. ImageShack moved to a subscription model in 2014, and Levin co-founded Nventify, whose Imagizer engine resizes and compresses images on the fly so that customers do not have to store every size.

The dependency question

Owl put the obvious question to him. He built on Twitter, and Twitter replaced him. Now X1 is built on Solana’s code. Is that the same risk?

He does not think so, and the distinction he draws is clean.

“You’re not dependent on a company if their code is open source. You can just take it, modify it, and as long as you maintain the open source yourself, you can continue to build and iterate. You can take their ideas and implement them, and remove the ideas you don’t like. We can also give them ideas so that they can improve their chain too.”

A closed platform can switch you off. Open code cannot.

The angel years

Through the 2010s Levin was an active angel investor, writing seed and pre-seed checks into consumer apps, SaaS tools, media and ad tech. He has described doing five to ten deals a year at his peak. Public databases document roughly a dozen; reported check sizes run from about $55,000 to $150,000.

Other names that appear in his portfolio include Petcube, Giftcoin, Hercules, Zero App, Five, GogoCoin and 200 Labs. Outcomes are what angel outcomes are: a few exits, several shutdowns.

One of those checks paid off in an unexpected way. When Levin’s X account was hacked in September 2026, he wrote to Bier, by then head of product at X. “I’ve been investor in his old company, he certainly remembers me.” The account was back within the hour.

The list matters less than what it built. Between Google, ImageShack and a decade of early checks, Levin ended up with the network of someone who was in the room for Web 1.0 and Web 2.0 both: early Google engineers, Sequoia partners, the founders he backed. Some of those people followed him into X1.

Bitcoin at a dollar

In 2010 a friend from Google told Levin about Bitcoin. He liked it at once: peer-to-peer money, simple and powerful, built on ideas that were 20 or 30 years old and had only now become practical.

He started mining in 2011. He had a data center with four racks of 35 servers each, three Nvidia and ATI graphics cards in every server, and spare electricity. Bitcoin was trading between $1 and $3. The plan was to offset his data center bills. He never sold the coins to pay them.

He moved to ASICs in 2014 and stopped when his hash power could no longer compete. Along the way he gave bitcoin to friends as gifts. One of them later bought a house with it.

XEN: the coin you mint

Levin launched XEN in 2022 to bring the ethos of first principles back to crypto: self-custody, trustless consensus, decentralization.

It came out of what he first called the Fair Crypto Foundation. XEN has no pre-mint, no owner, no admin keys and no fixed supply. Nobody buys it from a team. People mint it for themselves and wait, and the waiting is part of the design.

In the interview he tied XEN straight back to Gmail.

“XEN was not the coin you buy. It was the coin that you mint. So you can basically argue that it’s a free coin, and there is nobody controlling the supply. The community controls the supply by minting. And that’s why it became so popular.”

The name of the foundation was the thesis. Most of crypto is player versus player: someone launches a coin, others buy, the launcher sells.

“We wanted crypto to be fair, where people do not make money against you. XEN has no player that messes with you. It’s just you and the protocol.”

XEN launched on Ethereum and repeatedly pushed gas prices up on its own. When it went live on PulseChain on day one, Levin says, it was the top application on the chain for months. He recalls telling Richard Heart he should say thank you, since XEN was burning PLS and giving the chain the usage it wanted. Today, he says, about 185,000 Ethereum wallets hold XEN.

He gives Heart credit for understanding marketing, and agrees that every chain needs what Heart calls pumpamentals at its core. But he thinks PulseChain arrived too late. “EVM was on the way out,” he says, and eight-second blocks against Ethereum’s twelve were still slow.

Owl, who has been in the community since the start, described what XEN meant at the time: it arrived just as the centralized lenders were collapsing, and brought self-custody and first principles back into the conversation. Levin’s own description of the people who gathered around it is shorter.

“We have a community of explorers. We explore all kinds of new things, and those things are super fun. Not all of them make money.”

X1

X1 is where the threads meet: the infrastructure engineer, the open source believer, the fair-launch designer and the investor with a deep contact list.

Levin founded X1 Labs, a Delaware corporation, to build the chain, and has announced that it “is now registered with SEC”. Among the people who put money in the company are Greg McAdoo, the former Sequoia partner who was at the firm when it backed ImageShack, and Steve Ptucha of Solomon Labs, formerly a crypto product manager at PayPal. One comes from the venture world that funded Web 2.0, the other from the payments company that took crypto to the mainstream. Both chose to back a chain built by a man they had reason to trust with infrastructure.

Why Solana’s code

Levin’s community started on EVM chains. Leaving them was a deliberate decision, and he explains it with a survey of who actually won.

Ethereum, he says, won the first battle on vision, not engineering. “Ethereum is pretty bad technology. It’s very slow. It’s expensive. But people believed in the vision.” The EVM chains that came after mostly succeeded only where a large company already had the users: Binance, Coinbase with Base, Robinhood. Technically excellent chains without that struggled. He names Fantom, an EVM chain he knows well and calls “way ahead of its time, one of the best, fastest blockchains”, which never got real traction, and Avalanche, fast and interesting but short on revenue.

Solana had a vision simple enough to say in a sentence. Anatoly Yakovenko was an engineer who kept arriving late to trades, and built a chain where being fast was the whole point.

“I believe that Solana will win as a platform, and there will be more people familiar with that platform. EVM ultimately will fragment into the powerful companies that run EVM, which is fine.”

What X1 changes

If Solana wins as a platform, the opening is in who gets to run it. On Solana, consensus votes are transactions that cost SOL, every day, which is why Levin says a validator needs a very large stake before it earns anything.

“For us, the network does not charge you the fees. You can add your validator, you just pay for the validator itself, and you can earn your rewards by validating the blocks. Super cool that anybody can join.”

That is X1’s economic bet: a credibly decentralized chain where joining the validator set costs a server, not a fortune.

X1 mainnet turned one on October 6. “Knock on wood, we haven’t gone down, pretty much ever,” he says. “Maybe we were lucky. Maybe we were just very, very careful.”

Two fixes he sent upstream

Levin had never written Rust before X1. He learned it by improving the Solana validator, and sent the improvements back.

The first was almost embarrassingly simple. Running a validator with no traffic, he saw one thread pinned at full load. The component that processes consensus vote messages was polling for data in a tight loop with no pause.

“I installed a 10 millisecond delay between reads. As soon as I did it, it improved the validator speed and removed the CPU being abused by the bug. I gave it to them and they’re like, wow, this is really simple, I don’t know why we didn’t do it.”

The second was deeper. When consensus fails on a Solana-style network, he explains, every node ends up on its own fork, and the chain needs a coordinated restart to get back to a common state. That recovery code was not working properly. Levin spent a couple of months on it with Anza CEO Brennan Watt, feeding him reports until the fixes landed. Now, he says, a stalled chain can be revived in about ten minutes.

He is clear about why he did it.

“I think that’s how they started to believe me that I am not a scammer. I’m not just there to take their baby and make a ton of money with it. I’m actually contributing back to their codebase, to show that we have goodwill towards their success.”

Owl reminded him of the day Yakovenko posted: “@mrJackLevin you are our only hope.” Levin thinks it had less to do with the fixes than with licensing. Other teams were taking Solana’s code and closing it. X1 stayed fully open.

“I do not own X1 code. The X1 Foundation does not own the code. Our mandate is to create value on the chain.”

Fantom is his counter-example. It became Sonic and closed its source, and in his view the team was worrying about controlling the code when it should have been asking how to make people excited to be on chain. “They were thinking about the wrong things at the wrong time.”

He is also thinking about the next limit. In today’s design a single leader receives every validator’s vote directly, which is fast but caps how large the network can grow. Aggregating votes through subcommittees with BLS signatures, as Solana’s coming consensus does, lifts that cap at a cost he puts at perhaps 10 to 20 milliseconds. He treats it as an engineering trade, not a creed.

How he thinks

Technology is not enough

Asked for the hardest lesson of twenty-five years of building under pressure, Levin gave an answer in two halves.

“It taught me that we can do anything. Everything is possible. In technology, anything is possible. In business, maybe not everything is possible, because your business ideas are stupid.”

He has watched too many beautiful systems die to believe otherwise. “The technology is not enough. You have to have the vision, and you have to explain this vision to people so that they can understand what you believe in, and see if you can get like-minded individuals to build together with you.”

What not to build

This produces one of his more surprising management rules.

“If me or somebody on my team gets excited about the code, I generally tell them: do not build this. Because that’s the wrong thing to be excited about.”

The right thing, he tells them, is value on the chain: whether XNT is worth more because of what you shipped. “Technology will always be here. But if people are not coming to use your technology, then it’s useless and worthless. You’re the only person who will be excited, and the rest of the people will think you’re boring.”

His reasoning is not cynical. When people on a chain make money, they can afford to build, and more builders arrive. He points to Aptos, which he says has a dozen PhDs working on it and whose one breakout app was a game about tapping a cat. “I promise you, not a single PhD thought this would work.”

“We don’t care about how things work. We want to be entertained, we want to make money, and we want to believe in something. If you have those things combined, you have the recipe for success.”

Do not tell people what to do

Levin holds to first principles: open source, decentralization, self-custody. He also sees plainly that most people arrive for other reasons. The believers are the smaller group, and they are the ones who stay.

Where he parts ways with critics of memecoins is on what follows from that.

“People like to do it, and we shouldn’t tell people what to do, even if they lose all their money. If somebody wants to spend everything on something and have fun doing it, why should we tell them no? There’s no government looking at you and telling you how to do things in crypto. That’s ultimately one of the tenets of the first principles.”

He had the token launchpad idea himself, a year before pump.fun, and called it Token Factory. He did not build it, because he was on Ethereum then and knew his own community could push gas to 200 gwei with a single launch. “We would just blow it up into pieces.”

Say yes first

Owl asked about the power of saying no. He turned the question over.

“I say yes more than I say no. I say yes a lot, and then I’m like, 90 percent of it is wrong, let’s rethink this. It’s better to have ten ideas and use one or two of them than to have no ideas at all.”

“It’s important to be optimistic and open-minded and hopeful, and a dreamer.”

AI, he thinks, has made this way of working available to everyone. Take ten ideas, most of them bad, and have a model lay out the case for and against each.

“Now no one can say that they are not smart enough, because with AI everybody is smart. We live in a golden age of intellect and cognition.”

The other experiment

Owl introduced the subject as biohacking and human performance. Levin’s summary was shorter: “I just don’t want to be fat.”

He has trained for about ten years, and says that over the last couple he got lazier. So he ran an experiment on himself with testosterone, though his levels were already medium to high.

“I believe that I am a high performance human. I wanted to live my life at the peak level. Peak with thinking, peak with sleeping, peak with the human body.”

By his account he gained 12 kilos in three months and went from six or seven pull-ups to twelve, which he demonstrated on camera. “It used to hurt a lot, and now I feel like there’s no gravity.”

The change he dwells on is not physical. For years he would wake in the night with a list running in his head: do this, do that, why am I not doing the other.

“One night I woke up and my mind is absolutely quiet. Not a single thought. And I’m thinking to myself, where are all the thoughts? It became completely silent, and I had the best sleep that night.”

“I can look at you and I can think nothing, and there’s no interference.”

He knows the objection, that pushing the body this hard may shorten his life. He has an answer ready.

“I would rather die earlier than be a snail for the rest of my life. A snail that lives a hundred years is not worth living. A jaguar has a different type of life, because every year, every minute, it is as if you live a ten times richer life.”

This is one man describing his own choice, not advice. But it is the same instinct that wired 2,000 servers in a week: find the limit, then run at it.

What he is working on now

Levin’s focus today is the one he sets for his team: bringing value to the chain. He is working on new initiatives to do exactly that.

The pattern

Put the chapters side by side and they are the same chapter. Google gave away search and two years later found the revenue. ImageShack gave away hosting. XEN gave away the coin. X1 gives away the right to validate.

Each time, Levin made the hard infrastructure work first, kept the team small, and let people in for free. Each time, the thing grew larger than the idea it started as.

“We’re all becoming part of history, one way or the other,” he told Owl. He has already been part of it more than once.