Deep Profile Paradigm Matt Huang: What should a great founder look like?

Original title: Paradigm Shifts

Original author: Dom Cooke, Colossus Review

Compiled by: Heilsmann, ChainCatcher

“ Sometimes I feel like I’m managing the X-Men Academy ,” Matt Huang describes Paradigm, the $12 billion crypto venture capital firm he founded, as a gathering place for smart mutants with superpowers.

For example, Charlie Noyes, who is the number one employee of Paradigm. The 19-year-old MIT dropout who can't operate the calendar, was five hours late for the first time at 10 a.m. meeting and was unapologetic, but that didn't delay him becoming a general partner now; Georgios Konstantopoulos, CTO of Paradigm, who transformed from a fan of World of Warcraft to one of the most prolific engineers in the cryptocurrency field; or the developer known only for X nickname transmissions11, Paradigm found him on the Discord server, when the anonymous person was still in high school.

“They can be very outrageous at times, especially create chaos and drive you crazy,” Matt Huang said. “But once you see what they can accomplish, you think, oh my god, no one else in the world can do that.”

On a cold morning, I visited Paradigm's San Francisco-based company, and two Huang's team members are working on a mechanism that can reshape the way hundreds of billions of dollars of digital funds flow in the financial system. In a top-floor conference room, the curved chamber resembles the cathedral’s echo gallery, partner Dan Robinson explains their latest breakthrough at high-frequency trading speeds while pounding the floor with the “Paradigm Green” Nike Air Force 1 shoes. Research partner Dave White, wearing hexagonal glasses and a messy beard, leaned over in front of his laptop, and stopped from time to time to discuss the equation behind the concept he invented. Huang listened carefully, his athlete-like body wrapped in a simple black Japanese sweater, with a quiet authority that always leads people.

“Everything he can get in touch is good,” said Doug Leon, who was head of Huang’s 2014-2018 at Sequoia Capital. "He's very smart and extremely humble. Few people have seen Matt Huang without thinking he's particularly outstanding."

Through two huge arched windows above Union Square in San Francisco, the traditional finance concrete towers appear to the east, while the SoMa area startups extend south. It's a perfect view for a company that connects traditional finance with cutting-edge technologies—and for Huang, who has been known for discovering revolutionary potential in his career.

In 2012, during a week-long vacation in Beijing, he visited a startup that operates in two apartment units. Founder Zhang Yiming is developing a personalized news app—Huang believes that the idea is doomed to fail. But sitting at a fridge covered with dust and an old IKEA table, Huang noticed something beyond language as he watched Zhang Yiming speak through a translator: "I remember having a deep intuition at that time, this person was extremely capable, focused, motivated, and balanced, and would not let himself collapse. He had a very clear understanding of what he wanted to build, and a strong ambition to conquer the world." Zhang Yiming was the most impressive person he had ever lived with - so impressive that Huang had to invest. That company later became the creator of ByteDance, TikTok, and Huang's shares in it had become nine or ten digits - Huang does not keep the spreadsheet, so he can't tell you the exact number.

That instinct to discover talent is at the heart of Paradigm. Paradigm started in 2018 when Coinbase co-founder Fred Ehrsam found Huang at Sequoia Capital to propose a vision of a unique investment company. They started as equal partners, but Ehrsam later stepped back and distributed time to cryptocurrency and his newly founded brain-computer interface startup. Ehrsam commented that Huang is "a person born to run Paradigm".

Huang's parents were world-leading financial theorists and pioneering computer science professors, who grew up at the intersection of mathematics, economics and technology. Over six years, his company has grown from $400 million in management to over $12 billion by making early, concentrated investments in basic crypto projects while also building an important part of the core crypto infrastructure. Paradigm researchers are also investors, developing fundamental innovations and then opening them up for use throughout the industry. This is an unusual way for a financial company, but Paradigm is not a typical investment company. It's more like a combination of a research lab and engineering team, wrapped in a delicate shell of Wall Street on the West Coast.

" He 's very smart and very modest. It's hard not to think he's very special after seeing Matt. "

–Doug Leone, Sequoia Capital

Back in the top floor conference room, Robinson and White are working on “bullseye liquidity,” a breakthrough technology that could change the way stablecoins are traded. Stablecoins have become an important part of the cryptocurrency financial system, but their trading infrastructure remains primitive, and each pair of transactions requires an independent capital pool. Their innovations integrate these fragmented markets into a single, efficient system. Although this could bring significant advantages to Paradigm's portfolio companies such as Uniswap and Noble, they still plan to open their research results on their blog. "If someone implements this idea and improves the cryptocurrency ecosystem as a whole, we're totally fine," Robinson said.

White paused its partnership with OpenAI's o1 Pro - he is validating mathematical proof - to refine a view on n-dimensional space. On screen, Robinson shows a mathematical visualization image that looks like a quarter of the film of Captain America's shield. Huang listened most of the time, and he always preferred to listen, but when he spoke, it was obvious that he had completely absorbed the complexity they presented.

Many years ago, when they were kids, Robinson recalled that they would always argue endlessly until Huang opened his mouth. "He didn't talk much," Robinson said, "but we always ended up doing what he suggested." The person who knew Huang best described him as someone with extraordinary abilities hidden under his quiet appearance. "When he communicated with Matt Huang, he had a high insight every minute, even if there were a lot of times he didn't say a word," Stripe co-founder Patrick Collison observed. He included Huang on the board in 2021. His attention was meticulous, extending to everything he touched—from the speed of Paradigm websites, to the unpopular Japanese streetwear he loved, to the people he hired. "He has a high standard of excellence and will never tolerate mediocrity," said Coinbase CEO Brian Armstrong.

However, under this strong personality, there is a relaxing humility hidden. As Leone said, "He has a good sense of humor, but because he has so many beautiful qualities, you will involuntarily take him seriously." Perhaps the most telling thing is that these insights come from others - Huan is a kind of person whose greatest achievements are whispered rather than hyped, and his influence is more felt than publicly advertised . "Not every great investor or leader is a great person. Huang passed with excellence in all the character tests about 'whether this person can be the godfather of your child's."

Paradigm is not a typical investment company. It is more like a combination of a research lab and an engineering institution, draped in the exquisite cloak of Wall Street on the West Coast.

This combination of technical excellence and silence and integrity helped Paradigm establish its position as one of the most important institutions in the field of encryption. In an industry that grew from zero to $3 trillion through waves of speculation and crashes, Paradigm’s open source tools support 90% of smart contract development. Their innovations have helped hundreds of billions of digital dollars flow more efficiently. And its investments have won the trust of the world's most savvy investors, including Harvard, Stanford, Sequoia Capital and Yale.

wake up

One of Matt Huang's earliest memories is that he walks alone on the streets of Tokyo, a nine-year-old traveling through the world's largest city. Every morning, he would pass through narrow alleys and bustling streets and spend an hour to and from school. This kind of independence from childhood shapes his worldview. “Once you have N=2,” he said in comparison to Tokyo and New York, “this changes the way you treat everything.”

In 1997, Matt Huang's father Chi-fu Huang was ordered to establish the Asia office of Long-term Capital Management Company (LTCM), and the whole family moved to Japan. Prior to the move, Chi-fu Huang managed LTCM's Asian trading business in Greenwich, working from 4pm to 3am every day to match market time. Born in Taiwan, Chi-fu Huang is the only boy of five children, he has four sisters, and his parents sent him to the United States alone with their meager savings. He has been studying hard, eventually becoming a professor of economics at MIT, then joins Goldman Sachs, establishing and leading the fixed income derivatives research division, and finally entering LTCM, a company with Nobel Prize winners, representing the perfect blend of academic theory and market practice.

Huang's mother Marina Chen opened her own path in the academic world after immigrating from Taiwan. At Caltech, she pioneered parallel computing research under the guidance of legendary technologist Carver Mead, developing technologies that are still used in modern processors today. Although Chen, one of Yale’s first female computer science professors, seemed destined to achieve outstanding academic achievements, she chose to leave the academic community and concentrate on raising and educating her three sons.

Huang's family dinner is like an investment committee. Every night, when children hear the sound of the garage door opening, they will rush to complete the reading task assigned by their father - carefully selected articles that suit their age, from the principles of economics to "Scientific American". At dinner, they have to answer questions about the assigned topic. Each child has formulated different response mechanisms to the strong demands of his parents. According to boss Matt Huang, his instinct is to resist.

In 1998, the LTCM model failed during the Russian financial crisis, and their Tokyo division suddenly ended. The crisis emptied Huang's savings. However, Chi-fu Huang emerged from the ruins and founded Platinum Grove Asset Management in 1999 with his LTCM colleague, Nobel Prize winner Myron Scholes. In less than nine years, the company's assets grew from $45 million to $6 billion, becoming one of the world's largest fixed income hedge funds before the 2008 financial crisis. This pattern of finding opportunities in chaos and creating new things in a systematic collapse is no stranger to his son.

Scarsdale, a New York suburb, is Huang's fourth home in 11 years after Tokyo. As one of only three Asian students in this so Jewish-dominated school, these constant relocations and acculturation have trained his ability to understand social dynamics and interact with people of different personalities.

In class, Huang could not sit still. He was restless and repeatedly disrupted the order of other students, resulting in his expulsion from Chinese weekend schools. "Uncontrollable," his parents later described him at his wedding. However, when he lives as he wants, he shows the ability to focus. Huang seems to be unique in his "not cool, but academically inclined" circle of friends. He makes movies, debates liberal philosophy and is proficient in games. His attitude towards StarCraft—the semi-professional game on international servers—presents his attention to detail will become his mark, even extending to his current obsession with handstands.

Zhang Yiming speaks through a translation, but Huang has been attracted by the founder 's nonverbal expression - his gestures, expressions and strong emotions portray a picture that can be understood without words.

When he discovered mathematics, everything changed. The Mathematics Club gave him a natural affinity for mathematics. Although he was not a top player in national competitions, it also allowed his parents to see that as long as they were properly challenged, their uncontrollable son could perform well in school. Poker and chess have also become another way of venting his analytical thinking.

The reinvented student was admitted to MIT with his own efforts. In 2006, he found himself in "one of the most weird places on Earth", where he studied math. He once took a semester off school to play online poker and opened 8 gambling tables at the same time. But the decisive moment (besides meeting his future wife) comes when one of his close friends, Albert Ni, announced his withdrawal from school and joined a small startup called Dropbox, becoming its sixth employee. For someone who has been trained to become a PhD, it seems unimaginable to give up a bachelor’s degree. However, you did not fail - he was one of the most capable people Huang knew, and he made careful choices to create something new. This prompted Huang to read all of Paul Graham’s articles, through which he discovered the temptation of Silicon Valley and the ultimate rebellion: to open his own path.

Huang and his roommate applied for Y Combinator in his final year at MIT, which was initially rejected. Graham told them: "We like you guys very much. But your ideas are terrible." Six months later, they successfully joined YC with a viable prototype. MIT graduates drove across the country to San Francisco in six days. At YC, they created what Huang now calls “bad ideas”—a TV guide website in the streaming era called Hotspots. This "failed entrepreneurship" lasted for two years, and this experience made him deeply resonate with the founder. Later, Hotspots was acquired by Twitter, when Twitter was in its pre-market phase, and he observed that it was a "badly managed" company.

In 2012, Huang is preparing to open up a new path. In his opinion, Silicon Valley has become too dull; the consumer sector can be easily predictable and cannot find interesting jobs and convincing returns. During his week off Twitter, he came to Beijing to meet with six founders with the idea of ​​starting a tech company in China. One of them is Zhang Yiming, who is developing a consumer app that seems doomed to fail. Zhang Yiming talks to Huang Renxun through the translation, but Huang Renxun finds himself attracted by the founder’s nonverbal expression – his gestures, expressions and concentration paint a picture that can be understood without words. Huang Renxun thought to himself as he left the apartment, “I have to find a way to support this person.”

He invested a sum of money in ByteDance at a valuation of $20 million and $30 million, his largest personal investment at the time. Now ByteDance is worth US$300 billion, and its investment has increased by about 10,000 times. He still owns a majority of the shares, and while he is "increasingly calm" about it, he admits "it will definitely mess up your mind because it's probably the best investment I've ever made." That same year in San Francisco, he made seed investments in YC startups Instacart, Benchling, PlanGrid and Amplitude -- all of which are now multi-billion-dollar businesses.

In 2014, Huang was working on Twitter, a recruitment email from Sequoia Capital appeared in his inbox. He initially thought it was spam. Despite his outstanding performance, he has no intention of becoming an investor. But curiosity prevailed, and his interview task—a page of the company that Sequoia Capital should invest in— prompted him to write about Coinbase, which had only seven employees at that time.

At Sequoia Capital, Huang discovered the "highest standard place" that he has experienced. On the second day of his entry, Facebook acquired WhatsApp for $19 billion, and the partners briefly gathered in the hall. The champagne was poured out, but no one clinked glasses to celebrate. Within five minutes, everyone returned to work. The 10 billion dollar exit transaction is overshadowed by the legendary portfolio of trillion-level companies such as Apple, Google and Nvidia . This kind of culture made his already great ambition even more expanding.

“ You can see here how far the axis can extend , what a great founder looks like ,” Huang said of his four years at Sequoia Capital. "If you don't get exposed to these, your overall awareness of possibilities wouldn't be so high." Sequoia Capital also showed him that excellence comes in many forms. Working with investors who are very different in style but consistent on key levels gives him confidence in developing his approach: “I feel very free to realize that I can do things the way I do.”

Sequoia has gained some valuable returns. "In previous poker tournaments, Sequoia America was beaten by Sequoia China, but Matt Huang won the game for us after joining Sequoia America. Thanks to him, we brought back the jacket with an exaggerated color of Don Valentine," Leone recalled. You probably won't hear this story from Huang, he never boasts about his achievements. There are many things you can only learn from others or know what to ask.

Leaving home

Matt Huang first came into contact with Bitcoin at MIT in 2010, and he was immediately attracted by the idea that Bitcoin perfectly integrates mathematics, economics, computer science and game theory.

“I thought it was just a very beautiful idea in my heart,” Huang recalled. But in the early days, his exploration of Bitcoin was more of a curiosity than investment. It wasn't until 2012 that he bought some bitcoins on the then mainstream exchange Mt. Gox and experienced his first big bubble. "It was almost the first time I lost money," he reflected. "And then you kind of give up and write it off as something dead. And when you see it come back and not die, you start to be curious."

According to multiple sources, legendary investor Michael Moritz called Matt Huang " the only regrettable loss in Sequoia's history." "He was the first person in my career to volunteer to leave Sequoia," Leone said.

At Sequoia Capital, Huang found few colleagues resonate with his increasingly firm belief in crypto importance. The company supports his interests—he led several cryptocurrency investments on behalf of Sequoia—but he is increasingly looking for conversation partners outside the company. He began attending monthly dinners in San Francisco, exploring ideas on emerging technology frontiers with six to eight investors who are also interested in cryptocurrencies.

It was during this period of 2017 that Fred Ehrsam, who had just stepped down from the position of Coinbase president, wrote a blog post advocating that cryptocurrencies are the Meta Universe. Huang, who was still in Sequoia, took the initiative to contact him to discuss this idea. “I know I’m not going to start a company around this idea,” Ehrsam thought, “but it would be fun to sell the idea to Sequoia for fun.”

This exchange that was originally just a desire to seek knowledge has evolved into an in-depth discussion of 40 emails between the two, digging deep into the possibility of cryptocurrency. Their backgrounds are perfectly complementary: Ehrsam co-founded and operated the most important companies in the cryptocurrency space, while Huang brought top investment experience.

“Nothing felt completely right before meeting Matt,” said Ehrsam, who had previously discussed with several potential partners that set up a fund focused on cryptocurrencies. Over the course of six months, they methodically explored the possibilities of cooperation, from investment philosophy to fund structure, testing their fit in all aspects. They are particularly focused on ensuring a real partnership—everything will be divided into five or five, a principle that “was driving some people crazy”, but is crucial to both of them.

Leaving Sequoia was painful for Huang. That was the first place he really felt belonging: "It felt like a place that would allow me to retire if they would leave me behind." Several sources revealed that legendary investor Michael Moritz called Huang "the only regrettable loss in Sequoia's history." "He was the first person in my career to volunteer to leave Sequoia," Leone said. But Huang is already convinced that cryptocurrencies will be one of the most important technology trends in the coming decades. "I can easily let go when he told me that he thought it was a once-in-a-lifetime opportunity. Follow your dreams and go after it," Leone said with a hint of regret. "I'm angry with myself because he's talking about bitcoin all the time, and I usually have a good sense of smell. If I'm really smart, I'll go in that direction and create an opportunity for him to set up a fund inside Sequoia."

Subversion of status quo: Huang and Ehrsam

Huang and Ehrsam founded Paradigm in June 2018, based on two core fulcrums: First, cryptocurrencies will become one of the most important technological and economic changes in the coming decades; second, this field lacks the type of investor they want as entrepreneurs—an investor who is well versed in the qualities of “crypto-native”.

Graham Duncan, founder and Paradigm consultant at East Rock Capital, Huang called the most helpful person in the early days of the company, was shocked by their beliefs from the very beginning. “They are always planning what can happen from a scale perspective,” Duncan said. "It shocked me, but it was by no means arrogance. Their time horizons were completely different, and what they planned eventually came true."

They raised their first fund at the end of 2018, raising $400 million from top institutions such as Harvard, Stanford and Yale, as well as Sequoia Capital — universities that make major investments in cryptocurrencies for the first time. The fund is structured in novel ways: open-ended, without a fixed timetable for return on capital, allowing them to hold public crypto assets and private investments. Then they made a bolder move: Instead of gradually calling funds like most venture capital firms, they quickly demanded full $400 million and began buying average Bitcoin and Ethereum, which initially accounted for about 90% of the fund. The average buying price of Bitcoin is between $4,000 and $5,000 per coin – a huge bet that the crypto winter, where Bitcoin fell by more than 70% in 2018, will eventually thaw.

Their top three employees embody different aspects of their vision. Charlie Noyes was the first employee Huang met in a Telegram group chat discussing Bitcoin Cash Fork. "From the messages he sent, I thought he was a 40-year-old bearded man, very cynical and weather-beaten. When he showed up at dinner, I was really surprised that he was only 19," Huang recalled.

Noyes has been immersed in cryptocurrencies since he discovered Bitcoin through the gaming circle at the age of 12. He has published papers on crypto applications, won two Intel Science Competitions, then dropped out of school to MIT, and then dropped out of school to join Paradigm. At first, he had some difficulties in adjusting to office life. He felt that it was normal to express his opinions on investment proposals through email and come to the office once a week. When he was late for work on the first day, Huang sat down to explain his career expectations. This patience is rewarded.

Today, 25-year-old Noyes is one of Paradigm's general partners. Huang compared him to an artist, able to make a huge intuitive leap by compressing a large amount of scattered information, and ultimately leading to a clear investment argument. For example, he identified MEV (miners can extract value) as a key issue in blockchain in 2020 and became the leader of Flashbots. Its infrastructure now touches almost every transaction on Ethereum and has established key market rules for this $450 billion ecosystem.

Dan Robinson, Huang's middle school friend, is also the smartest person he met when he was a child, and has excellent encryption technology. After disillusioned him at Harvard Law School, Robinson turned to learning programming and explored cryptocurrencies while working at blockchain company Stellar. Huang and Ehrsam have tailored a unique role for him, covering investment, research and helping portfolio company building. This compromise later became a template for Paradigm's research-driven approach, and Robinson then invented key mechanisms for Uniswap, the leading decentralized exchange in cryptocurrency.

She is the third partner of Matt and Fred , and has been involved in the company 's construction from beginning to end .

–Graham Duncan, East Rock Capital

Huang, Ehrsam and Palmedo outside the top floor of the San Francisco office

Alana Palmedo joined four weeks after Paradigm’s founding, when the company was still renting office spaces on a weekly basis, bringing the institutional rigor needed to connect cryptocurrencies to traditional finance. While she is not “a deep understanding of cryptocurrencies,” her complex operational experience in managing the Boston University Foundation and the Bill Gates investment office during the 2008 financial crisis proved invaluable. At first she was skeptical, but Huang and Ehrsam’s thoughtful approach to building an institutional-level company, and her intuition as a value investor—Bitcoin fell so low, “this is definitely the bottom”—ultimately kept her here.

"She is the third partner of Matt and Fred, and she's been involved in the company from beginning to end," said Duncan, who helped interview Palmedo. She initially took charge of everything from transaction settlement to finance to compliance, and then recruited experts for each function to allow the investment team to focus on transactions. Now, as a managing partner, she builds Paradigm’s high-performance culture that requires everyone to remain extremely transparent and daily self-reflective regardless of their role. “Everyone has to be in the top 1% in their field,” Palmedo insists.

In mid-2019, cryptocurrency prices began to rebound, but most investors remained cautious about the industry and Paradigm entered the market again. Its initial investor group added $360 million in investment. This time reflects Huang's style of doing things: raise funds when others suspect, from partners who are as firm as he believes that cryptocurrencies will completely reshape finance.

Although cryptocurrencies have not fully fulfilled their transformative commitments, Paradigm’s investment has already delivered good returns. According to public documents, its first flagship fund grew from $760 million to $8.3 billion by the end of 2024. Sources close to the company revealed that Paradigm also returned all initial capital from its limited partners, and has paid more than $1 billion from the fund so far.

Long-term vision

Despite Paradigm's early success, we can't help but ask, why is Huang interested in the wild world of cryptocurrencies? For Huang, money is no longer a problem, and it seems that there is a perfect job at Sequoia Capital.

Brian Armstrong, CEO and co-founder of Coinbase, also thought about a similar question—“Who would leave a job like Sequoia?”—and replied: “He is a silent killer. Our industry needs more people as upright as him, who have been in this industry for a long time and have legitimate reasons. He has the belief that he is going to take a different path.”

For Huang, the answer is simple: "I think I've always had some kind of doubt about authority, so when I see authority play out, I do wonder: Is this how the world we want to work?"

“Americans say that China looks like a dystopian country,” Huang said. “I don’t think they fully realize that the same thing is happening in the West.”

Taking the Obama administration’s “Slaying Point Action” as an example, the U.S. Department of Justice is trying to limit access to banking services in certain industries. Operation Kill Point 1.0 took place between 2013 and 2017, targeting industries deemed to be “high risk” such as payday lenders and gun dealers. During the Biden administration, Operation Kill Points 2.0 focused on "debanking" cryptocurrencies. Even individuals like Uniswap founder Hayden Adams and Gemini co-founder Tyler Winklevoss have found their personal bank accounts suddenly closed for no reason.

Huang believes that the development of cryptocurrencies is divided into three key stages: first, the currency, then the financial system, and finally the Internet platform. Each stage is based on the next stage and facilitates the formation of the next stage. “Currency is the upstream of cryptocurrency,” he explained. “Buying the first bitcoin or setting up the first wallet is usually the first step to using other cryptocurrency apps. It’s like getting an AOL account for the first time and connecting to the internet.”

The currency phase has produced amazing results. Bitcoin has grown from a 2008 white paper to an asset worth nearly $2 trillion today, becoming the most valuable startup since its launch. It is worth noting that nation-states—even the United States—have begun to adopt it.

Institutions that mocked the industry in 2018, such as BlackRock CEO Larry Fink, who called Bitcoin a “money laundering index”, now embrace the technology. In 2024, BlackRock’s Bitcoin ETF raised $50 billion in just 11 months, the fastest growing ETF in history. Traditional portfolio models are also changing, with Fidelity now recommending 1-3% cryptocurrency exposure. As institutions make special allocations for crypto assets, the classic 60/40 portfolio is becoming "59/39/2".

The second phase - building a brand new financial system - is accelerating. Traditional finance operates through layers of intermediaries, while cryptocurrencies enable near-instant transactions, all-weather markets and novel financial instruments, all built on a license-free track. The rise of stablecoins (blockchain-based digital currencies, linked to stable assets such as the US dollar) proves this potential, with its circulation growing from $500 million to over $200 billion since its inception.

The third stage - the Internet platform - is still in its infancy and is the least clear. Unlike today’s Internet, platforms control user data and online assets, while cryptocurrencies can achieve true digital ownership and enable direct user-to-user interaction without intermediaries. High transaction costs have hindered the development of daily applications such as social media and games, but Huang Renxun believes that this is changing as new expansion technologies significantly reduce costs. Today's infrastructure that supports NFT and meme coins will eventually support more serious applications, just as YouTube has evolved from Cat Video to one of the most important platforms in the world.

Of course, like every new technology, cryptocurrency has its dark side. Scam and hacking are commonplace, memecoin promotes short-term thinking rather than building what really needs, token prices fluctuate violently, projects crash frequently, and the industry as a whole often looks more like a casino than a financial future.

However, Huang took a long-term perspective. Just as the early Internet attracted both outstanding researchers and scammers and fraudsters, the open frontier of cryptocurrencies has both contributed to innovation and bred bad behavior. Every new wave, including the seemingly irrational speculative bubble from the outside, brings new talents and capital to build critical infrastructure.

Stablecoins are a perfect example. The 2017 ICO bubble brought mainstream attention to cryptocurrencies and gave birth to a new generation of crypto rich investors. Some of this capital flows into development of stablecoins, resulting in significant improvements in its infrastructure. On Ethereum, the cost of sending USDC has dropped from $12 in 2021 to $1 today. On Coinbase's popular Layer 2 network base, the same transaction costs less than a cent. In turn, circulation has grown exponentially, up 400 times since the bubble burst, and real-world use cases have emerged.

SpaceX uses stablecoins to remit income from emerging markets, convert local currencies into digital dollars, and enable instant transfer. Scale AI pays its global network of data markers through stablecoin tracks, eliminating cross-border friction and costs. Financial teams like Ramp have found another advantage: While savings accounts only have a few percent interest, stablecoins backed by Treasury bonds can capture most of the gains banks usually leave themselves.

The data confirms this trend. Trading volume has grown at a rate of 120% per year for five consecutive years. In 2024, stablecoins processed $5.6 trillion in payment transactions, almost half of Visa’s $13.2 trillion. This momentum prompted Stripe to acquire Bridge, a stablecoin payment platform, in October 2024. "Stablecoins are room temperature superconductors for financial services. Thanks to stablecoins, global businesses will benefit from significant improvements in speed, coverage and costs in the coming years," wrote Patrick Collison, co-founder of Stripe.

Matt has a unique personality. He is calm, rigorous and patient—these traits are uniquely suited to deal with any complex technology, especially those with lagging consequences like cryptocurrencies. ”

– Patrick Collison, Stripe

This adoption reflects a broader evolution of cryptocurrencies: Bitcoin was launched in 2009 and reached its first million users in 2011.以太坊（Ethereum）随后在2015年推出，并在2017年达到了相同的里程碑。接着是2019年的稳定币，2021年的去中心化金融（DeFi），2022年的NFTs，以及2023年的社交应用。

批评者经常强调加密货币对日常业务缺乏影响。 Huang 认为稳定币是下一个杀手级应用，但他也区分了像AI这样的“单人”技术和像加密货币这样的“多人”技术，前者提供即时效用，后者需要协调采用。 “这就像采用一种新语言或定居一个新城市，”他解释说。“如果你独自去做，它毫无用处。”他指出电子邮件作为一个有指导意义的类比。早期批评者称其为“技术上引人入胜但经济上幼稚”，就像今天对加密货币的怀疑一样。

与Huang 交谈时，他对整个加密货币的从容态度令人注目。Patrick Collison 将Huang 加入Stripe 董事会，既因为他的加密货币专长，也因为他更广泛的商业洞察力，他说：“Matt 的气质与众不同。他冷静、严谨且耐心——这些特质独特地适合处理任何复杂的、后果滞后的技术，比如加密货币。”

他的独特之处在于他能够同时持有投资论点的两面。Collison 说：“他能处理熊市案例，这些案例通常比牛市案例具体得多。然后他理解技术的可能性，看到小的萌芽事物如何在未来变成非常重要的事情。”

最近，人工智能已成为科技的新前沿，其清晰且即时的应用激发了全世界的想象力。Huang 和他在Paradigm 的团队甚至考虑过扩展他们的关注点以包括AI。但他们坚持了对加密货币的承诺，Huang 解释说：“AI 有没有我们都会发展得很好。加密货币是一项非常重要的技术，需要与AI 共存，但却没有太多优秀的捍卫者。我们要确保加密货币的成功。”

invention

这种确保加密货币成功的承诺，促使Paradigm 发展出一种不寻常的投资方式。虽然大多数风险投资公司等待支持赢家，但Paradigm 帮助创造了获胜的条件。这意味着不仅仅是分析趋势或开支票——而是解决扩展整个行业所能实现的基础技术问题。

公司以研究驱动的风格几乎是偶然形成的。当Huang 聘请他的中学朋友兼伴郎Dan Robinson 时，并不清楚一个从律师转型为自学程序员的人如何适应一家投资公司。“我们希望Dan 加入团队，因为他是我认识的最聪明的人，”Huang 说，“但他并非最具商业头脑，我们不确定他如何参与投资流程。”主要是为了让他加入，他们创造了一个新颖的角色，包括让Robinson 有时间从事开源项目，他们共同称之为“探索性研究”。

“事实证明，这种特定类型在加密货币中是极其重要的事情，”Robinson 解释说。“大多数投资研究是关于收集和分析现有信息。我们试图发明新东西。”研究团队的突破往往来自在公司甚至还未意识到需要答案之前探索理论问题，比如靶心流动性（bullseye liquidity）。

加密货币之所以独特，是因为数学机制创造的巨大杠杆效应。传统交易所可能需要数千台服务器和数百名员工来匹配买卖双方。但当Vitalik Buterin 在2016年在Reddit上提出一个简单方程（x*y=k）时，他为万亿美元市场在区块链上自主运行奠定了基础。挑战在于，这个优雅的解决方案虽然计算效率高，但通过将流动性分散在所有可能的价格上浪费了大量资本。

Robinson 从早期的以太坊研究社区认识Hayden Adams，后者将Vitalik 的概念发展成了Uniswap。加入Paradigm 几周内，他写了一份关于Uniswap 的备忘录，这促成了种子投资，并开始积极改进。他的贡献促成了Uniswap v2，使得任何基于以太坊的代币之间可以交易，帮助该协议的交易量从20亿美元扩展到超过1万亿美元。

但Robinson 和Adams 在2019年的大部分时间都在寻找更根本性的突破。通过数学探索，团队发现了一种在特定价格范围内有效集中流动性的方法——让交易者将资本集中在实际需要的地方。这项创新成为Uniswap v3，将资本效率提高了高达4000倍。500万美元的头寸现在可以提供与20亿美元分散在所有可能价格上相同的交易深度。到2022年10月，Uniswap 的估值达到17亿美元。

“当他们与其他公司竞争时，Paradigm 实际上可以帮助你建立一家加密货币公司。他们有协议设计、安全、法律甚至政策方面的专家。”

– Brian Armstrong, Coinbase

这种研究促成突破性产品的模式在Paradigm 的投资组合中反复出现。去年，当Blur 向公司提出增加保证金交易时，团队面临一个根本性挑战：如何安全地针对流动性差、价值难以确定的NFT 放贷？研究团队花了四个月开发了一种全新的借贷协议Blend。Robinson 指出：“如果你能解决针对NFT 的借贷问题，你就可能解决针对任何非流动资产的借贷。” Blend 推出后几个月内，就创造并主导了一个全新的借贷类别。

与将技术资源与投资决策分开的传统风险投资公司不同，Paradigm 的研究人员是每项投资的核心。他们参加每场pitch 会议并帮助做出每个决定。这种整合意味着他们经常发现其他人错过的机会，因为他们已经在处理类似的技术挑战。当像Terra 这样的算法稳定币变得流行时，Paradigm 选择避开——多年尝试设计更好的稳定币机制让公司明白，这些项目并未解决根本问题。

这种深入的技术工作在寻找和完成交易以及招聘人才方面创造了强大的竞争优势。“当他们与其他公司竞争时，”Coinbase 的Armstrong 解释说， “Paradigm 实际上可以帮助你建立一家加密货币公司。他们有协议设计、安全、法律甚至政策方面的专家。”

“我们流程的最大部分是弄清楚什么才是最重要的问题，”Robinson 解释说。这需要紧跟加密货币快速演变的前沿。“互联网的世代非常短暂，”Huang 指出，并将其比作夏洛克·福尔摩斯的街头流浪儿网络，这些流浪儿为他提供了伦敦街头的重要情报。“即使两年的时间差异，也可能影响你对加密货币文化的理解。”

这种洞察促成了Paradigm Fellowship，该计划旨在发现仍在学校里的杰出年轻开发者。该计划部分源于公司与transmissions11 的经验，团队在Discord 上发现了还在读高中的他。有一次，transmissions11从学校集会上拨入了一场pitch会议，这让他意识到与加密下一代创新者合作的机会。

Konstantopoulos、Noyes 和Robinson

C rypto C rypto C rypto

2023年5月的最后一个星期四，加密新闻公司The Block 的记者们发现了一件引发行业广泛争议的事情。通过Internet Archive 的Wayback Machine，他们注意到Paradigm 已悄然从其主页和社交媒体上删除了所有“加密货币”（crypto）的提及，重新定位为一家“研究驱动的技术投资公司”。这一看似明显的发现——尽管这一变化已有一个月之久——立即引发了强烈反弹。对于异常重视忠诚度的加密行业来说，这感觉像是背叛。

“我们不想再为你们工作了，”一家投资组合公司在推特上写道，提到这次品牌重塑以及Paradigm 对FTX 的投资，“这成了我们整个行业的伤疤”。这种批评刺痛人心，但体现了加密货币界残酷的坦诚。 Paradigm 不仅恢复了“加密货币”这个词，还加倍努力，在主页上添加了闪烁的霓虹灯ticker，上面写着：“crypto crypto crypto” 。

这次变化背后的现实更为平淡。团队中的两名研究人员抱怨说，潜在的AI 合作者看到Paradigm 主页上的“加密货币”后不再回复他们的邮件。“我们想，好吧，所有加密货币圈的人都已经认识我们，他们从不去我们的主页。他们看我们的博客，每一家投资组合公司和博客文章都与加密货币有关。所以这有什么大不了的？” Huang 解释说。“事后看来，这显然是个错误。人们把网站视为你引以为傲的集体声明。”（Patrick Collison 就曾指出，Paradigm 的网站“可能是你今年用过的最快的”。）

然而，这一事件暴露了更深层次的紧张关系。到2022年11月，比特币从2021年的峰值下跌了75%，跌至1.6万美元以下。以太坊下跌了80%。同月，ChatGPT 推出，引发了AI 热潮，让加密货币看起来像是昨日的前沿。主要风险投资公司已开始将注意力和资本转向AI。

对Paradigm 来说，网站争议标志着一个谦逊时期的顶点。就在18个月前，这家公司似乎无人能敌。其比特币头寸增长了15倍。最早的投资之一Coinbase 以850亿美元的估值上市。它募集了一只25亿美元的风险基金。然而，2021年的狂热甚至考验了加密货币界最自律的投资者。

Fred Ehrsam 看到了警示信号。2021年3月，他向投资组合公司发送了一封题为《度过加密货币周期》的信。他指出，价格在短短两个月内翻倍，比特币市值超过1万亿美元，“像素化的加密艺术品动辄卖出数百万美元”，他警告说：“参议员甚至都有激光眼！狂热无处不在。”凭借他在Coinbase 的经验——2014-2017年低谷期间三分之一的员工离开——他提出了具体准备建议：对系统进行10-100倍使用高峰的压力测试，趁资本充足时考虑融资，并提醒新员工注意加密货币的残酷周期。

“下跌的年份比上涨的年份更容易应对。信号与噪声的比例很高，好吧，价格下跌了，但从长远来看，这并不困扰我们。”

– Matt Huang, Paradigm

他的警告被证明具有预见性，但仍不足以应对一切。“我们在那个时期犯了很多错误，” Huang 反思说。 “当你过于关注竞争对手时，你会变得更像对手。” 他解释说，看着竞争对手a16z 募集巨额资金让他们质疑是否需要匹配那种规模。Paradigm 从18人增长到62人。 “我们确实让质量标准下滑了，” 他承认。“我记得有些时候做出了妥协，感觉如果我们不这样做或不雇用那个人，我们就会落后。事后看来，这些都是错误的决定。”

Huang 不是一个热衷于电子表格的人，他不记得公司从峰值到谷底的最大回撤，但有一个时刻深深印在他的记忆中：FTX。Paradigm 向这家交易所投资了2.78亿美元，这是公司历史上最大的投资之一。到2022年，FTX 已成为加密货币的公众形象，创始人Sam Bankman-Fried 在会议上演讲、在国会作证、登上杂志封面。那年10月，他在Paradigm 的有限合伙人会议上发表了主题演讲。几周后，FTX 因欺诈指控和客户资金被挪用的揭露而崩溃。

这次投资彻底失败，但背叛感更深。在尽职调查期间，Paradigm 已经识别出关键风险：FTX 与Bankman-Fried 的交易公司Alameda Research 之间的关系。团队提出了直接问题，却得到了虚假保证。后来Huang 在Bankman-Fried 的刑事审判中出庭作证，这段经历强化了关于创始人契合度的重要教训。

“即使在当时，也很明显他并不认同我们改善加密货币的愿景，” Huang 说。“对他来说，这是一个赚大钱然后捐出去的方式。”这种分歧在政策讨论中变得明显，Bankman-Fried 主张的妥协被Paradigm 认为会损害加密货币的核心承诺。

FTX 并非Paradigm 唯一的失误。该公司曾在NFT 热潮顶峰时与他人共同领投OpenSea 的3亿美元C轮融资，估值达133亿美元。此后，该NFT 市场的交易量下跌了98%。另一家投资组合公司BlockFi 因FTX 曝光而破产。

“在风险投资中，总会有一些投资不如你所愿，” Huang 说。“这总是一个反思的机会，我们已经做了很多反思。”他坚持认为熊市实际上比牛市提供了更清晰的信号。“下跌的年份比上涨的年份更容易应对。信号与噪声的比例很高，好吧，价格下跌了，但从长远来看，这并不困扰我们。”

公司从这段时期中走出来时规模变小但更加专注。2021年增长到20人的投资和研究团队被裁减到11人。标准收紧，并为新投资增加了一个明确过滤条件：创始人必须与Paradigm 推进加密货币前沿的使命保持一致。

网站事件以另一种方式提供了启示。迅速的负面反应表明，加密社区已将Paradigm 视为不仅仅是一家投资公司。它是行业的标杆。

书写未来

Georgios Konstantopoulos 的办公桌后放着一把迷你电吉他，偶尔在即兴会议中拿出来使用。想象公司首席技术官在讨论区块链架构时弹起吉他riff 的画面，捕捉到了他方法中的本质：技术上的精湛与对实用性的直觉结合。

2019年，Konstantopoulos 是一位备受追捧的研究员和软件工程师，因其开发技能在加密圈内闻名。他的技术工作如此彻底，以至于Paradigm 的投资组合公司不断提到他的名字。

Huang 第一次在会议上见到他时，Konstantopoulos 正在权衡是扩展他的咨询业务还是加入一家初创公司。但Huang 以他一贯发现非传统人才的能力，看到了不同的道路。他提议创建一个类似于Robinson 的角色，让Konstantopoulos 将技术研究与投资评估结合起来。

这个角色以意想不到的方式演变。2020年，在帮助投资组合公司Optimism 实施其研究时，他注意到许多项目都在为相同的基础问题挣扎。挑战不在于想法，而在于构建它们所需的工具。Konstantopoulos 想，与其逐一支持公司，他能否构建开源基础设施来推动整个行业前进。

人们告诉你很难的事情通常并不难。只是因为你无法掌控你的工具，它们才显得难。

– Georgios Konstantopoulos, Paradigm

这种哲学促成了他的第一个重大贡献Foundry。Konstantopoulos 花了一个周末构建了一个工具，大大简化了编写安全智能合约的过程。可以将其想象为区块链代码的拼写检查——只是它不是捕捉拼写错误，而是防止数百万美元的漏洞。几个月内，它已成为行业标准，现在拥有90%的市场渗透率，保障了超过1000亿美元的智能合约（迄今为止）。

但Foundry 的成功凸显了一个更深层次的挑战。以太坊，这个驱动加密货币大部分创新的平台，运行在低效的软件上，扩展几乎不可能。就像试图通过拨号连接流式传输4K视频。Konstantopoulos 提出了一个大胆的解决方案：从头重建以太坊的核心节点软件。

“你疯了，”他的团队说。“这需要五年。”但Konstantopoulos 赢得了他们的信任，他通过独特的招聘方式比任何人都更了解他们的能力。他没有采用传统面试，而是通过他们在开源项目中的贡献找到工程师。“代码不会撒谎，”他说。“我想看到人们如何思考真实问题。”

由此产生的项目Reth 只用了18个月就完成了。虽然它的功能看似简单——下载交易、在本地执行并写入数据库——但其影响深远。通过优化这一基本流程，Reth 的体积缩小了80%，速度比替代方案快10倍。Coinbase 的Base、WorldCoin 和Optimism（2022年最后估值16.5亿美元）等主要平台已依赖其卓越性能（它于2024年6月推出）。

这些技术贡献形成了一个良性循环。Paradigm 的研究人员在评估投资时发现问题。他们构建的开源解决方案成为行业标准。这些工具吸引了最优秀的开发者，他们要么加入投资组合公司，要么自己成为创始人，或者在某些情况下加入Paradigm。

这一策略在去年10月达到高潮，Ithaca 从Paradigm 分拆出来，获得了2000万美元的投资。作为首席执行官（同时保留Paradigm 首席技术官角色），Konstantopoulos 旨在将其团队所构建的东西商业化。“其他团队需要20-30名工程师、种子资金和两年时间才能做到的事情，”他指出，“我们几周就能完成。”

他的信心来自于构建了从底层加密到用户界面的每一层，中间有Reth 和Foundry。“人们告诉你很难的事情通常并不难，”他争辩说。“只是因为你无法掌控你的工具，它们才显得难。”这种激进自力更生的哲学——构建行业所需的任何工具——改变了Paradigm 在加密货币中的角色。

至于他自己的转变，Konstantopoulos 用典型的希腊式语言描述：“Matt 是我唯一无法超越的导师。”大多数导师最终都会被超越，但在Huang 身上，Konstantopoulos 找到了一位与团队共同成长的领导者。虽然像他这样水平的工程师大多会离开去创业，但他和其他人留在Paradigm，因为Huang 与他们一起成长。“他们每天都推动我成为更好的自己。” Huang 说，“我不想被超越。”

在Huang 的MacBook 背面，一个细节引人注目：三个贴纸形成一个完美对齐的行，Foundry 和Reth 的标志夹着Paradigm 的标记。这既反映了他对细节的关注，也透露出他对风险投资不断演变的愿景。 “如果红衫资本不仅支持了Google，还创立了Google 会怎样？”他越来越常思考这个问题。这个问题指向一个未来，投资者与建设者之间的界限变得模糊。

Konstantopoulos 从游戏迷到加密货币核心基础设施架构师的旅程，实现了Huang 和Ehrsam 的最初论点： 加密货币需要一种不同的投资者。 不仅仅是能评估技术的杰出异类，而是能通过代码塑造金融未来的建设者。在一个天生敌视中央权威的行业中，Paradigm 通过专注于创造而非控制，成为加密货币最受信任的机构之一。Huang 和他的团队不仅在投资未来——他们一行一行地书写着未来。