Part IThe Story
Two Coffees and a Beer
The story is always told with the drinks first. Two coffees and a beer. People who repeat it, and in crypto it gets repeated constantly, tend to treat the order as a recipe. A venture capitalist named Steve Vassallo once explained the arithmetic on LinkedIn: the coffees sharpened the focus, and the beer quieted the instinct to overthink. Then he noted, with what seemed like real delight, that this was how a business he valued at $75 billion began.
What we know is less tidy. Sometime in 2017, Anatoly Yakovenko and a friend sat at Café Soleil in San Francisco talking about mining. Their subjects were proof of work, the consensus scheme Satoshi Nakamoto had stitched into Bitcoin, and why electricity mattered so much to the whole arrangement. The two had a side project building deep-learning servers, and they had started mining cryptocurrency on the GPUs to pay for the hardware. That arrangement had raised an awkward question, which Yakovenko later recalled plainly: why would anyone pay for the AI products? The conversation drifted. The caffeine did not.
By four in the morning he was awake in bed, tossing and turning by his own account, and he had an idea about time. Specifically, he saw that the passage of time could be encoded into a data structure. He did not know whether it was a good idea. He did a rough calculation, the kind an engineer does on the inside of his skull, and the result said a blockchain built this way could move somewhere between 1,000 and 10,000 times more throughput than Ethereum or Bitcoin did at the time.
There is a detail in a 2022 Fortune profile that the legend tends to omit. Caffeine, Yakovenko said, makes him "too jittery." He drinks tea now, or a decaf Americano. The founding fuel of one of the most valuable networks in the history of software is a substance its founder mostly avoids. I find this more revealing than the drinks themselves. The man is a systems engineer, "through and through," as Ali Yahya of Andreessen Horowitz put it, and the anecdote has the shape of a systems failure: a controlled variable briefly let out of control, one night of overclocking that produced a result, and then the variable put back where it belongs.
The idea itself came from an older machine. Yakovenko had spent nearly thirteen years at Qualcomm, where the cellular network depends on a protocol called Time Division Multiple Access. TDMA dates to the 1960s and 70s, and it is almost embarrassingly simple. You divide time into slots and give each transmitter its turn. Two radios that would otherwise shout over each other take turns instead, and more information gets through. Bitcoin, he realized, had the same problem. When two miners produce blocks at once, the network forks, one block gets thrown away, and the bandwidth spent on it is wasted. If the producers could take turns, and if everyone could agree on whose turn it was without first stopping to talk about it, the waste would disappear.
The hard part in a distributed system is agreeing on the clock. Every node on a blockchain is a stranger to every other node, and none of them trusts anyone's wristwatch. So Yakovenko proposed a clock that proves itself. A SHA-256 hash is computed, and its output is fed back in as the next input, again and again. The function cannot be run faster by running it in parallel, so the length of the chain is itself evidence that time has elapsed. "There's a way to create a historical log," he said years later, still sounding faintly surprised by it, "that has proof that events had some amount of time passed between them." He called it proof of history.
He would build a company around this, then a network, then a phone. He would watch the token that measured the network's worth rise from twenty-two cents to $260 and then fall below $8, in the space of about two years. He would be called the father of high-throughput blockchains and the enabler of digital slop, and he would accept both descriptions, the second one partly about himself. Under all of it sits the four-a.m. intuition of a man born in a country whose clocks, currency, and information had all been controlled from the center. A clock nobody owns. A history anyone can verify.
By the Numbers
The Clock and Its Keeper
$350Total money among seven family members when the Yakovenkos left for the U.S. in 1992
~13 yrsYakovenko's tenure at Qualcomm before founding Solana
100M+CDMA flip phones running BREW, the OS he worked on as a core kernel developer
Mar 16, 2020Date Solana's mainnet produced its first block
$260 → <$8SOL's November 2021 peak and its low after the FTX collapse
$314M2021 token round led by Andreessen Horowitz
42% → 29%U.S. share of open-source blockchain developers, 2018 to 2022, per Electric Capital
The Currency Was Bad
Seven people. Three hundred and fifty dollars. That was the sum of the Yakovenko family's portable wealth when they left the collapsing Soviet Union for the United States in 1992, about fifty dollars a head to carry across an ocean into a country whose prices they did not yet understand. Anatoly was eleven. He has said that the adults' anxieties were mostly kept from him. "As a kid, I wasn't privy to these problems that my parents had," he recalled on the Unlayered podcast. "But I was aware enough to know that the Soviet currency was bad."
Eleven-year-olds do not generally have opinions about monetary policy. This one did, in the way children sense weather. He has also said, in a sentence that breaks off in the transcript, "I do remember that time in 91," and then the recording moves on and we never learn what the time was like. The fragment is enough. A boy watched an entire state system dissolve, the money first.
Before that there was a dacha on the Black Sea. Asked once what made him happiest as a child, he gave an answer that reads like the opening of a Soviet children's film: "Making bows and arrows and running around our dacha on the Black Sea." It is the only pastoral image he has offered of his early life. After it comes Chicago, which in the early 1990s belonged to Michael Jordan. Yakovenko has said he arrived "right at the height of Michael Jordan's Chicago Bulls career, and was completely immersed in that craze." A child arrives from a country that has just ceased to exist and lands in a city where one man can apparently do anything. You can draw your own conclusions from that collision. He seems to have drawn some.
"I had this idea of a businessman that goes around and makes stuff happen," he told the journalist Ksenija Pavlovic McAteer in 2021, describing his childhood picture of America. "I thought that was exciting." Then, almost immediately, he corrected the record: "As an adult, I am an engineer, I like to get into the nuts and bolts of problems and solve them." The two sentences sit side by side in the interview like a before-and-after photograph. The boy wanted to be a man who makes things happen. The adult discovered he was a man who makes things work. The tension between those two ambitions never fully resolves, and the rest of his career keeps returning to it.
His parents were first-generation immigrants. They are proud of him, he says, and they have no real idea what he does. "It's hard for me to explain to them what I'm doing, because they don't really know," he said. "They barely understand how computers work, and what a decentralized computer is." He said it, by Pavlovic's account, warmly. The distance between a family that crossed the Atlantic with $350 and a son whose protocol would one day underwrite tens of billions of dollars in value is so large that it stops being a gap and becomes a different language.
As a teenager he taught himself C, his first programming language. He built computers and fooled around with early versions of Linux. He read. Asked once to name his biggest influences, he listed three names, none of them engineers: "Kafka, Burroughs, Godel." He explained the last one more than the others. "The idea of different kinds of infinities and Gödel's incompleteness theorem seemed really tied to existential questions of how we perceive the universe and what it really is." This is an unusual canon for a future blockchain founder. It contains a novelist of opaque bureaucracies, a writer of fractured and paranoid systems, and a logician who proved that any sufficiently powerful formal system contains truths it cannot prove about itself. A reader could connect these to a Soviet childhood if he wanted to. Yakovenko does not, at least not in public.
Of his adolescence, the period most profiles mine for trauma, he offered almost nothing. "It was just a boring time," he said. "You have all the energy, but no opportunities to have impact. Tons of reading, soccer, running, and tinkering with the internet." Energy with no outlet, which is a description of potential. His advice to young people, offered in the same interview, comes from someone who remembers that frustration exactly: "Don't listen to old people. The world is going to change and whatever advice they will give you is already stale."
He had reason to distrust the elders. He enrolled at the University of Illinois at Urbana-Champaign to study computer science at the moment the dot-com bubble was swelling. As a teenager he had felt "this magical possibility of writing a piece of code that just solves some incredible problem for the world," with a shot at becoming the next Steve Jobs or Bill Gates. Then the bubble burst while he was still in school. "Some advisors literally [told] me, 'maybe computer science is not a good career choice,'" he recalled. He ignored them.
Pavlovic, in her interview, noticed something about this second collapse: it was not his first. A boy who had watched a superpower's currency become worthless was not going to be frightened off his field by a NASDAQ correction. The bubble's bursting confirmed something he already knew. Systems fail, the wreckage is survivable, and the people who stay keep building.
The Flip Phone Years
The first company was called Alescere. Yakovenko co-founded it in his third year of college. It was a voice-over-internet-protocol startup building, in his words, "what kinda looks like Google Voice now," a telephone system for small businesses. He led development of the SIP and RTP protocol stacks, the plumbing that carries a voice across a packet network in real time, along with the server components. The work came out of a VoIP project in South Central Illinois, and it got him a job interview at a company in San Diego working on similar problems.
Alescere died. The Qualcomm interview did not. In 2003, with the dot-com wreckage still smoking, Yakovenko moved to San Diego and began what would become nearly thirteen years at one of the most consequential and least glamorous companies in technology.
There is a particular type of engineer Qualcomm produces, and the type is easy to describe because it is defined by its constraints. You write software for chips that cost almost nothing and live inside devices that people sit on, drop in toilets, and carry everywhere. Power is scarce, memory is scarce, and every clock cycle matters. Yakovenko became a core kernel developer for BREW, an operating system platform that ran on every CDMA flip phone, more than a hundred million devices. If you owned a clamshell phone in the mid-2000s and it ran on that standard, some of the code deciding what happened when you flipped it open was his. He has said as much himself, with the detachment of a man stating a tolerance: your phone likely uses Qualcomm products, and it may also use products he helped develop.
He did not stay on flip phones. He rose to Senior Staff Engineer Manager. He led a team of ten as lead architect for a software stack that offloaded high-performance work onto Qualcomm's Hexagon digital signal processor. The applications were augmented reality, virtual reality, 3D camera processing, 4K video post-processing, and Google X's Project Tango, the effort to give phones spatial awareness, which his team's technology made possible on Qualcomm hardware. He holds two patents on high-performance operating system protocols. None of this would make a magazine cover. All of it would turn out to matter.
The lesson he carried out of San Diego was a lesson about time, and specifically about the arrogance of software that assumes the hardware will stand still. "If you didn't write your software in such a way that it's just going to get better by the next chip you have to rewrite it," he explained on the Lightspeed podcast in August 2023, recalling the Qualcomm years. "It was like a waste of time to build it in a way that doesn't scale well with more hardware." At Qualcomm the next chip always came. The engineers who did well were the ones who wrote for the chip that did not yet exist.
He was also watching something else. Bitcoin had caught his attention in the aftermath of the 2008 financial crisis. One retrospective says he tried mining it early, and he later grew more interested in Ethereum's virtual machine, which he considered more inventive than Bitcoin even if it reminded him of JavaScript. But the thing did not "click," as he put it, until much later. "It wasn't until like, I think 2017," he said, "is when DeFi and protocols and Ethereum started getting, like not real traction, but at least like funding traction, is when it clicked in my head, this is a new environment for software."
A new environment for software. He has used versions of that phrase repeatedly. He had seen it happen before. "I kind of saw the birth of Linux and servers and kind of running the web as a platform," he told an interviewer in 2025. Then he had spent a career inside the mobile revolution, which he described elsewhere as participating in "all the mobile phone R&D projects you can think of." He was not a crypto evangelist who discovered engineering. He was an engineer who had lived through two platform shifts and recognized a third.
In July 2016 he left Qualcomm. He spent the next nine months at Mesosphere building a distributed operating system for data centers. In May 2017 he moved to Dropbox to work on compression, and he left in October. In the span of about a year and a half he worked on three of the hardest problems in computing in quick succession: the embedded kernel, the distributed system, and the compression algorithm. Balaji Srinivasan, interviewing him years later, called his path a "relatively unusual late in career migration to crypto." He was in his late thirties. In a field that celebrates twenty-two-year-old dropouts, that is late.
"I started thinking about how we solve scaling for these kinds of problems at Qualcomm with wireless protocols," he said, describing that 2017 turn, "and that's what really got me to go deep down the rabbit hole."
The phrase "rabbit hole" fits. Alice falls because she is curious, not because she is ambitious. Yakovenko had spent his career solving other people's synchronization problems for a paycheck. Now he had found one worth solving for its own sake.
Dividing Time
The technical claim at the center of Solana is easy to state and hard to appreciate. Every blockchain has to answer one question before it can do anything else: in what order did things happen? Bitcoin answers it slowly and expensively, by making miners burn electricity in a lottery whose winner writes the next page of history. Proof-of-stake systems answer it by having validators vote, which requires them to exchange messages, which takes time, and the time grows with the number of participants. Agreement about order is the bottleneck. It is the cost of trust among strangers.
Yakovenko's proposal was to separate the question of when from the question of what. Proof of history generates a verifiable sequence, a cryptographic clock, before consensus begins. Validators do not have to argue about timing because the timing is already encoded in the ledger. "You get faster 'time consensus,'" as the Wikipedia summary of his early descriptions puts it, while keeping the validation decentralized. The whitepaper, Solana: A new architecture for a high performance blockchain, was typeset in LaTeX. Its source still lives in a public GitHub repository, with 110 stars and 76 forks at last count.
The shape of the solution was Qualcomm's. TDMA had taught a generation of radio engineers that you could get extraordinary throughput out of a shared channel if everyone agreed on the slots. Yakovenko transposed that lesson from radio spectrum to a global ledger, swapping the trusted tower for a hash chain. The breakthrough was not invention so much as transplantation, the rare event of one field's obvious idea arriving in another field where nobody had thought of it.
A great deal followed from that one decision. A network with an agreed-upon clock can produce blocks in quick, rotating succession; an early testnet ran 400-millisecond block times and claimed 50,000 to 65,000 transactions per second across 50 nodes. A network that wants to use all that speed has to execute transactions in parallel. So Solana requires every transaction to declare up front which pieces of state it will read and write, which lets the system see where contention is. "When you have a lot of writes to a single part of the state, that becomes sequential," Yakovenko explained. "There's no way to parallelize it because somebody has to go first, somebody has to go second." Somebody always has to go first. Even in the fastest system, time keeps its order.
And a network built this way refuses, on principle, to split itself up. In the early years the industry's fashionable answer to scaling was sharding: breaking a chain into many smaller chains running side by side. Yakovenko thought it was a dead end and said so with the bluntness that would become his public signature. Asked in an early interview for his most controversial opinion, he answered: "Sharding is not going to succeed commercially. People will ship these systems, but they are going to be slow, expensive and hard to program." Solana would be one machine, one state, one giant ledger where any program could call any other program in the same atomic instant. He has described the goal in terms that sound like physics: "I really, really want to solve this like entropy problem. Why don't we minimize the error between all the useful stuff in the world that people care about and this representation of it and one kind of giant atomic view."
The project needed people, and the first people came from San Diego. Greg Fitzgerald and Stephen Akridge were both Qualcomm alumni, engineers formed by the same constraints Yakovenko had absorbed for over a decade: scarce cycles, unforgiving hardware, and the habit of thinking in microseconds. They were not crypto people any more than he was. They were people who knew what a fast system felt like from the inside. Several of the founders surfed, and Yakovenko, like some of the others, is an Ironman and an underwater hockey player. Underwater hockey, if you have not encountered it, is played on the bottom of a swimming pool with short sticks and a breath held. It rewards a high tolerance for oxygen debt and the ability to coordinate with teammates you cannot hear. As a metaphor for consensus among validators it is almost too neat. Yakovenko does not use it as a metaphor. He just plays it, and has said it should be an Olympic sport.
Then there was Raj Gokal, the one co-founder who did not come from chips. Gokal had come out of health technology, specifically Omada Health, where Balaji Srinivasan sat on the board and first knew him. He had also worked in venture capital and product. If Yakovenko and the Qualcomm engineers were the machine, Gokal was the person who could explain the machine to people with money. He became chief operating officer, and in a December 2020 op-ed the two of them would sign together as CEO and COO. It was an old and reliable division of labor between the man who knows how it works and the man who knows who needs to hear about it.
The project was first called Loom. Then an Ethereum-based project called Loom Network turned up, and the founders needed a new name. The way they found one is so casual it almost undercuts the mythology. Yakovenko and several co-founders surfed at Solana Beach, a small town north of San Diego. Lots of California companies, he reasoned, had named themselves after places in the state. "We were arguing on Slack," he recalled, "and eventually I just proposed 'Solana,' and that kind of stuck."
It is worth pausing on the word. In Spanish, solana means the sunny side of a mountain or valley, the slope that faces the sun. It is the opposite of umbría, the shaded side. A man who spent his early childhood in a country famous for its long winters, and whose family crossed into the American Midwest with almost nothing, named his life's work after the warm side of the slope. He has never, as far as I can find, remarked on this. He named it after a surf break. The etymology arrived uninvited.
— Anatoly Yakovenko, Unlayered podcast, 2024As a kid, I wasn't privy to these problems that my parents had. But I was aware enough to know that the Soviet currency was bad.
The Runway and the Crown
The worst time to start a cryptocurrency company in the last decade was early 2018. Ethereum and its peers were, by one account, shedding roughly ten percent of their value week after week. The initial coin offering mania of 2017 had flooded the market with tokens attached to whitepapers attached to nothing, and the hangover had arrived. Solana Labs formed its core team that year, published its official whitepaper, launched an internal testnet, and began raising development money, with Multicoin Capital among the first backers. Kyle Samani, Multicoin's co-founder, would become one of the network's loudest public champions, and later, along with Yakovenko, a named defendant in a 2022 class action over token offerings and disclosure. The plaintiffs voluntarily dismissed their claims against Yakovenko and the Solana Foundation, without prejudice, and the case proceeded against the remaining defendants.
The bear market shaped the company's character. In a down market a crypto startup cannot raise its way out of trouble, so it learns to do without. Yakovenko has described the posture plainly: avoid over-hiring, keep the team small enough that everyone matters, and keep enough money in hand that you are not forced back to investors every few months. It sounds like advice from a management book. It was survival.
The decision he describes as the hardest of his tenure came just before launch. To extend the runway, he had to let go of "really good people," engineers he still speaks of with obvious regard. "So, that was kind of really gut-wrenching because they were really good, really good engineers," he told the podcaster Luba Yudasina in July 2026. He added, in the slightly defensive tone of a man who has thought about it many times, that he is still friends with them and that they are not "mad" at him. There is no reason to doubt him, and no reason to think the gut-wrench has entirely gone away either.
The same interview contained the most direct statement he has made about why he was the one making those calls. "I wanted to be the CEO," he said. "It was my idea. I kind of just wanted to drive it." There is nothing apologetic in it and nothing grandiose either; it reads like a description of a job assignment. Then he widened the point into a philosophy. Without a "clear leader," problems arise. "You need, I think at some level, somebody that is, you know, wears the crown, no matter how heavy it is."
The crown is a strange image for a man who dresses, according to the reporters who have met him, in a gray sweatshirt, shorts, sneakers, and a black baseball cap. At Solana's Hacker House in Miami in 2022, a Fortune reporter observed that he looked more like one of the developers than like the chief executive. The authority he means is not ceremonial. It is the authority of the person who decides which engineers go so the runway lasts a few months longer. In a decentralized industry that prefers to pretend nobody is in charge, he has been unusually willing to say that, at the company level, somebody has to be.
The Series A closed in 2019 with $20 million raised in private token sales. That same year the team ran Tour de SOL, an incentivized testnet that paid outside validators to try to break the network. Then, on March 16, 2020, Solana's mainnet produced its first block.
The date deserves attention. Mainnet beta went live during the same week that COVID-19 sent global markets into free fall. A network whose entire thesis was the precise measurement of time was born into one of the most disordered weeks in modern financial history. Yakovenko has listed that crash, alongside FTX, among the near-death experiences the network survived. Most startups get one existential crisis in their early years. Solana got a pandemic in its first week.
SOL, the network's token, had been sold to early investors for twenty-two cents.
Summer, and the Man Who Moved Fast
What happened next has a name in crypto: Solana Summer. The phrase belongs to 2021. When the financier Anthony Scaramucci hosted Yakovenko on a SALT Talks episode that year, the moderator introduced it with the cheerfulness of a man riding a market: "This summer, it's Solana's summer."
Speed sold. Developers who had been priced out of Ethereum by gas fees arrived on a chain where transactions cost a fraction of a cent. Decentralized exchanges appeared. NFT collections with names like Degenerate Apes found buyers. A space game called Star Atlas promised a universe built on-chain. By the fall, SOL was trading above $190 one week and around $245 the next. In November 2021 it touched $260, more than a thousand times its original sale price.
Into this summer walked Sam Bankman-Fried. The founder of the FTX exchange and the trading firm Alameda Research became Solana's most visible institutional patron. FTX built Serum, a decentralized exchange, on Solana. FTX launched a Solana NFT marketplace. Bankman-Fried's firms held enormous quantities of SOL, and his public enthusiasm functioned as a kind of credit rating. In 2021, asked about the relationship, Yakovenko gave an answer that has aged like milk: "Sam Bankman-Fried moves fast and that's why I think we work well together."
It is easy, in hindsight, to make that sentence carry more than it should. It was a compliment about speed from a man obsessed with speed, given to the industry's most celebrated young operator at a moment when nearly every sophisticated investor in the world held the same opinion. But the line reads differently now. Speed was the single variable Yakovenko had spent his life optimizing. He recognized it in Bankman-Fried and mistook it, as almost everyone did, for the whole of what mattered.
The serious money came that year too. Across six funding rounds, Solana Labs would raise $374 million, including a $314 million round in 2021 led by Andreessen Horowitz. Ali Yahya, the general partner there, became the person who described Yakovenko most precisely: a systems engineer "through and through." Investor guides began appearing, a whole genre whose titles promised readers a place in the fast lane; Solana Rising: Investing in the Fast Lane of Crypto is representative of the mood. Wall Street journalists began calling Solana an "Ethereum killer," a phrase Yakovenko visibly disliked.
And the network kept going down. The architectural choices that made Solana fast also made it fragile in specific ways, and during the bull run it suffered a series of lengthy outages, stretches when the chain stopped producing blocks and validators had to coordinate restarts. For a network whose selling point was a clock that never stops, a stopped clock was the worst possible failure. Critics who had always considered the design reckless felt vindicated, and Wikipedia's coverage of the network came to dwell heavily on outages, hacks, lawsuits, and price crashes, enough that by 2024 Solana's partisans were publicly complaining that the encyclopedia made the project look like a list of disasters.
Yakovenko's own response to criticism tended toward the empirical. In December 2020, before the summer, he and Gokal had co-authored an op-ed for CoinDesk's year-in-review arguing that the industry had been chasing the wrong scaling problem. Throughput was not the real issue, they wrote. The real issue was decentralization, and specifically censorship resistance, "the single most critical property of blockchain networks." They measured it with the Nakamoto coefficient, the minimum number of entities that would have to collude to compromise a system, and they produced the numbers. The top three Ethereum mining pools controlled 51 percent of hash power. On Cosmos, seven of 125 active nodes controlled more than the one-third of stake needed to mount an attack. On Polkadot it was eight of 258.
It was a characteristic move. It made a claim, attached a measurement, and named the competitors. It also left Solana with a standard it would itself be judged by. That is the trouble with being the person who brings the yardstick. Everyone gets to use it on you afterward.
Ninety-Seven Percent
In November 2022, FTX collapsed. The exchange that had been crypto's most respectable face turned out to have been commingling customer deposits with Alameda's trading. Within days, Bankman-Fried's empire was in bankruptcy and Solana's most prominent patron had become its most toxic association. The token that had peaked at $260 a year earlier fell below $8. Measured from top to bottom, the drop was about 97 percent.
The price was only the visible damage. The deeper wound was to the companies building on Solana. FTX had not just been an investor. It had been a bank. Startups in the ecosystem had raised money and, in the logic of the moment, parked it at the most trusted exchange in the world. Armani Ferrante was one of them. He had built Serum's developer tooling and was now building a wallet and exchange called Backpack, and he had just raised a large round and kept it on FTX. Overnight the runway was gone.
"We were worried about ecosystem startups," Yakovenko said afterward. The sentence is mild to the point of understatement, which is how he tends to describe catastrophe. When he has discussed the episode at greater length, the interviewers have supplied the chapter titles: "FTX Holy Sh*t Meltdown," read one, and "Solana Dropping 97%."
What happened next is the part of the Solana story its partisans like to tell, and it has the advantage of being true. Teams that had kept their runways small, the posture forced on Solana Labs itself back in 2018, did not hire, did not panic, and shipped. Ferrante's team launched Mad Lads, an NFT collection that revived interest in the ecosystem at exactly the moment it seemed dead, and that some observers credit with helping keep Solana alive. Jump Crypto had already been building Firedancer, an entirely independent validator client written from scratch, which meant the network would no longer depend on a single codebase written by a single company. Help also came from an unexpected direction. Yakovenko has spoken of how Vitalik Buterin's support helped during the darkest days. Buterin is the founder of Ethereum, the chain Solana had been built to outrun.
There is a small joke buried here. Years earlier, asked to name the most humorous thing he had seen in crypto, Yakovenko had answered in two words: "Vitalik rapping." The rival he once found funny turned out to be the rival who stood next to him when it counted. It is not quite a story about friendship. It is a story about an industry small enough that its competitors understood one another's survival was linked.
In February 2023, the hosts of the Bankless podcast invited Yakovenko on for an episode with a title that sounded like a coroner's inquest: "Will Solana Make It?" His answer was not about Solana at all. "My guess is if you like looked at all the product launches from inception of crypto to now to what's gonna happen in the next like twelve to eighteen months," he said, "then the next twelve to eighteen months is probably gonna be way more than everything else combined up to this point. And if people are launching products, they're grinding for product market fit, they're gonna get users, means crypto's gonna survive. It's just like that's gonna happen."
Read the quote again and notice what it leaves out. It doesn't mention price, sentiment, or Bankman-Fried. It is a forecast stated as a conditional, and the condition is shipping. If products launch, users come. If users come, the system survives. It is the reasoning of a man who has seen platforms collapse before, once at eleven and once in college, and who knows the wreckage does not settle the outcome.
He was right, as it turned out. By October 2025, the network's market capitalization had been described as $110 billion. SOL had recovered above $130. A wave of retail speculation, much of it from people who had never heard of FTX, was about to make Solana the busiest chain in crypto. That wave was the one development its founder would describe as slop.
— Anatoly Yakovenko, All-In podcast, 2025Solana is fundamentally a protocol; it's like an email standard. It's a bunch of software, and the people that run it don't report to me, I can't fire them, so I can't stop it if I wanted to.
Communities Are Useless
The memecoins came in waves. First Bonk, then Dogwifhat, then an entire industrial apparatus: token-minting platforms like Pump.fun and LetsBONK made it possible for anyone to launch a coin in seconds, attach a picture of a dog, and watch strangers bet on it. Solana's speed and low fees made it the natural casino. By January 2025, the network accounted for a dominant share of decentralized exchange traffic, more than Ethereum and all its layer-two networks combined, and much of that volume came from memecoins. Several of the most notorious, including tokens named for the American president and first lady and one named LIBRA, drew legal scrutiny.
Yakovenko did not celebrate. On July 28, 2025, in an exchange on X with Jesse Pollak, who created Coinbase's Base network, he called memecoins and NFTs "digital slop." He compared them to loot boxes in mobile games, the randomized rewards that keep players paying with no promised return. Pollak argued that the creative tokens launched on platforms like Zora were closer to art, valuable for the culture around them regardless of price. Yakovenko pushed back on who actually pays: was it good for retail buyers to hold tokens after the price had collapsed, however "valuable" the content? Asked whether Solana's current success would exist without memecoins, he conceded the point and then reframed it. Apple's fortune, he noted, was built partly on loot-box mechanics in the App Store. Popular and valuable are different words.
A week later he made it worse. Replying to a post by Nikita Bier, X's head of product, he wrote three words: "Communities are useless." The backlash was immediate and predictable. In an industry where nearly every project begins as a social movement, where "community" is invoked the way earlier generations invoked "the people," the founder of the network driving the memecoin boom had dismissed the thing the boom ran on.
He had said a longer version two years earlier on the Unchained podcast, and the longer version is more interesting. "I would say that community is useless in some sense," he said in 2023. "The core value that we provide is the code. And how good it is. At the end of the day, we don't run validators, we don't run any of that, we just write code, and then people use it."
Strip away the provocation and what remains is an engineer drawing a line around his responsibilities. He writes the code, and the code is the product. What people build on top of it, including dog coins, loot boxes, and casinos, is theirs. He does not run the validators. He cannot fire them. This is consistent with everything else he has said about the protocol, and it is also, plainly, a way of keeping the slop at arm's length while the slop pays the bills.
The contradiction gets sharper when you look at what he actually recommends to founders. In an October 2023 interview, asked whether Solana's habit of giving new developers direct access to senior engineers came from the top, he said yes, and pushed it harder than almost any other piece of advice he has given. "Anyone that's trying to build any kind of developer facing product, have your Senior devs be in Discord talking to people," he said. "It'll take some of their time to do it, but it's really important." The reason was humility: "there's no way I know exactly what every job's needs are. Where our docs are lacking, or where community devs are blocked."
So the man who says communities are useless has also built one of the most accessible developer communities in the industry and calls it essential product development. The two positions only conflict if you think "community" means one thing. For Yakovenko it seems to mean two. There is the community of builders, who find bugs, fill in gaps in the documentation, and change the roadmap. That one he values almost above anything. Then there is the community of believers, who hold the bag and chant the ticker. He seems to regard that one as noise at best and, at worst, as people being harmed by a machine he built.
His early answers to a questionnaire make the distinction clearly. Asked what made Solana's community unique, he did not mention loyalty or culture. "We are focused on performance," he wrote. "We geek out on shaving milliseconds of confirmation times and benchmarking." That was the community he had in mind. The one that arrived was different, and he never pretended otherwise.
There is something in this that rhymes with his childhood. The Soviet state also had communities: mandatory ones, slogan-driven ones, ones whose enthusiasm was a performance required for safety. A child who grew up watching a currency fail while everyone was told to believe in it may have permanent immunity to enthusiasm as a substitute for function. He has never said this. He has said that the currency was bad.
A Clock Nobody Can Stop
In September 2023, Yakovenko wrote an op-ed for Fortune that opened the way his profiles never do, with his own biography. "I was born under Soviet rule in modern-day Ukraine," he wrote. "We moved to America when I was 11. Even at that age, I was very aware of the differences between the place I grew up–where access to assets and information was controlled with an iron fist–and my new home–where opportunity was unbound."
Access to assets and information. It is precisely the pair of things a blockchain is designed to make uncontrollable. The rest of the op-ed was policy. Incorporating a normal startup costs less than $500 on LegalZoom, he noted, while incorporating a blockchain startup in a legally compliant way costs tens of thousands in legal fees, and still with no guarantee. Solana Labs itself, he has said elsewhere, spent about ten percent of a funding round on lawyers. The American share of the world's open-source blockchain developers had fallen from 42 percent in 2018 to 29 percent in 2022, according to Electric Capital. Founders were leaving. "Imagine if Google had been founded in Russia, or Reddit had been founded in China," he wrote. "How different would the internet look today?"
Never mind that the argument was self-interested; most op-eds are. What is striking is how personal it was beneath the policy language. The man who left a country where the center controlled everything was asking his adopted country not to drive the builders out. He considered leaving, he has said, and never seriously did. America was home. "The U.S. is the best place to start a company," he said in a later interview. "It's got the best engineers, the best talent, I think the best venture capital."
His handle on X became simply @toly, with an American flag emoji beside it.
The regulatory environment changed, and his ambitions widened with it. He began to speak of "internet capital markets," which he defined as "taking all these financial services and layers in capital formation and reducing them into software and letting the fees and everything else collapse." He imagined founders conducting direct listings through open-source smart contracts, with disclosures posted on-chain and tokens functioning as cap tables. He argued for a disclosure regime that scaled with the size of the raise: "If you raise a million dollars, you should have fewer disclosures than somebody raising a hundred million." He said, on a 2025 podcast, that he wanted "to see the world where vast majority of tradable assets in the United States in the world are on chain."
On the All-In podcast he was asked whether regulated incumbents like NASDAQ would beat the upstarts. His answer reduced Solana Labs, the company he runs, to almost nothing. "We're never going to build a company that is onboarding U.S. institutions and serving US customers; we want NASDAQ to do that, and to run it on Solana." And: "if we succeed, if the protocol is awesome and globally synchronous, and super fast, NASDAQ would make more money by just running a Solana node." The protocol is like an email standard, he said. The people who run it do not report to him. "I can't fire them, so I can't stop it if I wanted to."
The last clause is the one to underline. Most founders spend their careers accumulating control. Yakovenko has spent his trying to build something that does not need him, a clock that keeps running even if its inventor wants it to stop. From the outside it can look like modesty. It is closer to a design requirement. A system that depends on the goodwill of any one person, including its founder, is exactly the kind of system that once controlled his family's money with an iron fist.
He kept building at the edges. There was a phone, the Saga, announced during the 2022 downturn. Its premise: "every phone in the world is gonna have to be Hardware wallet at one point or another." The old Qualcomm kernel developer had returned to the flip phone's descendant, this time to put private keys in the operating system and challenge the app-store duopoly. There was NFT compression, which brought the cost of minting a million NFTs down to roughly $130 to $150. There was Alpenglow, a consensus overhaul he described at Consensus 2026 as pushing confirmation times closer to the physical limits of global communication. And in January 2026, there was a post on X proposing a new formula for token launches: more than twenty percent of tokens released on day one; no tokens for teams or investors at launch; any investors unlocked "100% on the same day 1 year after." His reasoning was a clock argument. "Markets are great at dealing with information that is well known ahead of time."
At Consensus 2026, sitting across from Ali Yahya, he looked back at the industry's first decade and delivered a verdict that sounded like absolution. "All the ideas were the right ideas," he said. "They were just the timing was off." It is the summary judgment of a man whose whole career has been about timing: TDMA slots, hash chains, a mainnet born into a pandemic, a token peaking just before the bottom fell out. His insight, from the beginning, was that you cannot get the order of events right unless you can first prove when they happened.
His net worth, mostly in SOL and Solana Labs equity, has been estimated by the on-chain analytics firm Arkham at somewhere between $500 million and $1.2 billion, which means it moves with the price of a token by tens of millions of dollars in an afternoon. He once told an interviewer that his definition of success was not money at all. It was "the freedom to do whatever it is that I want," including the option "to spend the rest of my life surfing, or not." "You can really come from anywhere and get to that point," he said. He still codes in the mornings. He still gets on a bike for two hours when he needs to make a decision, and comes back, he says, "with a lot of decisions made or internalized."
The canonical repository,
solana-labs/solana, is archived now. Development moved to Anza, a separate organization. Of the network's important clients, the one built from scratch by Jump owes nothing to Yakovenko's original code. The old repo is read-only, frozen at 14,900 stars, a record that can no longer be changed. Among his pinned projects is a Haskell implementation of a restricted Boltzmann machine from some earlier enthusiasm, and a small C library called toaster for incremental failure injection, which deliberately breaks software to see what survives.At the top of his GitHub profile, where other engineers list their employers or a link to their homepage, there is a single line.
Do not trust me, formally verify all code.
Part IIThe Playbook
What follows are the operating principles that can be extracted from Yakovenko's career. They are not the principles he would necessarily list himself. He is suspicious of self-help framing and would probably reduce most of them to "make it faster." They are patterns visible in the decisions: what he built, what he refused to build, whom he hired, whom he let go, and what he said when it would have been easier to say nothing. Several of them cut against conventional startup wisdom. A few cut against conventional crypto wisdom, which is a different and often stranger thing.
The principles
- 1.Steal the answer from the field next door.
- 2.Write for the chip that doesn't exist yet.
- 3.Pick one architectural hill and refuse to come down.
- 4.Treat a long apprenticeship as an unfair advantage.
- 5.Cut deep before launch, not after.
- 6.Somebody has to wear the crown.
- 7.Put your best engineers on the front line.
- 8.Build something you cannot personally stop.
- 9.Bring a yardstick, and accept that it will be used on you.
- 10.In a crash, the only metric is shipping.
- 11.Tell the truth about what pays your bills.
- 12.Make the future boring by announcing it.
⏱
The Solana Clock
Key dates in the life of the network, from the 4 a.m. idea to the consensus overhaul.
1992
The Yakovenko family emigrates from the collapsing Soviet Union with $350 among seven people.
2003
After a failed VoIP startup, Alescere, Yakovenko joins Qualcomm in San Diego.
2016
Leaves Qualcomm after nearly thirteen years; joins Mesosphere, then Dropbox.
2017
Two coffees and a beer at Café Soleil; drafts the proof-of-history whitepaper.
2018
Core team forms in a bear market; Loom renamed Solana; Multicoin backs the raise.
2019
Tour de SOL incentivized testnet; Series A closes at $20M in private token sales.
2020
First mainnet block on March 16, during the COVID market crash.
2021
"Solana Summer"; a16z leads a $314M round; SOL peaks at $260 in November.
2022
FTX collapses; SOL falls below $8; Saga phone announced.
2025
Memecoin boom; Yakovenko calls memecoins "digital slop" and communities "useless."
2026
Token-launch framework proposed; Alpenglow consensus overhaul discussed at Consensus.
Principle 1
Steal the answer from the field next door
Proof of history was not invented from nothing. It was carried over from a neighboring field. Time Division Multiple Access had been solving the problem of many transmitters sharing one channel since the 1960s. Every cellular engineer of Yakovenko's generation knew it as basic. Almost nobody in crypto in 2017 knew it at all. The insight lived in the gap between two communities that did not talk to each other, and Yakovenko happened to belong to both.
The lesson is not "be interdisciplinary," which is the kind of advice that sounds wise and changes nothing. It is narrower and more useful: the hardest open problem in a new field is often a solved problem in an old one, wearing different vocabulary. Blockchain people called it consensus. Radio people called it scheduling. Under the jargon, both asked whose turn it was.
This only works if you have actually lived in the old field long enough to know its answers by reflex. Yakovenko did not look TDMA up. It surfaced at four in the morning because thirteen years at Qualcomm had put it there.
Tactic: Before attacking the central problem in a new domain, translate it into the vocabulary of the field you know best and ask whether that field already solved it.
Principle 2
Write for the chip that doesn't exist yet
At Qualcomm, the next chip always came, and software that could not use it had to be rewritten. Yakovenko built that assumption into Solana's foundation. "If we are limited by bandwidth right now, in two years there's going to be twice as much," he said in 2023. "These are all Moore's law problems that just kind of solve themselves." The design maximizes parallelism across cores, disks, and network cards so that each new generation of hardware makes the network faster without a redesign.
Critics saw this as a weakness: Solana validators need expensive machines, which raises the cost of participating and arguably narrows decentralization. Yakovenko's bet was that the hardware curve would close that gap faster than competitors could close the performance gap. It is a wager on time, which may be the only kind of wager he finds interesting.
The general version applies far beyond blockchains. Most systems are designed around the constraints of the year they are built. A few are designed around the direction those constraints are moving. The second kind age into their advantages.
Tactic: Identify the one input to your product that is improving on a predictable curve, and architect so that every improvement in it flows straight to your users.
Principle 3
Pick one architectural hill and refuse to come down
"Sharding is not going to succeed commercially," Yakovenko said early, when sharding was the industry's consensus answer. He built Solana as a single, monolithic chain with one global state, betting that atomic composability, the ability of any program to call any other in one transaction, mattered more than the theoretical scaling of split chains. He spent years being told he was wrong and stayed put.
The value of the stance was not that he was certainly right; the debate is still live. It was that the refusal gave the project a coherent identity. Developers knew exactly what Solana was and was not. Every subsequent decision, from access lists for parallel execution to a fee market tuned for local contention, followed from the original commitment. A company that hedges its architecture ends up hedging everything.
⚖
Conventional wisdom vs. the Yakovenko approach
Where Solana's founder diverged from the default playbook, and what it cost.
| Conventional wisdom | Yakovenko's approach | Category |
|---|---|---|
| Scale blockchains by sharding or moving to layer twos. | One monolithic chain; scale with hardware. | Architecture |
| Raise big in a bull market; hire ahead of growth. | Found in a bear; cut good engineers to extend runway. | Capital |
| Celebrate the community at all costs. | "Communities are useless"; senior devs live in Discord. | Culture |
| Accumulate founder control. | "I can't fire them, so I can't stop it if I wanted to." | Governance |
| Praise whatever activity drives usage. | Calls the memecoin boom "digital slop." | Risk |
| Vest investor tokens gradually to smooth selling. | Unlock all investors on one known day, a year out. | Capital |
Tactic: Name the single architectural bet your product depends on, state it publicly, and judge every later decision by whether it strengthens or dilutes that bet.
Principle 4
Treat a long apprenticeship as an unfair advantage
Balaji Srinivasan called it an "unusual late in career migration." Yakovenko arrived in crypto in his late thirties with nearly thirteen years at Qualcomm, a stint building distributed operating systems at Mesosphere, and a few months on compression at Dropbox. His co-founders Greg Fitzgerald and Stephen Akridge came from the same Qualcomm environment. In an industry that glorifies the young dropout, Solana was built by people who had spent a decade learning how real systems fail under real load.
He is unsentimental about why this mattered. Asked for advice to aspiring developers, he recommended going deep on systems, distributed systems, and communication networks, and added that "it's almost like it's better to work at a big company sometimes for the first five years," for the mentorship and resources. Then the self-assessment: "I was too lazy. Personally, I needed to go build." He built at a startup first, failed, and then took the apprenticeship.
The underrated advantage of a long career inside a large incumbent is not the network or the résumé. It is a library of solved problems stored in muscle memory, which is exactly what Principle 1 requires.
Tactic: If you are early in your career, choose the job that will teach you the hardest technical problems in the most demanding environment, and treat founding as something you earn the right to do later.
Principle 5
Cut deep before launch, not after
The hardest decision Yakovenko describes came before the network existed: firing "really good, really good engineers" to extend the runway ahead of launch. Founded in a 2018 bear market, Solana Labs had to assume the next raise might never come. The cut was made while there was still time for it to matter.
The FTX collapse four years later showed the downstream effect. The ecosystem teams that recovered fastest were those that had kept runway small, did not hire through the panic, and kept shipping. The discipline Solana Labs had learned under duress became, in effect, a survival trait across its whole ecosystem. Founders who had parked large raises on a single counterparty learned the opposite lesson the expensive way.
The general rule: painful cost decisions made early, while optionality remains, are survivable. The same decisions made late, after the money is gone, are often fatal.
Tactic: Model your runway assuming no further funding is ever available, and make the personnel decisions that model requires now, not when the bank balance forces you to.
Principle 6
Somebody has to wear the crown
"I wanted to be the CEO," Yakovenko said of the founding. "It was my idea. I kind of just wanted to drive it." In an industry with a near-religious commitment to the absence of leaders, he has been explicit that a company, as distinct from a protocol, needs one. "You need, I think at some level, somebody that is, you know, wears the crown, no matter how heavy it is."
Notice the distinction he keeps drawing. The protocol should not depend on anyone; he cannot stop it and does not want to. The company that writes the code needs a single person accountable for hard calls, like which engineers go so the runway lasts. Decentralized outcomes, in his model, are built by centralized teams. Confusing the two produces either authoritarian protocols or leaderless companies, and both fail.
The crown metaphor also concedes something about the job: it is heavy. He does not pretend leadership is pleasant or that the firings did not hurt. He just maintains that someone has to carry the weight, and that the person whose idea it was is the obvious candidate.
Tactic: Even in a flat or decentralized organization, name in writing who makes the final call when the team disagrees, and make sure that person accepts the weight explicitly.
Principle 7
Put your best engineers on the front line
This is the advice Yakovenko gives most emphatically: "Anyone that's trying to build any kind of developer facing product, have your Senior devs be in Discord talking to people." The rationale is humility at the level of information. "There's no way I know exactly what every job's needs are. Where our docs are lacking, or where community devs are blocked."
The practice makes the feedback loop extremely short. A senior engineer who sees a developer stuck on a bug at 11 p.m. on a Tuesday changes the roadmap by Wednesday, with no product manager, ticket system, or quarterly survey in between. "They organically build a better project," he said, "just from that interaction." He acknowledges it breaks at scale, when the squeaky wheels drown out the real problems and a dedicated developer-relations function becomes necessary. In the early days, he insists, nothing replaces it.
This is the same man who called communities useless, and the two positions are compatible. He values the community of builders as a sensor network. He distrusts the community of cheerleaders as a substitute for one.
Tactic: For the first eighteen months of any developer-facing product, require your most senior engineers to spend a fixed block of time each week answering users directly in public channels.
Principle 8
Build something you cannot personally stop
"The people that run it don't report to me, I can't fire them, so I can't stop it if I wanted to." That sentence from the All-In podcast states the deepest design goal in Yakovenko's work. Solana is meant to function like an email standard, a protocol others adopt and operate, not a service he controls. He pictures NASDAQ running a Solana node and Solana Labs never onboarding a single U.S. institution itself.
The archived
solana-labs/solana repository makes the point concrete. Core development moved to Anza, and Jump built Firedancer, an independent client, from scratch. The network's survival no longer depends on the founder's company or the founder's code. For a founder this is either self-erasure or the highest form of success, depending on what you think you were building.There is a reason this matters strategically and not just philosophically. Institutions, he argued at Consensus 2026, do not fear decentralization. They fear "the ambiguity that they don't understand the system that they're operating in." A system nobody can arbitrarily stop is a system whose rules are knowable. Credible neutrality is a product feature.
Tactic: Identify the single point of control in your product that you personally hold, and ask whether giving it up would make the product more trustworthy to the customers you most want.
Principle 9
Bring a yardstick, and accept that it will be used on you
In December 2020, Yakovenko and Raj Gokal argued that the industry was measuring decentralization wrong. Node counts and developer headcounts were vanity metrics, they wrote, and the meaningful number was the Nakamoto coefficient: how few entities would have to collude to compromise the system. They then published the coefficient for competitors. On Ethereum, three mining pools held 51 percent of hash power. On Cosmos, seven of 125 nodes held more than a third of stake. On Polkadot, eight of 258.
The move did two things. It reframed the debate onto ground where Solana's thesis looked strong, and it handed critics a precise instrument to measure Solana itself. Every outage, every concentration of stake, every high-end hardware requirement could now be scored against the founders' own standard. That is the price of bringing the yardstick, and it is worth paying. A company that defines the metric shapes the conversation, even when the metric is used against it.
Tactic: When your category's dominant metric flatters your competitors, propose a better one publicly, with numbers, and commit to reporting your own performance against it.
Principle 10
In a crash, the only metric is shipping
When Bankless asked in February 2023 whether Solana would survive, Yakovenko declined to talk about price, sentiment, or Bankman-Fried. He talked about product launches. "If people are launching products, they're grinding for product market fit, they're gonna get users, means crypto's gonna survive. It's just like that's gonna happen."
This is the logic of a man who has lived through collapses as a child and as a student. Price is downstream. Sentiment is downstream. What determines whether a platform survives a crash is whether people keep building on it while it is unpopular. The ecosystem's recovery after FTX, with Mad Lads, Firedancer, and the teams that shipped without hiring, came about because enough builders believed the same thing at the same time.
At Consensus 2026 he generalized the point backward: "All the ideas were the right ideas. They were just the timing was off." The implication for founders is that a crash does not refute an idea. It often only refutes the timing, and whoever is still shipping when the timing turns gets the reward.
Tactic: During a market downturn, stop reporting price-linked metrics to your team and track only one number: features and products shipped to real users each month.
Principle 11
Tell the truth about what pays your bills
Memecoins drove much of Solana's resurgence, and Yakovenko called them "digital slop." He compared them to loot boxes. He questioned whether retail buyers were well served by holding tokens after the price collapsed. Asked whether Solana's success depended on them, he admitted it did and compared the situation to Apple's revenue from loot-box mechanics in the App Store.
The commercially rational move would have been silence or cheerleading. Founders almost never criticize the activity that fills their network. Yakovenko's candor cost him goodwill and drew accusations of elitism. It also preserved something harder to rebuild once lost: the credibility of a founder who will describe his own ecosystem accurately. When he says something is valuable, people have reason to believe him, because he has shown he will say when something is not.
There is a limit, and it should be named. Candor about the slop does not remove it from the network or the revenue from the balance sheet. Calling it slop while it pays the bills is honest; it is not a solution. That tension does not resolve, and he does not seem to need it to.
Tactic: Write down plainly which of your revenue sources you would not be proud to explain to a skeptical friend, and say so internally even if you cannot yet change it.
Principle 12
Make the future boring by announcing it
In January 2026, Yakovenko proposed a token-launch formula that goes against standard practice. Release more than twenty percent of tokens on day one. Give teams and investors nothing at launch. Distribute through airdrops to real users or fair auctions. If there must be investors, unlock all of them on the same day, one year later. "Markets are great at dealing with information that is well known ahead of time," he wrote. "The 1 year wait will force the secondary market to match investors that want to cash out with those who want to double down, and the primary market will anchor the price."
The insight underneath is the same one as proof of history: when everyone agrees in advance on when things happen, the cost of coordination collapses. Gradual, opaque vesting schedules create a constant low-grade uncertainty that markets price as risk. A single announced date turns that uncertainty into a calendar entry. It may sell off, but nobody is surprised. He ended the proposal by noting that product-market fit matters more than any tokenomics. The design only removes avoidable noise.
Tactic: For any event that will affect your stakeholders, from layoffs to unlocks to price changes, announce the exact date far in advance rather than letting it arrive gradually or by surprise.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Anatoly Yakovenko, CryptoMurmur interviewI am obsessed with making programs faster and optimizing complex problems. Most success comes from how much time you spend on something and not how smart you are.
— Anatoly Yakovenko, CryptoMurmur interviewDon't listen to old people. The world is going to change and whatever advice they will give you is already stale.
— Anatoly Yakovenko, Fortune, September 18, 2023Even at that age, I was very aware of the differences between the place I grew up–where access to assets and information was controlled with an iron fist–and my new home–where opportunity was unbound.
— Anatoly Yakovenko, Unchained podcast, 2023I would say that community is useless in some sense. The core value that we provide is the code. And how good it is.
— Anatoly Yakovenko, Consensus 2026All the ideas were the right ideas. They were just the timing was off.
Maxims
- Borrow before you invent. The hardest problem in a new field is often an old field's settled answer, waiting for someone fluent in both vocabularies.
- Design for the next chip. Software that does not improve automatically with better hardware will have to be rewritten, so build to ride the curve.
- Time is a data structure. When participants can prove when events happened, they no longer need to negotiate the order.
- One hill, defended. A clear, contrarian architectural bet held through ridicule gives a product an identity that hedging never will.
- Cut while it still counts. Painful cost decisions made before launch buy survival; the same decisions made after the money is gone buy nothing.
- Decentralized outcomes need centralized accountability. The protocol should need no one, but the team that builds it needs someone wearing the crown.
- Your best engineers belong in the chat room. Direct contact between senior builders and users is the fastest product-development loop ever invented.
- Build what you cannot stop. A system that does not depend on its founder's goodwill is the only kind a stranger can truly trust.
- In a crash, count shipments, not prices. Platforms survive downturns because builders keep launching, not because sentiment returns.
- Announced futures are cheap futures. Markets punish surprise far more than bad news, so put the hard dates on the calendar early.
