Revealing Matt, the founder of venture capital firm Paradigm: Met Zhang Yiming at 24, invested in Toutiao and made 2000 times profit

This article was published in 2019 on the public account, "Random Book Flipping," with the original title: "He Came to Beijing for a Tour from Silicon Valley at 24, Met Zhang Yiming, and Invested in Toutiao, Now Earning 2000 Times," by Cheng Tianyi, reprinted by Mars Finance.

As one of the top venture capital funds in the crypto industry, Paradigm has recently made swift and precise moves, not only earning tens of billions of dollars from a single investment in Uniswap but also investing in Cosmos, Mina, and in the DeFi sector, investing in MakerDao, Synthetix, and leading rounds in dydx, Opyn, and the staking protocol Lido. They even invested tens of millions of dollars in the Ethereum scaling solution Starkware. In the NFT sector, they have made significant investments in ZORA, Showtime, and others. While aggressively laying out their strategy across various blockchain tracks, it’s time to understand the founders behind it.

Last year, Random Book Flipping revealed that Toutiao's angel investors were Liu Jun, Zhou Zijing, and Wang Qiong, representing SIG (Haina Asia).

In January 2013, Zhang Yiming disclosed during a visit to NetEase Mobile that Toutiao had secured its first round of financing of $5 million in November 2012. This round of financing came from SIG (Haina Asia Venture Capital Fund) and Wall Street economists, investment bankers, and Silicon Valley startup CEOs. In fact, the financing was finalized in March 2012, but due to some policies regarding the VIE structure, the funds did not arrive until the end of the year.

In 2012, Toutiao's A round valuation should not have exceeded $30 million. Now, Toutiao's valuation is $75 billion, a 2500-fold increase in just six years.

Considering dilution, A round investors may have achieved a 2000-fold return on their investment in Toutiao.

As for the Wall Street economists, investment bankers, and Silicon Valley startup CEOs who invested in Toutiao in its earliest stages, their identities have remained unknown.

Until last week, during ByteDance's seventh anniversary celebration, Zhang Yiming shared a financing story:

About six months after the company was founded, an American entrepreneur visiting Beijing came to tour the company through a friend's introduction and discussed many product technologies. He was surprised and said that the team's technology in this apartment was on par with Silicon Valley. Could he make an investment? Later, he became our investor. ------ ByteDance's seventh anniversary celebration

Everyone knows that Toutiao's B round was invested by a Russian named Yuri. In fact, the A round was also captured by an American team (besides SIG). During Toutiao's most difficult B round financing phase, Zhang Yiming explained what he was doing to over 30 domestic VCs, but no one understood. Ultimately, it was the American investment team from the A round that helped connect him with the Russian investors for the B round.

So who are they?

This article is a sequel to Toutiao's financing story, further exploring the entrepreneur who came to Beijing for tourism, his legendary father, and the story behind the institutions involved.

Silicon Valley Startup CEO Matthew Huang

Matt Huang

To introduce Matt Huang before he turned 24, we need to mention his legendary father.

In 2008, the bankruptcy of Lehman Brothers triggered a global financial storm, taking down a number of shadow banks, one of which was a fund called Platinum Grove Asset Management. The partners, being older, chose to hand over the fund to younger individuals rather than spend another 10 years battling in the new bond market.

One of the retired founding partners was Huang Qifu. He held a PhD in finance from Stanford University and a PhD in finance from MIT. At 34, he secured tenure at MIT, leaving academia for the industry at 35, and over the next 15 years faced numerous ups and downs, earning and losing vast amounts of money. (The story of this remarkable figure can be found in Chapter Two of this article.)

Huang Qifu's retirement life was as follows:

So what do I do after retirement? I first spend more time with my sons, which I am most proud of. I have three sons, and I spend a lot of time with them. Otherwise, before retirement, I worked 24 hours a day, six days a week, and had basically no time. So I have a good relationship with my two younger sons; I talk to them every day. ------ Huang Qifu's speech at National Taiwan University in 2014

The eldest son, who did not talk to his father every day, is Matt Huang.

Huang Qifu's family photo

Even so, Matt Huang's growth experience shows that he had guidance from his family.

For example, he obtained a bachelor's degree in mathematics from MIT, while his father earned a bachelor's degree in economics from National Taiwan University. After Huang Qifu went to Stanford, he found that the mathematics education at National Taiwan University was very poor, having only taken calculus in his freshman year. Therefore, during his graduate studies at Stanford, he took mathematics courses alongside undergraduates and spent all his time in the mathematics department in his third year.

Of course, a background in mathematics also has its drawbacks. Huang Qifu recalled: "Stanford changed my destiny, allowing me to learn very complex mathematics in a short time and use it to solve difficult problems. Because I first asked an economic question and then looked for tools. People from a mathematics background have many tools but don’t know what questions to ask."

Thus, during his college years, Matt Huang spent a lot of time dealing with good questions.

Goldman Sachs New York Office

In 2008, during the summer of his sophomore year, Matt interned for three months at Goldman Sachs in New York, working on real-time bond price prediction.

In 2009, during the summer of his junior year, Matt worked as a researcher at MIT's economics department, studying financial crises, Knightian uncertainty, and flight-to-quality. His father's fund had just encountered a financial crisis a year earlier.

After graduating from college, Matt did not choose to stay in Boston to delve deeper into academia, nor did he go to New York to join Wall Street in finance.

Instead, he partnered with two classmates majoring in computer science—Alexander Spicer (a classmate of Matt, now an engineering manager at Flexport) and Ashutosh Singhal (a junior who interned at Google)—to start a tech company in California, becoming a "Silicon Valley startup CEO."

This company was incubated by Y Combinator W11 and was initially called Moki.tv, aggregating Netflix, Hulu, and Amazon, serving as a "personalized guide for online streaming video."

It later transformed into a media advertising analytics product, known as Hotspots.io.

According to early media reports, Hotspots primarily provided services such as measuring the media impact of Super Bowl ads, scraping the number of mentions of various companies on Twitter, and overall social coverage. It also tracked advertising spending by large companies like Coca-Cola, H&M, and Anheuser-Busch to determine whether their marketing was yielding returns.

In 2012, the content recommended to users by Toutiao was all generated through machine-driven social interest mining, calculation, and release. Users only needed to bind a certain SNS account, and the program would discover user interests in 3 seconds, analyze interest DNA in 5 seconds, and actively understand users' "likes," without requiring any subsequent user actions or the need for users to think about what they liked to "hot start recommendations."

This logic of achieving precise recommendations through social networks for cold starts was something Matt Huang, who was engaged in social media advertising analysis, could easily understand.

So when he came to Beijing for tourism in 2012 and was introduced to Zhang Yiming, hearing the story of Toutiao, his reaction was, "I was surprised and said that the technology of the team in this apartment was on par with Silicon Valley." Ultimately, he became an early investor in Toutiao.

In 2011, Twitter's advertising revenue was only $259 million. According to a 2012 report by LostRemote, Facebook's user base was only 1.69 times larger than Twitter's, yet its advertising revenue was 22 times higher. Therefore, due to the need for talent and advertising analysis tools, in April 2012, Twitter acquired this startup, with the transaction amount remaining undisclosed.

Twitter's advertising products

Including Matt, three founders joined Twitter's advertising department, reporting to Adam Bain, the head of Twitter's commercialization team. Bain later became Twitter's COO and left Twitter in 2015 to become a personal angel investor, achieving great success and becoming one of the investors in Reddit.

After the acquisition, Matt was responsible for advertising-related business in Twitter's revenue engineering team, creating analytical products for advertisers and VIT.

The investment story of ByteDance mentioned at the beginning of this article also took place during this period—Matt began making early investments in 2011, building a luxurious portfolio:

• Toutiao/Bytedance, entered between A and B rounds at a price of tens of millions of dollars, now Bytedance's valuation has exceeded $75 billion
• Instacart, a U.S. fresh grocery delivery service platform, raised $600 million in Series F funding led by D1 Capital Partners in October 2018
• PlanGrid, a construction drawing management application service provider, was acquired by Autodesk for $875 million in November 2018
• Benchling, a U.S. cloud platform for DNA research management tools, raised $14.5 million in Series B funding led by Benchmark in June 2018
• ……

Among these projects, the most successful is undoubtedly Toutiao. Matt discovered this project during his trip to Beijing in 2012, subsequently persuading his father to invest, leading to a significant additional investment in Toutiao's A round, which greatly helped with Toutiao's challenging B round financing.

On February 19, 2014, Fortune magazine reported an exclusive story stating that Sequoia Capital had "stolen" talent from Goldman Sachs and Twitter, bringing in a pair of junior partners to assist the fund in growth equity practice. At that time, Sequoia was raising $900 million to $1 billion for its sixth growth fund.

Andrew Reed came from Goldman Sachs, where he worked in TMT-related investment banking, while Matt Huang came from Twitter.

Matt worked at Sequoia Capital for four years, during which he led multiple investments in blockchain startups, leaving Sequoia in June 2018 to create the cryptocurrency fund Paradigm.

Sequoia Capital also invested in this fund, and Sequoia partner Alfred LinkedIn stated to the media: "Matt will still be a good friend of Sequoia, and we wish him great success."

In October 2018, Yale University, which has the second-largest endowment fund in higher education (with a scale of $30 billion), was revealed to have invested in Paradigm, which planned to raise $400 million for its first fund, making it one of the largest cryptocurrency funds in the U.S.

Matt is a partner at Paradigm, which has connections to his father Huang Qifu and co-founder from Coinbase.

Wall Street Economists and Investment Bankers, Huang Qifu and John Meriwether

Huang Qifu entered the industry in 1993 and shifted from Goldman Sachs to hedge funds in 1995, managing two funds—one managing a trillion dollars by 1997 and another that quickly became the industry leader but faced consecutive world-class failures.

He had a smooth sailing academic career in his early life, moving from Stanford's doctoral program to MIT at 28, obtaining tenure at 34, and becoming a full professor at 35. That year, he authored the classic textbook "Foundations of Financial Economics," which has been used by Tsinghua and Peking University. At 36, he became a chair professor.

After retirement, he made a comeback in early-stage investments, achieving a 2000-fold return on one case with Toutiao.

Huang Qifu's "Foundations of Financial Economics"

The fund, which became immensely wealthy, was called Long-Term Capital Management (LTCM), a hedge fund primarily engaged in bond arbitrage trading, initiated by Wall Street legends and geniuses in 1993:

The founder John Meriwether is known as the father of bond arbitrage on Wall Street, and prominent figures such as the former U.S. Deputy Secretary of the Treasury and former Vice Chairman of the Federal Reserve were also partners in the company. One of my (Huang Qifu's) other mentors, who won the Nobel Prize in Economics with Merton and is regarded as a financial master, Myron Samuel Scholes, was also here. ------ The Fall of the Father of Financial Models

In 1995, these geniuses sought to find a Chinese person to bring their trading strategies to Asia. Huang Qifu, due to his outstanding work in derivatives trading, caught their attention:

"The founder called me and said, 'Qifu, I'm in New York this afternoon; can we have dinner together?' The founders of Long-Term were basically among the top in Wall Street, very famous and influential. When you get a call asking to have dinner with him, you just have to go. When we sat down for dinner, he didn't talk about business at all; he only asked about my family and children, discussing personal matters all night, and then we went home."

Huang Qifu (far left), JW (far right)

However, Goldman Sachs had already promised Huang Qifu that he would become a partner by 1996. Huang Qifu's wife, a computer PhD from Caltech and a professor at Yale, was also not keen on returning to Asia.

Thus, this dinner continued for several weeks, and eventually, his wife agreed to let him try, so Huang Qifu left Goldman Sachs to become a member of the Long-Term Capital "dream team," responsible for the Asian market in the Tokyo office.

In the following two years, Huang Qifu helped Long-Term establish its Asian market. He used an idea to turn the derivatives and bond trading that had been losing money in Japan into a profitable business, and as a result, he was promoted to partner in 1997.

At that time, Long-Term managed assets worth $1 trillion (mostly financial derivatives), while Soros' Quantum Fund peaked at only $22 billion, and Bridgewater managed only a few hundred million dollars. In 1997, China's GDP was only $961.6 billion, highlighting the enormous financial leverage and terrifying scale.

Scenes from the Russian financial crisis

However, in 1998, Russia declared bankruptcy, and no one came to its rescue. Long-Term Capital, due to its excessive leverage and firm belief that "other countries would step in to help Russia," suffered heavy losses in this crisis:

Russia declared bankruptcy, causing many international banks to incur losses. They held an emergency meeting overnight to sell assets for cash. These large banks held many easily sellable assets, primarily bonds from the G7 countries. Our disaster stemmed from this because most of the financial assets we held were also G7 bonds. A massive sell-off of G7 bonds ensued. On August 21, international banks began to sell off their G7 bonds significantly, causing the largest fluctuations in bond prices in history. In the ensuing sell-off, LTCM lost nearly two years of hard work in a single day. ------ The Fall of the Father of Financial Models

On September 23, the Federal Reserve intervened, organizing 15 institutions led by Goldman Sachs, Merrill Lynch, and J.P. Morgan to inject $3.725 billion to purchase 90% of LTCM's equity, jointly taking over LTCM and preventing its collapse. This was the only time in history that the Federal Reserve intervened in a hedge fund.

After this disastrous defeat, Huang Qifu lost all his savings invested in Long-Term Capital. However, he quickly teamed up with Scholes and others to establish Platinum Grove Asset Management, starting from scratch and strictly controlling leverage. The fund initially managed only $45 million, and the fixed costs were high, leading to almost daily losses in the first year relying solely on management fees.

With the extraordinary talents of Huang Qifu and Scholes, the fund's scale grew rapidly, reaching $5 billion by 2007. The busyness of the business can be glimpsed from his account:

"Before the financial crisis, our fund was the largest in the world doing this, and we only had one office with two locations. In Beijing, we hired a lot of programmers, but all trading was done in the U.S., and our company operated basically 24 hours a day. Around two or three in the morning, we worked with European traders, then at 7:30 AM, the U.S. team came in, and in the afternoon, the Asian team joined in."

In 2008, the bankruptcy of Lehman Brothers triggered a global financial crisis, and Huang Qifu's new fund repeated the mistakes of Long-Term Capital—they believed that since the U.S. government intervened to save Bear Stearns, it would do the same for Lehman Brothers, rather than letting it fail.

However, tragedy struck, and Platinum Grove Asset Management's net worth rapidly diminished. Huang Qifu and Scholes announced their retirement after this battle, allowing younger individuals to adapt to the newly formed post-disaster bond market.

After retirement, besides spending time with his children, Huang Qifu and other current or former finance professionals established Starling Ventures in 2011 to invest in early-stage and growth-stage tech companies, with a portfolio including stars like Toutiao, Zuiyou, and Coinbase, achieving a 2000-fold return on the Toutiao case alone.

His eldest son Matt Huang's new fund co-founder, Fred Ehrsam, is a co-founder and board member of Coinbase, where he worked for six years overseeing daily operations. Coinbase is also part of the Starling VC portfolio.

Toutiao's ability to attract attention from DST and Yuri, completing its B round financing in September 2013, was also aided by Huang Qifu and Starling VC.

Sequoia's Self-Reflection and Shen Nanpeng

It is unclear how the information-sharing mechanism between Sequoia China and Sequoia U.S. operates. In any case, Matt Huang joined Sequoia U.S. a bit later; he left Twitter to join Sequoia in February 2014, thus missing the opportunity to share his experiences and insights with Sequoia China and Shen Nanpeng earlier.

The person who discovered Toutiao and actively promoted it at Sequoia China was Cao Yi, who is a few years older than Matt Huang.

Cao Yi later revealed the details of Sequoia missing Toutiao's B round. Toutiao's B round closed in September 2013, and although it had been approved internally at Sequoia, it was ultimately abandoned due to price factors. At that time, grassroots entrepreneur Zhang Yiming did not get to meet investment mogul Shen Nanpeng.

Cao Yi's recommendation for Toutiao's B round was not approved internally at Sequoia, and later he left to start Source Code Capital, becoming the only personal investment company for Zhang Yiming.

An interesting Easter egg is that Source Code Capital, founded by Cao Yi after leaving Sequoia, shares the same English abbreviation as Sequoia China, both called SCC, often humorously referred to as "Little Sequoia."

Shen Nanpeng later shared his experience of missing Toutiao's B round on various occasions, bravely facing this miss and re-entering eight months later at a tenfold increased valuation, knowing that at this point, Toutiao had almost no revenue.

Sequoia Capital China founding partner Shen Nanpeng

When I met Zhang Yiming, I was a bit drowsy because what he wanted to do carried enormous risks. He wanted to aggregate news, stories, and pictures through the internet and push them to users based on their different preferences. No one in the U.S. had tried this approach, but he believed there was space for such a product. I made a huge mistake. I didn't invest during his first round of financing. Why? I did a lot of due diligence. As investors, we rationally visited many of his competitors and found that all the big companies wanted to do this product—Sina wanted to do it, Sohu wanted to do it, Xiaomi wanted to do it, and Tencent wanted to do it. So after discussion among our partners, we felt that the market competition was too fierce. As a small company, there was no opportunity. Sometimes we investors are too smart, thinking too rationally. So I passed on his A round financing. In the following nine months, Toutiao proved us wrong. In fact, I have always regretted not participating in that meeting with Zhang Yiming. It seems we all lacked this judgment. Of course, if more people had participated, perhaps we could have made more complete or accurate decisions. Missing out is always a regretful and particularly painful thing, but at the same time, you know this is part of investment life. In a sense, the most interesting part of investing is that you will always have many misses. Even if I had invested in Toutiao back then, I would still feel I invested too little; you can always think this way. Venture capital is indeed a game of regrets. This is a mistake, but such mistakes may be unavoidable for a fund. -- "Meeting the Big Shots" Interview with Shen Nanpeng

This is also what makes Sequoia admirable.

Due to Shen Nanpeng's hunger for success, Sequoia possesses both the courage to always be at the forefront and the ability to leverage its management scale to salvage cases like Toutiao, allowing it to invest in the most popular cases in the market while also daring to invest in cases that most people do not understand.

This strategy of daring to enter at relatively high prices has not only succeeded with Toutiao but has also been repeatedly validated in other cases, showcasing the advantages of Sequoia's large management scale and the reasons for its large management scale.

References:

1. The Fall of the Father of Financial Models

2. What Twitter's acquisition of 'Hotspots.io' means for social TV

Some content is organized by Random Book Flipping from Huang Qifu's speech at National Taiwan University.