Hyperliquid’s Jeff Yan in Focus: An 11-Person Team, Zero VC Funding, $900M Annual Profit—What’s His Secret?
TechFlow Selected TechFlow Selected
Hyperliquid’s Jeff Yan in Focus: An 11-Person Team, Zero VC Funding, $900M Annual Profit—What’s His Secret?
Why a Physics Olympiad gold medalist gave up financial freedom to rebuild the entire financial system.
Navigating Web3 tides with focused insightsAuthor: Dom Cooke
Translated and edited by: TechFlow
TechFlow Intro: Colossus magazine conducted an in-depth visit to Hyperliquid’s Singapore office, producing the most comprehensive profile of founder Jeff Yan to date. Eleven people. Zero VC funding. Over $900 million in annual profit—how did this three-year-old protocol capture 37% of the decentralized perpetuals market in just two years?
From Yan running quant strategies on a TV screen in Puerto Rico, to rejecting a $1 billion valuation VC check and airdropping $16 billion worth of tokens to users, to today’s independent teams deploying crude oil, gold, and S&P 500 perpetuals on Hyperliquid—the piece unpacks one central question: Why would an International Physics Olympiad gold medalist abandon financial freedom to rebuild the entire financial system?
Full Text:
On a Friday morning in January this year, a 43-year-old man was taken from his home in Saint-Léger-sous-Cholet, western France. He was driven 30 miles to Basse-Goulaine, beaten, bound, and dumped roadside. Twelve hours later, in Verneuil-sur-Seine, a suburb of Paris, three armed men kicked down a family’s front door, assaulted the couple in front of their children, zip-tied all four family members, ransacked the house, and left by train.
This was the 70th such attack globally in less than a year.
Two days later, I boarded a flight to Singapore.
Bodyguards and Plush Cats in the Office
I came here to visit an 11-person team—but the first person I met upon arriving at the office wasn’t part of that team. He was a burly American with a crew cut and stubble, sitting behind a small table in the lounge area, facing an Apple laptop. His build clearly signaled he wasn’t there to write code.
He was a bodyguard.
One of Hyperliquid’s co-founders walked me from the hotel to the office. Her handle is iliensinc—an abbreviation of Aliens Incorporated. As we passed under rain trees lining the street, she told me they hadn’t always been based in this part of Singapore. The company originally occupied a shared workspace in the financial district, but her co-founder—the only team member who didn’t use a pseudonym—began drawing attention. At first, it was just curious glances from passersby trying to place his face. Then strangers started approaching him for conversation. Later, people followed him into his apartment elevator. So the company relocated—to a quieter building, one nobody would think to look for them in.
Even their cleaning lady didn’t know what they did. In her mind, she cleaned for a plush cat toy merchandising company. There were 34 plush toys scattered around the office—making the misconception entirely reasonable. The company mascot is a cat named Hypurr; 12 of them perch on shelves, but there are also sharks, lizards, koalas, penguins, and dragons—several lounging atop Dell monitors like furry gargoyle statues. Most were brought in by an engineer whose wife had banned him from bringing more home, so he brought them to work instead. The team never corrected the cleaning lady’s assumption.
Because Hyperliquid is one of the highest-profit-per-employee companies on Earth. Last year, its 11 employees generated over $900 million in profit. It’s only three years old, valued at $10 billion, and has never accepted a single dollar of venture capital. Its central figure, Jeffrey Yan, is 31—and in an industry where “the bigger your success, the higher your kidnapping risk,” he’s become one of its most recognizable faces, almost against his will.
The TV-Screen Trader in Puerto Rico
Before Hyperliquid, Yan lived in Puerto Rico, operating one of crypto’s largest anonymous trading operations almost single-handedly: Chameleon Trading—“Chameleon” was his gaming ID from high school. He launched it with $10,000 of his own savings, growing at several thousand percent per year for two and a half years. After telling me the returns, he immediately tried to downplay them. I noted his humility. I also noted that Chameleon made him very rich. At 27, he achieved financial freedom. To surfers, bartenders, and waitstaff in San Juan, he was just another young guy in board shorts.
Now he sits cross-legged on a gray armchair in a Singapore office guarded by a bodyguard—barefoot, wearing black shorts and a navy T-shirt—explaining why the entire financial system needs to be torn down and rebuilt. What I wanted to know was: Why did he trade his first life for this second one?
Not for money, he said. Yan didn’t come from wealth, nor does his lifestyle suggest any interest in the trappings of affluence. He wears the same Lululemon shorts and T-shirts every day—15 pairs of shorts and 10 T-shirts, each in just three colors. There’s no sign of wealth in the office. The furniture is leftover from the previous tenant. The only things the team added were two board games in the lounge, NFTs on the wall, and those plush cats. I found four books on a shelf and recognized one as Frank Slootman’s Amp It Up, a management book whose core thesis is “most people don’t try hard enough.” I mentioned it to iliensinc. She shrugged. That book wasn’t theirs—but the philosophy was. In the kitchen, three unopened bottles of Grey Goose and Macallan weren’t theirs either—they’d been left over from a community event two years ago that failed to meet its minimum spend requirement. This team drinks tea.
Not out of love for crypto either. Bitcoin had fallen roughly 30% from its early-October peak—ironically, the asset meant to replace gold dropped 30%, while gold rose 7% over the same period. Most tokens fared even worse. I asked Yan how he viewed the pervasive pessimism in the industry, and he offered no defense. “There really is a lot of questionable behavior in this industry,” he said. “People are starting to realize many things aren’t what they were advertised to be—that may actually be a good thing.” He doesn’t consider Hyperliquid a crypto company. “Nobody calls us an internet company anymore,” he told me. “We use crypto technology, but that doesn’t define us.”
Of the 11 team members, only two—including Yan—had prior crypto experience. That was intentional. Early crypto circles, in Yan’s words, were mostly filled with people seeking quick riches. He was building something long-term—a pursuit better aligned with technical thinkers than traders. But it was also a supply issue. Hyperliquid hires from the podiums of international math and science olympiads. Yan won a physics gold medal at 18. One engineer earned a silver in informatics; another trained with the U.S. national team. Yan wants to hire more, and he added two people after my visit—but candidates at that level willing to work in crypto are scarce. Many have been burned by scams and empty promises over the years, and more recently, diverted by AI.
So what, then, is Yan—someone who already earned enough to do anything—doing here?
There’s Always Someone Above You
The answer, at least to outsiders, is becoming increasingly clear.
Hyperliquid is a blockchain, with its own exchange built directly on top. In traditional exchanges, companies hold your funds and control the infrastructure. On Hyperliquid, you hold your own funds, and the platform is open-source. Yan’s vision for it—stated without irony—is to host all finance. Whether that’s ambition or absurdity depends on whether you’re looking at the plush cats or the platform’s data. Because in the months following my visit, markets that had used the same trading methods for over a century began shifting—in small but measurable ways.
Hyperliquid launched in 2023 with perpetual contracts. Perpetuals are derivatives—and the largest single market in crypto. They’re essentially bets on the price of an asset you never own. Unlike traditional futures, they never expire. This market is six to eight times larger than spot trading—roughly $7 trillion per month. Until recently, nearly all volume ran on centralized exchanges. Binance was the largest. No decentralized platform could challenge it. Hyperliquid was the first—and grew to about 14% of Binance’s market share.
Then, in October 2025, Hyperliquid did something centralized exchanges couldn’t: Anyone could launch a new perpetual market on the platform for any asset with a price feed. An independent team called Trade[XYZ] became the most active deployer. They started with silver. By January, their 24-hour volume reached roughly 2% of CME’s (the world’s largest derivatives exchange, founded in 1898). Then Trade[XYZ] launched crude oil. Oil trades on markets that close weekends. On a Saturday late February, the U.S. and Israel began bombing Iran. CME was closed. Hyperliquid wasn’t. Crude oil’s daily volume surged from $21 million to $3.7 billion. A month later, Trade[XYZ] launched S&P 500 perpetuals—officially licensed by S&P Dow Jones Indices—and traded them 24/7, including weekends.
The most influential product on Hyperliquid is now built by people outside Yan’s team—and will remain so.
Trade[XYZ]’s founder requested anonymity. He bought his first Bitcoin for $66 in 2013 and had remained an investor—not a builder—never intending to start a company. He told me he’d likely have left crypto altogether if not for Yan. “Hyperliquid has a chance to save crypto,” he said.
But none of this explains why Hyperliquid became what Yan describes as—within an industry that “looks like it’s about to succeed until suddenly it isn’t”—nor why he abandoned life in Puerto Rico to prove it. These questions followed me on my first afternoon in the office, as iliensinc and I chatted in the lounge, a plush cat on the table, the air still carrying the scent of ginger and sesame from lunch. She told me that three years earlier, when Yan announced Chameleon’s end, the team had asked him the same question. Her answer didn’t begin with crypto—it began with who Yan is. “You should ask him about his mother,” she said.
There’s Always Someone Above You—and Always Something Beyond
Yan likes holding meetings outdoors. We sat on a covered patio—four gray lounge chairs, a coffee table. Cars passed below. Every few minutes, a gardener fired up a lawnmower. Crosswalk chimes buzzed intermittently.
Yan tucked his feet beneath him. When I asked about his mother, he paused. She had a catchphrase, he said—a Chinese idiom: “There’s always someone above you—and always something beyond.” She wasn’t the kind of mother who pushed her children, but she wanted him to understand that however impressive he thought he was, he’d only seen a tiny fraction of the outside world.
She raised him and his sister alone in Redwood Shores—the most profitable geographic corridor in U.S. business history—sandwiched between San Francisco and Palo Alto. Oracle’s mirrored headquarters loomed over the neighborhood. Neighbors were engineers and product managers whose kids were already preparing for the kind of life Yan would later build. His parents were Chinese immigrants who divorced when he was in third grade. His father left. His mother, an accountant, worked overtime every tax season—he could feel it. “I could tell other families were richer than ours,” he said, “but I never resented it. Going out to play didn’t cost much.”
His school had no culture of academic competition. Despite that idiom, his mother never pushed him. Before he became a teenager, no one pushed him at all. He played outside, went to school, returned home, kept playing. By ZIP code standards, he was the rarest kind of kid: one left entirely to his own devices.
Photo: Yan and his dog Max in Redwood Shores
In eighth grade, a friend who’d just transferred from private school dragged him to a math competition. The friend needed a partner. Yan had never seen anything like it. School math bore no resemblance. No formulas to memorize, no rote calculations. Just a problem—sometimes a single sentence—and you had to find your own way in. The answer wasn’t a number, but a proof—a complete logical argument explaining why something must be true. Finally, they ranked participants like sprinters. For Yan, it fused the best parts of sport and understanding the world.
That summer, he woke at 5 a.m. daily, downloaded past contest papers online, and solved them alone in his room. No tutor, no expensive summer program, no one demanding he do it. “Turns out I’m extremely competitive,” he said. “There was a competition I’d never heard of before—kids had been training their whole lives, and I was way behind.”
A year in, he made the U.S. Math Olympiad training camp in ninth grade—the top 50 high schoolers nationwide. He was among the youngest in the room. He didn’t make the national team. He said he didn’t care. For three weeks, he sat with teenagers who could stare at three sentences for five hours and extract truths invisible to most. “There’s no Federer in math,” Yan told me, “but at the highest level, there’s something like what Federer possesses. There’s a style to the work, an elegance to constructing proofs—I saw it up close for the first time at that camp. ‘It’s like playing football with Tom Brady,’ he said, ‘but nerd version. Most people never get that feeling.’”
The next year, he was eliminated in an intermediate selection round for math. He was 16—forced to wait a full year before trying again. I asked if that was his first real failure. “Losing is common,” he said. “Most people are losers. Usually only one person wins.”
The problem wasn’t losing—it was emptiness. “There was a hollow feeling,” he said. “I should’ve been learning something.” So he picked up advanced physics textbooks. His school didn’t offer physics until tenth grade, but he’d just learned calculus—and for the first time grasped what it was truly for. He discovered Feynman Lectures. “I watched them like binge-watching a show.” Within a year—entirely self-taught—he ranked among the top five young physicists in the U.S.
He made the U.S. Physics Olympiad team, traveled to Estonia (his first trip to Europe), and won silver. The next summer in Copenhagen, he took gold—ranked 24th worldwide. At 18, back in the Bay Area, he understood his mother’s words about “something beyond” were right. Above him stood exactly 23 people.
Harvard and Hudson River Trading
Harvard covered nearly all his tuition. In his freshman spring semester, Yan enrolled in CS124—Data Structures and Algorithms. Mostly upperclassmen took it—and it was notorious for being brutal. Students called it a “necessary evil” in course evaluations. One warning read: “No social life. You’ll have no girlfriend.” Of 150 students, Yan—the freshman—topped the class, decisively.
At Harvard, freshmen are assigned to upperclass dorms after their first year. Yan drew Pforzheimer House, where he befriended Scott Wu. Wu was two years younger; Yan first met him at an Olympiad-kids summer program. Wu won three consecutive gold medals at the International Olympiad in Informatics (IOI)—his last perfect score—and later co-founded Cognition AI. When Wu, a sophomore, was also assigned to Pforzheimer, he messaged Yan: “Yo, I’m in Pfoho.” Yan replied: “Let’s go!”
Wu often found Yan teaching himself jazz at the upright piano in the common room, repeating phrases until flawless. They played chess, Go, poker—and spent hours debating what “being the best at something” truly means. Yan talked about Faker—the greatest League of Legends player ever—as well as elite Go players and top HFT traders. “He was always thinking about what makes someone exceptional,” Wu told me. “What is the essence of this field? What does true mastery mean?”
Wu recalled Yan’s unusual contrarian thinking. Most Harvard students, exposed to identical information in the same environment, arrive at similar conclusions. Yan never did. Wu also said he was deeply funny. “A master deadpan comedian. He’d say something completely unexpected—but deliver it in the flattest tone imaginable.”
Yan worked every summer. He interned at Google X, building tools for autonomous vehicles before Waymo existed. He interned at Tower Research Capital, a trading firm. In his senior year, he freelanced at Nuro (another autonomous vehicle company)—mainly because he felt college was stretching into a fifth year.
That winter, junior year, he and Wu joined Hudson River Trading’s inaugural internship cohort—10 interns total. HRT is one of the world’s most successful quant firms. Fellow interns included Alexandr Wang and Jesse Zhang—later founders of Scale AI and Decagon. The internship was a three-week competition. Each round, Wu and Yan swept the top two spots.
After graduating with a bachelor’s in math and master’s in computer science, Yan joined HRT full-time in late 2017, placed on the U.S. equities algo team. He met weekly with his manager. This manager had trained many newcomers. Usually, these meetings followed a rhythm: newcomers hit walls in code, they solved them together, newcomers hit the next wall. Yan never hit walls, his manager recalled. He arrived with ideas. Meetings were efficient—but something felt off. It took a while to pinpoint it. Yan did everything right—but it seemed to mean nothing to him. Eight months in, when Yan announced he was leaving, his manager understood. His departure email—by HRT standards—was unusually warm.
Yan liked HRT. He considered trading the purest game possible in real life—you were either right or wrong, and the market told you which. The world’s smartest people competed with you, and in that brutal contest, they produced something incredibly valuable for the world: liquid, efficient markets. But spending eight months improving an already excellent system, at a company that would thrive without him—meant he couldn’t answer the question haunting him: What value did he add to the world?
Crypto’s Christmas
In December 2017, the answer found him. Bitcoin neared $20,000. Coinbase was the top-downloaded app in the U.S. Billions poured into ICOs like “Jesus Coin.” Crypto’s Christmas. Yan first heard of Bitcoin during his HRT internship, when two former partners pitched it to interns. Nobody was impressed. But while still at HRT, he discovered Ethereum’s Yellow Paper—a description of a computer whose outputs everyone agreed on, impossible for any individual to shut down. Working in finance daily, he saw what finance ran on. That paper described replacing trust with code. “I thought: I can build something that transforms finance entirely.”
By April 2018, he left HRT to build a prediction market—where users could bet on weather, elections, or sports. Anything with an outcome. It ran on blockchain, with no single entity controlling funds. Its architecture centered on an idea Yan and his co-founder believed they pioneered: off-chain matching, on-chain settlement—because Ethereum was too slow for a real exchange. Funds lived in smart contracts governed by code, but users experienced fast, seamless trading. Crypto’s decentralization promise—without friction. He and his college roommate Brian Wong (also ex-HRT) built it in Binance Labs’ first San Francisco incubator cohort—named Deaux.
Kalshi launched in 2019 with the same concept. Polymarket followed in 2020. Today, Kalshi and Polymarket combined are valued at over $40 billion.
Deaux reached 100 users.
As Yan recounted this, Singapore’s sky opened. Heavy, monsoon-style rain—capable of flooding gutters in minutes. From the patio, we heard rain hitting streets below, tires hissing through puddles, amplifying traffic noise.
“That thing had zero chance,” he continued. When Deaux launched, Bitcoin had fallen over 80% from its peak. “Jesus Coin” was dead—and stayed dead. Nobody cared about betting on tomorrow’s weather. More critically, Yan and Wong barely considered regulation. Kalshi later spent three years fighting regulators before launching.
When Deaux shut down, Scott Wu was the only person on Earth who mourned it. He was one of five regular users.
Yan refunded most of the $450,000 raised. He was still under HRT’s non-compete, so he headed to Lake Tahoe with a friend also under restriction—skiing until the snow melted. Then he backpacked China, Japan, and Peru. He tried convincing me travel had techniques. He had none.
By late 2019, his non-compete expired. Yan moved to Puerto Rico—where capital gains taxes legally drop near zero. He arrived with $10,000 and a strong intuition that something big was coming.
His girlfriend joined him in Puerto Rico. They shared a one-bedroom apartment renting for under $2,000/month by the sea—but “shared” implied some cohabitation, and Yan reserved no time for it. Without a monitor, he commandeered their TV, setting up camp in the living room. For the first year or so, she got roughly 30 minutes of his attention daily. The rest belonged to trading algorithms scrolling across the screen.
Yan worked at least 14 hours daily—easily 100+ hours weekly. Starting with Python scripts, he coded connections to crypto exchanges, letting programs trade for him 24/7. He monitored them, tweaked logic, tracked data—and scrapped everything when unsatisfied.
He could do this because crypto’s openness surpassed anything traditional finance had ever offered. At HRT, executing a U.S. equity order required connecting to 13 exchanges across three New Jersey data centers, complying with SEC’s Reg NMS rules, sourcing CME futures data via Chicago microwave links—and millions in infrastructure costs. In crypto, everyone—HRT employee or TV-screen trader—used the same HTTP infrastructure built for websites. All you needed was an AWS server.
For nearly two years, his girlfriend had no idea what happened on that TV. Their life didn’t change. Same rent, same food. She knew he was passionate and driven, assumed he was doing well—but had no material evidence. Then, one Friday night in summer 2021, she tried pulling him out for dinner—booked a week earlier. He refused.
“You don’t understand,” he told her. “If I don’t fix this bug now, I’ll lose $100,000.”
Chameleon Trading—and Not Taking a Single VC Dollar
After that night, Yan decided to formalize it as a company. He needed someone who handled everything besides coding. At Harvard’s Pforzheimer House, one person seemed to manage every aspect of life flawlessly—a skill utterly foreign to him. Last he’d heard, iliensinc was in Asia, serving as chief of staff at a VC fund—flying between Tokyo, Seoul, and Hong Kong.
When he contacted her, she was in San Francisco. COVID had halted travel, turning her Asia-hopping job into midnight Zoom calls from her apartment. Yan explained what he needed. He offered no job description, title, or concrete duties. But after three years of vetting founders, she sensed he wasn’t someone to short.
The company got its official name: Chameleon Trading. iliensinc began joining Zoom calls with exchange business development teams—adding professional polish to what was essentially one man’s operation on a San Juan beach. Beneath giants like Jump Trading, Tower, HRT, and Jane Street existed a layer of anonymous trading firms—size unverifiable. Chameleon was among the largest.
By 2022, Yan grew restless. Four years into crypto, having plugged into every market—centralized and decentralized—he began caring about the industry itself—not just his P&L. Bitcoin gave the world a way to hold and transfer value without intermediaries. Ethereum gave the world a computer no one could shut down. Between them, they’d laid nearly all groundwork to rebuild finance. Yet the industry had built almost nothing. Binance and Coinbase—the two largest exchanges—remained centralized. Crypto kept re-introducing what it was meant to eliminate.
That summer, iliensinc arranged a team retreat at a countryside UK hotel. She’d grown Chameleon to six people. Yan gave her a Bitcoin budget. The team flew to London, visited the British Museum, stayed at a country estate. Their leader—leaving screens for the first time (at least in anyone’s memory)—was visibly uncomfortable.
Back in Puerto Rico, trading resumed. But Yan told the team they’d begin building something new. He wasn’t sure what. He had ideas—none convinced him. He only knew Satoshi’s original Bitcoin vision was being quietly buried by the industry he’d helped build—and this disturbed him more than it should have for someone who’d made millions from that industry’s failures.
To his team, Yan had breathed too much fresh air.
In November 2022, FTX—the world’s third-largest crypto exchange, valued at $32 billion—collapsed in nine days. It had been lending user deposits to Alameda Research, a trading firm run by its founder’s girlfriend. When users demanded withdrawals, the money was gone. Less than six months earlier, Terra—a $50 billion crypto ecosystem—vanished in three days. It attempted a dollar-pegged stablecoin sustained solely by internal logic. The algorithm meant to preserve the peg accelerated its collapse. The industry’s two largest projects died within half an orbit of the sun.
Yan had seen enough. He told his six-person team trading was over. They could disagree—but Chameleon was done. If he was wrong, they could return to trading anytime. Several disagreed; several left. But Yan’s decision stood. No investors to consult, no board to convince—it was his money, his call, and he’d found a new mission.
“I was overconfident that FTX would be the death knell for centralized exchanges,” Yan told me, “but that belief was useful—it gave me the resolve to attack this massive market.”
Building a Blockchain from Scratch
The market he meant was perpetuals. Born from economist Robert Shiller’s insight in the 1990s: Traditional futures expire. At expiry, traders either take delivery of the underlying asset (oil, wheat, pork bellies) or close and reopen positions—paying fees each time. Shiller asked an obvious question: If almost no one trading pork belly futures actually wants pork bellies, why force expiration?
Traditional markets already had working solutions—no reason to change. In 2016, BitMEX—a crypto exchange—saw the reason. Since then, perpetuals became crypto’s dominant trading format. Contracts never expire. Traders use high leverage—often 10x or 20x their capital. Fees and liquidations made centralized crypto exchanges among the industry’s most profitable companies.
By end-2022, no usable decentralized version existed. The reason was foundational tech. Most modern markets run order books: buyers bid prices, sellers ask prices, matches execute. More participants mean tighter spreads. From NYSE to Binance, that’s the model. But order books don’t just process trades—they must track traders constantly updating prices, often adjusting dozens of times before execution. Existing blockchains fail at this. Too slow, too expensive, too clunky. Every update costs money and requires confirmation. Running an order book on them is like using dial-up to run the NYSE.
By end-2022, Yan and his team reviewed every blockchain other projects used—none met their needs. So they built their own. Three months later, Hyperliquid had a custom blockchain robust enough to host an exchange. Yan spent the next half-year defending Hyperliquid’s offering on Twitter, explaining why it surpassed existing industry solutions.
The exchange’s dilemma: It’s useless when empty. A buyer entering an empty market finds no sellers. Traditionally, you pay market makers—guaranteeing counterparties regardless of traffic. You compensate them with cash, equity, or token shares. Several market makers approached Hyperliquid. One told iliensinc outright: “My firm makes kings. If you don’t pay us, you’ll never succeed.”
They refused. They paid no one. Hyperliquid launched end-February 2023. Through March–April, its user base consisted mainly of NFT collectors who’d never traded perpetuals—placing $10 orders, learning leverage via simulated contests. No serious users.
Then, in May, Yan deployed Chameleon’s most successful trading strategies—those that made it crypto’s top anonymous operation—into an on-chain vault called HLP (Hyperliquidity Provider). You could deposit $10 or $10 million. No management fee, no performance fee. The vault ran automated strategies—every dollar of profit flowed to depositors. All records lived on-chain. Deposit $10, and watch it grow in real time. Had FTX been built this way, Alameda’s hole would’ve been visible to the world.
HLP solved two problems at once: The exchange gained liquidity. Depositors gained something traditional finance never offered. An early Hyperliquid user described it as “the first time in history an ordinary person could invest in a high-frequency trading strategy for free.”
“I’d happily pay Jeff 2% management fee plus 50% performance fee for access,” they told me. “Instead, a nameless nobody, sitting anywhere on Earth, accesses one of crypto’s best market-making strategies. People still don’t grasp how extraordinary this is.”
But few understood at the time. By fall, crypto prices surged daily, yet depositors watched HLP balances decline while Bitcoin climbed. The algorithm did its job—trades were profitable—but because everything ran on-chain, it couldn’t hedge exposure to the broader market. Traditional market makers offset this risk elsewhere. HLP couldn’t, by design. So while winning trade after trade, it was effectively short a relentlessly rising market. People grew angry. Other projects attacked Hyperliquid on Twitter and Discord; Yan fought back. Back then, it still mattered to him.
But HLP was never the final answer. Yan built it to bootstrap liquidity until independent market makers arrived—and he saw the opportunity was obvious to them. Demand exceeded supply; wide spreads meant easy profits for anyone willing to quote. He wrote documentation, posted long Twitter threads explaining market making, and guided market makers step-by-step through integration. Most hesitated. Every other exchange paid them. Yan refused—and HLP couldn’t scale to fill the gap. “Alameda was central to FTX’s operation,” he said. “We didn’t want HLP to be central to Hyperliquid’s.”
Data rose, complaints rose. Market makers should theoretically appear any moment. But if they didn’t—and users left first—the project failed.
Yet there’s one group you can always count on: VCs.
Their analysts quietly used the exchange themselves—on personal time—then individually urged partners: “This thing is genuinely good.” Partners picked up the phone. Yan and iliensinc did zero outreach—no pitch deck. The protocol generated fees, but Yan insisted the team take none from day one. When VCs called asking for decks, Yan and iliensinc just talked—and eventually realized: there truly was none.
By January 2024, funds began visiting in person. iliensinc knew the process—she’d been an investor. She began walking Yan through terms he needed to know, rights clauses to watch. For about two weeks, he cooperated. “It felt almost natural,” he told me. “VCs were calling—so maybe it was time to raise.”
His sole condition: only consider term sheets valuing Hyperliquid at $1 billion. The protocol had been live less than a year. The team burned hundreds of thousands monthly—all from Yan’s personal funds. When a fund met his condition, Yan spent a weekend thinking.
He’d asked founders and VCs themselves what fundraising was truly for. None convinced him their money was worth more than money itself. At some point, refusing felt right. Once that feeling clicked, nothing more needed saying.
Monday morning, he told iliensinc: “We’re not taking it.”
“What the fuck?”
She couldn’t believe it. She managed the money—watching it burn daily. Now a fund offered ~$100 million, and after two weeks of prep, he rejected it. Others on the team couldn’t accept it either.
He called the fund and declined. They didn’t believe him. Surely he’d accepted another offer? No. Hyperliquid wasn’t a company—it was a protocol, and neutrality was paramount from day one. “If Bitcoin had taken a VC round,” he said, “I truly don’t think it would be Bitcoin. Its entire value proposition would’ve been destroyed.” And he didn’t need the money. To this day, many team expenses still come from Yan’s pocket.
On January 28, 2024, he tweeted four lines:
No investors. No paid market makers. Team takes no fees. No insiders.
The Token Airdrop: Crypto’s Largest Wealth Transfer
Hyperliquid holds one meeting daily: a morning standup. I watched one on my second day in Singapore. The team crowded around an engineer’s screen. A plush dragon sat atop the monitor. They tested a new feature called “portfolio margin,” discussing potential failure modes. Long stretches passed without dialogue. Yan stood arms crossed, head bowed, staring at his bare feet. The engineer beside him did the same. These silences weren’t awkward or brief—no one in the room found them strange.
Partly due to personality. The team is young—ages 24 to 31—almost uniformly brilliant introverts. But when I asked Yan if he read regularly, he hinted it went beyond shyness.
“I read far less than conventional wisdom deems optimal,” he smiled, adjusting dark-rimmed glasses. “Reading a book deeply enough to permanently reshape your thinking is extremely time-consuming. The ROI isn’t great.”
He shifted his jaw—a habit I later recognized—like equalizing ear pressure mid-flight. A specific risk in writing about young technologists is that they’ll inevitably tell you they don’t read. I appreciated Yan adding he reads roughly one book every two months—and hopes someday to sit down and finish the ones he hasn’t. Then he explained why reading more must wait.
“If you’re not the first to do something,” he said, “it’s probably not worth your time. I genuinely believe that. Operate on that assumption, and reading becomes less helpful. If useful references already exist for what you’re doing, it’s likely already been done. If it’s already been done—why do it?”
By late 2023, Hyperliquid faced another crypto industry problem with a ready-made playbook. Yan routinely ignored such playbooks. A crypto project’s token grants holders stakes in its success. Deciding who gets tokens first—and under what conditions—typically uses points programs. Projects announce “use the platform to earn points.” Users assume points convert to tokens. They flood in, racing to accumulate maximum shares before conversion.
The problem: Most influx aren’t users. They’re professional teams reverse-engineering formulas, running bots to farm maximum rewards, then leaving. Real users—the ones points programs should serve—get scraps.
Hyperliquid’s version launched November 1, 2023. Users earned points weekly by trading—but the formula remained unpublished. Nobody knew how it worked. Every Friday, iliensinc announced that week’s points, turning it into a ritual. Users watched her Discord ID flicker “typing…” then gathered to compare allocations, share screenshots, and theorize how the system worked. “Rewarding real users is critical,” Yan said. “Defining ‘real user’ is hard—but Hyperliquid’s points program likely reduced bot participation from 99% to 20%.”
Around the same time, market makers Yan refused to pay directly began arriving. One—among Binance’s largest—was wary of new platforms post-FTX. But mutual contacts spoke highly of Yan, and at a September 2023 conference in Singapore, he met Yan and iliensinc for the first time. “Jeff has ambition but zero arrogance,” the market maker told me. “He describes his work with precision—every claim checks out.” He messaged his team immediately: “We should integrate.” They went live two weeks later.
What this market maker saw confirmed what users noticed. Infrastructure design nuances only traders spot. Hyperliquid embedded a “speed bump” mechanism—making it harder for aggressive quant firms to front-run other market makers. The industry later copied this feature. The result: Market makers could display deeper liquidity without needing millisecond latency to survive. Yan deliberately sacrificed some trading volume—specifically, volume from institutional front-running—to improve pricing for ordinary users. This trade-off reduced Hyperliquid’s own revenue.
During the same conference—Token2049—Yan and iliensinc decided to relocate. U.S. regulatory prospects for crypto derivatives looked uncertain; Yan told me building in America felt like unnecessary risk. A lawyer I interviewed described that period as U.S. regulators “using every tool to drive this technology overseas.” iliensinc evaluated Hong Kong, Switzerland, and Singapore—choosing Singapore for its modernity, safety, and lack of distraction.
In spring 2024, the team moved. The city suited Yan—it was boring. He operated in two modes: work and fitness. He swam, ran—anything exhausting but injury-free. This principle stemmed from a Puerto Rico e-bike accident—leaving a facial scar and forcing him to avoid keyboards for a week. Exercise cleared his mind for coding. His sole leisure concession was Sunday mornings. The rest of the week belonged to Hyperliquid. He even cut his own hair—barber visits wasted time.
He didn’t see this as abnormal—or rather, he saw most people’s attitudes toward work as abnormally lax. “I think people are generally too soft,” he said. “The brain is an organ. If you need to work more hours, you can train it.”
He’d learned not to impose this on his team. They ate lunch together daily—family-style, around a black wooden table. Thursdays were Chipotle. Since Singapore lacked Chipotle, they gave chefs the recipes—now they cook it. Lunch conversations typically drifted to what the team had recently watched or listened to. Yan often grew quiet then—likely lost in thought.
By spring 2024, Hyperliquid’s perpetuals daily volume surpassed $1 billion, and infrastructure groaned under user load. One afternoon, alarms blared continuously. The platform couldn’t handle incoming traffic. It was Hyperliquid’s first outage. But outside, everyone cared about only one thing: When would the Hyperliquid token arrive?
In May, Yan tweeted a six-month roadmap—packed with technical goals. No mention of tokens.
In preceding months, Hyperliquid expanded from derivatives into spot trading. Its first token—Purr—was named after the cat. Spot was necessary: To issue Hyperliquid’s token, the team needed a spot market to trade it. But it introduced problems derivatives exchanges never faced. With perpetuals, no one holds the underlying asset—you bet on price. With spot, someone must custody assets. Yan refused. The point was user-controlled assets.
To solve this without becoming a custodian, he realized he had to stop viewing Hyperliquid as an exchange built on a blockchain—and start seeing it as a blockchain with a built-in exchange. The chain they’d built to run the exchange—already processing hundreds of thousands of orders per second—could become programmable. It would become an open system where anyone could write code and build financial applications—just as thousands of developers already did on Ethereum. The difference? Ethereum was too slow for a real exchange—precisely why Yan built his own chain.
If he opened the chain, assets could enter Hyperliquid via decentralized bridges secured by the protocol itself—no single party acting as custodian. And anyone building on the programmable layer could tap the exchange’s order book and all its liquidity. A developer could build a lending platform, stablecoin, or mobile trading app—plugging directly into the same market where professional market makers quoted billions daily.
Yan disliked analogies. He’d tell you Hyperliquid has no counterpart in traditional finance—people prefer cramming new things into old categories rather than understanding them on their own terms, and that’s a mistake. But for those of us who aren’t Yan, it’s like Amazon building cloud infrastructure to power its e-commerce platform—then discovering the cloud service dwarfed e-commerce. Yan used the phrase for the first time in that tweet: Hyperliquid will host all finance.
He’d hesitated to make this shift. He told me he subconsciously resisted signing that commitment. Embedding a virtual machine into Hyperliquid was a massive engineering lift—the team didn’t know if they could pull it off. They didn’t know how much work would start from scratch. But at some point, he said, the answer became obvious. If they didn’t do it, they’d spend years cobbling components together—part Binance, part Ethereum, but neither—and regret it.
The community exploded. They expected an airdrop. Instead, they got infrastructure tweets. A 1,000-like comment quoted *Breaking Bad*: “We had a good thing.” “I hate this. You betrayed us.” Users didn’t want a blockchain. They wanted money. Team member Xulian—who joined after a 15-minute user interview stretched to 90 minutes—absorbed this anger. “Jeff thinks about what’s best long-term,” he told me. “We genuinely don’t care how something looks short-term.”
The loud voices, iliensinc said, eventually tired themselves out. The team spent the next six months building spot, developing the programmable layer, testing on testnet, and preparing staking. Then, on Friday, November 29, HYPE arrived.
Hyperliquid airdropped 31% of total supply to ~94,000 early users—no conditions, no lockups. If you’d used the platform and earned points, you woke up that morning richer than you went to bed. At launch price, the airdrop was worth over $1 billion. At its all-time high, it peaked at $16 billion—the largest wealth transfer in crypto history, with every dollar going to users.
The team’s own allocation—23.8%—was smaller than the community’s and vested over years. On airdrop day, they received nothing. VCs received nothing. If they wanted tokens, they’d buy them on the open market—at the same price as everyone else—on Hyperliquid (since it wasn’t listed elsewhere). That was another thing you had to pay for.
That morning, Yan didn’t need to explain anything on Twitter. “Woke up to six-figure airdrop,” one user wrote. Another replied: “HYPE changed my life today. Enough to live comfortably for years, help my family, go all-in on the bull market.” Someone else said: “Seven-figure airdrop—thanks, Jeff.”
“I feel really good about it,” Yan told me. “It’s rare for early participants to truly share upside—and gain meaningful ownership of the network.”
I asked how he felt since then—now that everything carried a public price tag.
“Terrible,” he said.
The Jelly Jelly Attack and Big Exchanges’ Siege
On a Wednesday night late March 2025, iliensinc’s computer began alarming. She was on a call. She hung up. On-screen, HLP—the Hyperliquid community vault—balance was falling.
A trader had probed Hyperliquid’s defenses days earlier with small coordinated positions. Now the probe ended. He opened three positions on Jelly Jelly—a low-cap token with ~$15 million market cap and $72,000 daily volume. One large short, two longs. The short was designed to blow up. He shorted a token he planned to pump—when the position imploded, others would absorb it. Like handing someone a grenade with the pin pulled.
HLP absorbed it. On Hyperliquid, when the order book can’t absorb a trader’s liquidation, the community vault takes the position and unwinds it gradually. Normally routine. But Jelly Jelly had almost no order book—and when HLP got stuck, unable to exit, the attacker frantically bought Jelly Jelly on public markets. Price surged over 500% in under an hour. Each tick deepened the vault’s loss.
iliensinc watched losses cross $5 million, $8 million, $12 million. Nothing in the system could halt it. No one designed for this: using a $15 million-market-cap token as a weapon.
Validators across Asia and Europe came online. Hyperliquid’s blockchain is secured by ~20 validators—individual operators who verify every transaction, earning voting rights by staking large amounts of HYPE. Many used Hyperliquid before the token existed. They saw the same public ledger—visible to anyone—and concluded this wasn’t a trade. Within minutes, all voted to delist Jelly Jelly and settle positions at pre-manipulation prices. Every legitimate position holder was unharmed. Only the attacker lost.
This exposed critics’ long-awaited question: If ~20 validators can override market prices and settle contracts at numbers they choose—how decentralized is this system? Yan didn’t dodge it. The small validator set was intentional. A system releasing upgrades every few weeks couldn’t coordinate 1,000 participants for each. The set would expand over time—but never at the cost of the speed that brought Hyperliquid this far.
“The fix took a month. Learning lessons from attacks—not being told in advance—is terrible,” Yan said. Hyperliquid—never paying market makers, never taking fees—offered bounties up to $1 million for vulnerability reports. “But these people clearly didn’t want to notify us. They wanted to exploit it.”
As the attack unfolded, Binance and OKX—the world’s two largest centralized exchanges—launched Jelly Jelly perpetuals on their platforms. A Twitter user @-mentioned Binance co-CEO He Yi, urging her to list it. “If you list Jelly Jelly,” they wrote, “Hyperliquid might be finished.” He Yi replied in Chinese: “OK, received.”
This is the return on ambition. You leave a beach in Puerto Rico where no one knows you. You build from scratch—with a TV and your own savings. You reject $100 million. You send billions to strangers. What do you get?
War.
In 2023 and 2024, Hyperliquid was small enough to ignore. The airdrop changed everything. Market cap: $4.2 billion, then $9 billion, then more—meaning every major crypto company saw Hyperliquid as a future threat stealing their lunch. Binance announced its own decentralized exchange. Coinbase and Robinhood launched futures products. New protocols launched targeting Hyperliquid. Then someone followed Yan into his apartment elevator.
This might mean nothing—but violent attacks on crypto holders nearly doubled in 2025. In France, a hardware wallet co-founder had a finger sawn off, with photos sent to business partners as ransom. A Canadian family endured waterboarding. Crypto transfers are instant, irreversible, and require no bank approval. A person with a wrench and a wallet address can wipe out fortunes.
Yan moved to a safer location, hired bodyguards—and became, in a sense, confined to Earth’s safest island city. Two private security personnel accompany him when traveling. iliensinc began testing the team: If strangers ask where you work, what do you say? That’s why nearly everyone speaking for this article used pseudonyms.
The October 10 Stress Test
I asked Yan what was hardest in 2025. He didn’t mention Jelly Jelly, competitors, or bodyguards. He said: API servers.
All summer, Bitcoin breached $100,000, Hyperliquid’s monthly volume topped $400 billion, and servers connecting market makers to the blockchain buckled. Too many market makers integrated—each sending torrents of orders, cancellations, and updates—the relaying infrastructure couldn’t keep pace. An order meant to execute instantly took three seconds.
The chain itself never went down. User funds were never at risk. But three seconds—in a market where milliseconds decide fate—was a red flag. “If congestion happens outside extreme volatility,” iliensinc said, “then during actual events, it’s unacceptable.” Yan slept almost not at all for weeks—going to bed at 1:30 a.m., pinged awake at 3 a.m. when things broke again. The team rewrote the servers from scratch.
October 10, the event struck. Trump threatened 100% tariffs on Chinese imports. Within 24 hours, over $19 billion in leveraged crypto positions were liquidated—the largest in industry history. Over 1.6 million traders entered a self-reinforcing cascade—forced selling depressed prices, triggering more liquidations, further depressing prices.
Hyperliquid had zero downtime, no withdrawal pauses. The rebuilt servers held. The Jelly Jelly fix held. HLP absorbed billions in liquidations—earning $40 million. But because every Hyperliquid blockchain transaction is public, anyone could count its liquidations. Other exchanges reported liquidation data imprecisely—Binance publishes one per second. Major media rely on exchange-provided data—which is misleading. Reports claimed Hyperliquid processed more liquidations than any other exchange. It looked like the riskiest trading venue—simply because it was the most transparent.
Three days later, while the rest of crypto tallied casualties, Yan’s team released the upgrade defining Hyperliquid’s next phase: HIP-3. HIP-3 lets anyone staking 500,000 HYPE deploy new perpetual markets on-platform—setting parameters, choosing price feeds, keeping half the trading fees.
By year-end—its second full year of operation—Hyperliquid generated ~$900 million in profit. Zero dollars went to the team. 99% was automatically converted to HYPE and burned—permanently removing it from circulation—returning nearly all platform revenue to token holders.
I asked iliensinc how she viewed 2025. She said: “Feels like we’ve grown up.”
The Endgame of Finance
On my final afternoon in the office, Yan and I sat at the black dining table beside the kitchen—next to those unopened whiskeys—where the team eats lunch daily. I still had questions.
Hyperliquid has spent the past year distributing its components outward. Builder Codes—released before HIP-3—let independent developers build trading apps atop Hyperliquid’s order book, keeping a share of fees their users generate. Paradigm co-founder Matt Huang called it “an outstanding UX franchise model.” These teams have earned over $70 million since October 2024.
HIP-3 goes further. Within six months, seven independent teams deployed hundreds of markets—mostly non-crypto assets: crude oil, gold, stock indices, forex. Top deployer Trade[XYZ] grew 38% weekly since October 2025, amassing over $130 billion in volume across 192,000 traders. Markets deployed by independents now comprise half of Hyperliquid’s total volume. In February 2026, HIP-4 launched—allowing anyone to deploy options or prediction markets on-platform. HIP-3 opened Hyperliquid to any asset with a price. HIP-4 opens it to any event with an outcome.
The most influential product on Hyperliquid is now built by people outside Yan’s team—and will remain so. I asked how he felt about that. What the team should build versus what to leave to others.
“It’s a dynamic question—I don’t think there’s a right answer,” he said. “The most important angle is philosophical. Are you building a financial super-app—like Robinhood—or a financial system?” He admitted he didn’t know which would win. “But I believe a universally accessible financial system is a better outcome for the world—a system built on open rails, owned by no single company.”
“To build it, we constantly ask: What must we do to enable others to succeed on Hyperliquid—and build their own businesses? When people compete and own their own things, systems become more resilient and scalable.”
He said the path of least resistance is doing everything in-house—keeping it all within one company. They chose the opposite. “It’s harder—but we care how we reach our goal, because the path determines what we ultimately build.”
Trade[XYZ]’s founder told me he thinks someday no one will know they’re using Hyperliquid. “Maybe in the end state, it’s just financial infrastructure and the liquidity layer,” he said, “and Interactive Brokers, Phantom, or someone else faces users. That’s actually beautiful.”
Paradigm’s Huang invested heavily in HYPE on the open market shortly after launch. “More astonishing,” he told me, “is that this was built by 11 people.” Eleven people—and almost no AI. The office has dedicated AI laptops running latest models—used only for idea exploration. “We closely monitor AI capabilities,” Yan said, “but it’s not yet good enough to write critical code.”
I asked Yan about the largest shadow looming over all this. Hyperliquid’s cumulative volume since 2023 exceeds $4 trillion. It holds 37% of the decentralized perpetuals market. Yet the world’s largest capital market—U.S. users—can’t access it. Americans are blocked.
The barrier is Dodd-Frank—the 2008 crisis-era law requiring every derivatives trade to flow through regulated intermediaries. Ironically, Hyperliquid’s public ledger already gives regulators what Dodd-Frank aimed for: real-time visibility into all system leverage. But until the U.S. Commodity Futures Trading Commission (CFTC) issues new rules, Americans have no legal path to trade derivatives via decentralized protocols. Yan didn’t form his own policy team—consistent with his philosophy. A month after my visit, the Hyperliquid Policy Center launched as an independent nonprofit, led by Jake Chervinsky—a prominent crypto lawyer with a decade in the field. The independent Hyper Foundation—supporting the Hyperliquid ecosystem—donated 1 million HYPE ($28 million) to seed it.
Yan acknowledged Hyperliquid had grown beyond “just build, don’t worry about anything else” as a viable strategy. “There’s lobbying happening in the opposite direction,” he told me. “I can’t confidently predict the outcome. But regulation ultimately reflects the people’s will—I’m optimistic about the direction.”
Go and the Endgame
One question I’d saved all week: You don’t seriously think Hyperliquid can host *all* finance, do you?
He laughed. For someone who cuts his own hair, he laughed more than expected. “I mean, ‘all’ is definitely overstated,” he said. “That’s our aspirational target. But it’s extremely hard—and decades-long goals sound arrogant.”
“That’s the difference between Go and chess,” he continued. “In chess, the stronger you are, the farther ahead you calculate. In Go, possibilities explode. The focus is more on intuitive judgment for the next move—not exhaustively searching the entire tree.”
He saw I wanted more, so he reframed it. He tries to live by this principle: Be certain you’re moving in the right direction—and execute the current step flawlessly—but you don’t need to know precisely where you’re going.
That night—Friday—the team dined at a hotel’s Chinese restaurant. The engineer who’d colonized the office with plush toys couldn’t make it. Everyone else was there—including me. We were led through a quiet lobby, down a corridor, into a private room paneled in dark wood, with carved lattice screens and a round table. At the far end, behind a partition, armchairs surrounded a low table. We sat there first, sipping tea.
The room was freezing—AC set for hotter nights. Someone handed the youngest engineer a blanket. He draped it over his shoulders—reading “Christian Dior.” So he and Yan discussed luxury brands—neither showing expertise. One mispronounced “LVMH” as “LHVM.” Neither corrected the other. iliensinc sighed, adjusting her Ralph Lauren hat.
When we moved to eat, the lazy Susan spun—never stopping. Dish after dish appeared—until a wide blue-and-white porcelain bowl arrived, silencing the table. Shallow water covered river stones and small leaves—a miniature koi pond. In its center sat white scallop-shaped noodles. Three tiny orange fish circled endlessly in the moat between bowls. The waiter introduced the dish. These fish, he said, rest 30 days to work five minutes. We watched them circle—again and again—before they were removed, beginning another month’s vacation.
We left around 9:15 p.m., stepping into light rain. I said goodbye, hopped in a taxi to the airport. Minutes from the hotel, the cab climbed a left curve on the highway—financial district flashing into view: HSBC, J.P. Morgan, Standard Chartered, Deutsche Bank, Citi—logos glowing against the black sky. Then the road straightened eastward, and the towers receded one by one in the rearview mirror—until only wet pavement remained. Yan walked the opposite way—to work—his bodyguard waiting.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@joincolossus
