Brian Armstrong stepped out of his car, felt soft California sunshine on his bald head, and smelled eucalyptus. He gazed at the façade of Y Combinator: the one-story building, just five miles from Google’s Mountain View campus, looked more like a sleepy suburban office park than a famous startup school that had educated the founders of Stripe, Dropbox, and other billion-dollar companies. Brian didn’t care about the place’s humdrum appearance. He knew who had gone there before him. The founders of Airbnb, a company he’d just left, had come out of Y Combinator, and so had the CEOs of other Silicon Valley stars like Doordash, Twitch, and Reddit. Brian, pale and shy-looking at first glance, exuded a quiet confidence from his trim frame and wasn’t bothered that he’d broken up with his would-be cofounder just days before, making him the rare entrepreneur to do the program alone. It was the summer of 2012, and Brian was brimming with certainty that he would build Y Combinator’s next famous startup.
It wasn’t always this way. Twelve miles to the south, in San Jose, is where Brian had spent his early teenage years in the 1990s, restless and vaguely unhappy. San Jose is the tenth-largest city in the country and the hub of Silicon Valley, but it could still feel—then and now—like a lifeless parking lot where many people have nothing to do. Brian felt like that a lot. Until the internet.
As it had for so many other intelligent but introverted kids, the appearance of the World Wide Web brought friends to Brian as well as a flood of exciting ideas. Being stuck in poky San Jose didn’t matter now that he had a global community of hackers and philosophers at his keyboard. By the time he arrived at Rice University in 2001, Brian knew he wanted to use the internet to remake the world in the way an earlier generation of tech visionaries had done with microchips and desktop computers.
But there was a problem.
“I always had this thought, ‘I wish I was born a bit sooner.’ By the time I graduated from college and I was starting to work, I worried maybe I was too late,” Brian recalls. “The formative internet companies had been built, and the revolution had happened.”
He was wrong, of course. The internet revolution is still blazing, and entrepreneurs, for better and worse, are using it to remake our homes and our lives. And in late 2008, a mysterious person using the name Satoshi Nakamoto published a nine-page white paper on the web that would bring that same revolution to money. Brian discovered that paper a year later.
It was Christmas, and Brian was in his old room back at his parents’ house in San Jose, reading tech news on the internet, as usual. Someone had posted Satoshi’s paper on a computer discussion forum. Right away, he was rapt. He read and then reread what the paper described: a new type of digital currency known as bitcoin that operated outside the realm of any bank, company, or government. Bitcoin kept track of who paid whom just as a bank did, but the transactions were recorded by random people on computers scattered around the globe. It was real money without banks or borders. Brian began reading Satoshi’s paper a third time, ignoring his mother’s calls from downstairs to join the family for dinner.
Two and a half years later, as he walked through the doors of Y Combinator, Brian was more fixated on bitcoin than ever. By now, he had developed a special insight of his own about the currency, one that he would soon deliver to millions of people.
• • •
In his startup bible, Zero to One, mercurial billionaire Peter Thiel talks about “open secrets”—business ideas that are just there for the plucking by those who are not afraid to challenge conventional thinking. Thiel gives the example of Airbnb, whose founders saw a latent market for empty rooms, and Uber, whose founders realized it was possible to replace taxis with a GPS signal and a smartphone app.
The books of business writer Michael Lewis provide other examples of open secrets. In Moneyball, Lewis describes a general manager who built a winning baseball team by relying on data rather than the long-held wisdom of veteran scouts. And in Liar’s Poker, he recounts how a trader made a killing at his Wall Street firm by bundling home loans into mortgage bonds—an obvious idea, but a secret at the time because popular consensus dismissed it.
In 2012, Brian had grabbed an open secret of his own. He knew bitcoin could be a world-changing technology, but that buying it—for most people—was a confusing, convoluted experience. What if he could make it simpler? Y Combinator President Sam Altman understood the power of such simplicity and what Brian sought to do. “Making things easy to use is important to 99 percent of people, but technical people overlook that. When Dropbox launched, programmers would say, ‘I don’t get why anyone needs this when you can use these command line tools and make backups of all your files,’” he says, describing a computer process obvious to programmers but baffling to everyone else.
The same reasoning applied to bitcoin. More people would try it if only someone built a website where they could buy it the same way they bought stocks online. But the bitcoin devotees who could build such a site scoffed at the idea. They didn’t see the point. Instead, many sought to lift the technical principles of Satoshi’s paper and build a cryptocurrency of their own in hopes of getting rich. In Altman’s words, “Everyone in the crypto community wanted to start a new version of bitcoin. There was this mindset at the time of, ‘I’m going to get rich quick by making a new coin and keeping 20 percent for myself.’”
Brian saw it differently. Seizing on that open secret—the pent-up demand for easy access to bitcoin—he built a mockup of what would become the website Coinbase. And on August 21, 2012, Brian took the stage on Y Combinator’s Demo Day, a semiannual event where so many startups strut their stuff before venture capitalists and the tech press. It is a small moment of glory for the founders to savor before, inevitably, most flame out in the following months. That’s the ordinary fate of startups, but not all of them, including two other companies in Brian’s class: one was Instacart—now a billion-dollar grocery service—and the other Soylent, a meal-replacement product that’s since built a cult following in Silicon Valley and beyond.
When it was his turn to present on Demo Day, Brian stepped onto the stage with quiet confidence. He turned to the audience and shared his idea with the simple slogan: “Coinbase: The easiest way to get started with bitcoin.”
It seemed so obvious—in retrospect.
• • •
Brian’s early insight into bitcoin would make him a billionaire. But it would cost him a friend. In that summer of 2012, Brian had not planned on going to Y Combinator alone, where one-man bands were discouraged. The startup school wanted cofounders. Plural.
Despite Silicon Valley’s veneration of individual entrepreneurs, the reality is that tech startups, like so many creative endeavors, are very much a team sport—often a two-person partnership. In works like Collaborative Circles and Powers of Two, researchers have shown how genius is rarely solitary: John Lennon and Paul McCartney relied on each other to compose timeless Beatles hits; Pablo Picasso and Georges Braque used their brushes side by side to create Cubism; biologists James Watson and Francis Crick worked intensely together to discover the double helix and DNA.
Tech is no different. Apple is famously associated with Steve Jobs, but, in its early days, the computer company wouldn’t have gotten off the ground without the other Steve—Jobs’s partner and programming virtuoso Steve Wozniak. The same is true with Google. The Stanford graduate supervisor of Larry Page and Sergey Brin has remarked on the near total mind-meld of the search engine founders. And a garage in Palo Alto, known as the birthplace of Silicon Valley and now an official California state landmark, did not belong to a lone inventor but to two men: Bill Hewlett and Dave Packard, who founded HP.
Experience had taught Y Combinator’s overseers that a good cofounder is as important as a good business plan. “If you look at the history of successful companies, they’ve been founded by partners,” says Y Combinator’s Altman. “In our experience, it’s very, very hard to be a single founder. The ups and downs of a startup are so intense that you need to cheer each other up when someone is struggling.”
And right up until the start of the Y Combinator program, Brian had a cofounder. His name was Ben Reeves. A shy, young British kid, Ben was a programming wizard who believed in bitcoin with the same passion as Brian. The pair clicked upon meeting on a bitcoin discussion website. Before long, Brian and Ben made plans to start a company together. They applied to Y Combinator as a team, and the presti-gious school accepted them. But days before Ben was due to board a plane from the UK, the pair clashed on a key issue and Brian jilted him. “Cofounding is really like a marriage. Even though I think we have mutual respect for each other, we don’t work together extremely well,” Brian emailed Ben a few days before Y Combinator.
For good measure, Brian changed the passwords to the libraries of code they had been building together. In startup land, it was the equivalent of cutting off a spouse from a joint bank account. But it had to be done.
The point on which Brian and Ben had disagreed wasn’t an aesthetic one or even a strategic one. It was an existential one. Their dispute turned on a near-religious clash about what bitcoin was supposed to be.
When the pseudonymous Satoshi Nakamoto revealed bitcoin in his nine-page paper, he described the invention of a new and decentralized technology. That word, decentralized, is critical. It meant no single person, company, or government could control the network on which bitcoin is built. Meanwhile, people who bought and sold bitcoin could not rely on a bank or anyone else to manage their stash of digital money. Owning bitcoin meant using something called a private key—a long gobbledygook string of letters, numbers, and symbols—that opened and closed your online wallet. If a person lost that key, it was gone forever. It was the digital equivalent of a pile of cash in an unbreakable safe to which no one knew the combination.
That’s where Coinbase came in. Brian’s idea—the open secret he seized on—was to provide a service where you could own bitcoin without controlling a private key. Coinbase would do it for you.
It was a commonsense solution. But bitcoin purists saw it as heresy, against everything Satoshi stood for. It didn’t matter that customers could use Coinbase to buy bitcoin and then transfer it to a wallet they controlled with a private key. It was a matter of principle. In the eyes of the purists—the overwhelming majority in the crypto community back in 2012—Brian and his vision of Coinbase stood for the c-word: centralization. He was a heretic and a traitor to Satoshi’s vision.
Brian and Ben never reconciled. Ben would go on to build a successful bitcoin company of his own, but he has never forgotten how Brian had jilted him. Years later, he allowed Wired magazine to publish verbatim the text of Brian’s break-up email. His LinkedIn page still reads, “Coinbase founding team member.”
Brian today plays down the rift. The divorce with Ben came at the prodding of a senior executive at Y Combinator, and Brian believes it was necessary. But at the time, it was also a major problem. As a result of his last-minute breakup with Ben, Brian became the rare entrepreneur to go through Y Combinator as a single founder. In doing so, he had reaped the accelerator’s coaching experience and could tap into its fantastic Rolodex of mentors and investors. But he had no one to cheer him up or encourage him when things got hard. And they were about to get very hard.
While Y Combinator offered prestige and publicity because of the small number of companies it accepted into its fold, acceptance was not the same as success. The reality was, after the program’s much-hyped Demo Day, over 80 percent of the startups quietly ran out of money and turned to dust. And those companies typically had two or three founders pulling out all the stops. In the summer of 2012, Coinbase was little more than a marketing idea and an unfinished website with a single founder. The company needed much more to get off the ground—millions more lines of code, product testing, a business plan and, of course, real-life customers. If Brian couldn’t pull this off, Coinbase would share the fate of most startups: failure. Brian’s odds were grim.
• • •
Five miles south of Y Combinator in Mountain View is another Silicon Valley town called Sunnyvale. It has the same soft air, eucalyptus scent, bland suburban streets, and a stop for the Cal-Train, the region’s poky commuter rail service. It’s home to dozens of notable tech companies, including Atari, Yahoo, Palm, and the chip maker AMD. That same summer of 2012, it also become home to a young Wall Street refugee named Fred Ehrsam.
Fred was one of those golden kids everyone knows in high school. He had a model’s good looks—a chiseled face and a flop of blond hair—and he radiated swashbuckling athleticism. Growing up in Concord, New Hampshire, he had run with the popular crowd—of course he did—but it never felt right.
“I felt like an observer of my own life,” he says. He did what he was supposed to do: got good grades, excelled at lacrosse and basketball. The desire to please his father gnawed at him. Fred’s dad was a hard-charging engineer who had graduated from Harvard Business School and expected the world. Years later, staring out from a magnificent penthouse with bountiful views of the city of San Francisco and the ocean beyond, Fred still didn’t know if he measured up. “Even if you’re very good at a video game, the levels keep getting harder and harder,” he said wistfully.
Fred’s choice of metaphor is fitting. Video games are something he knows better than almost anyone else. Although the world around him in high school never felt right, the one he found on the internet sure did. Every day, he would leave lacrosse or basketball practice as soon as he could and rush to play World of Warcraft or Call of Duty, often staying up all night so he could stay competitive in two online leagues—one in the US and another in Europe. By the time he was a senior, he was a professional gamer, entering and winning tournaments around the country.
Video games gave Fred an escape from the pressures of high school and family life, but only a temporary one. Soon enough, it would be time to get a college degree, which he earned as a computer science student at Duke University, and then it was time to make a respectable living. And he did, taking a job as a foreign exchange trader at Goldman Sachs. “Being a forex trader at Goldman Sachs was the closest I could get to playing a video game in real life while also having a job that came with money and prestige,” he admits.
Fred looked the part, and he was good at the job. That didn’t mean he liked it. In fact, he was dying inside. His bosses at Goldman Sachs were old-school Wall Street types who had come up bellowing into telephones and jostling with other men in trading pits. And they didn’t like the new style of trading that was creeping into the finance industry, one that largely rewarded those who wrote the best algorithms. The prophecy of the famous West Coast venture capitalist (and future Coinbase board member) Marc Andreessen, “Software is eating the world,” was coming true. And it was going to swallow up those old-school traders. Even if they didn’t want to admit it.
“They called the software engineers ‘IT’ and treated them as second class,” Fred recalls. “They had this aversion to automation. If I wanted to do something that could replace half the trading desk, they didn’t want that. It was a very bizarre time.”
It was like high school all over again. On the surface, Fred looked and acted the part of a hotshot trader, and he was pleasing his parents, but deep inside he wished he was anywhere else. So he responded as he had back then, taking refuge late at night on the internet, discovering people and worlds and a place he belonged. This time, he became transfixed by blogs and Reddit threads about a new digital currency that anyone could access without a central bank—or, for that matter, a merchant bank like Goldman Sachs. Bitcoin, a currency free of governments, wasn’t just an intriguing idea, Fred felt. It was a necessary one. Day after day, he watched Wall Street gorge itself on Federal Reserve funds. The situation overseas was even worse—countries like Greece fumbled from bailout to bailout as a result of epic mismanagement by political leaders. In contrast, the once-crazy concept of bitcoin looked sane. Also, Fred saw in bitcoin a job for which he was born: he knew about digital money from years of using video game currency, and he knew about finance as a Wall Street trader. He wanted in on bitcoin.
There was just one problem. All of the action appeared to be taking place in Silicon Valley. This was a place he’d heard about, of course, but growing up in New England, he didn’t grok what it was all about. Gradually, though, he came to realize—just as painters flocked to Paris and moviemakers to Hollywood—Silicon Valley was where you went if you wanted to do great things with software. Even New York City, which supposedly had everything, didn’t offer that particular mix of business hustle and computer science wizardry. It was time to go. After two years at Goldman, Fred took leave of the tall buildings of Wall Street and struck out for suburban Sunnyvale.
• • •
Fred and Brian met at The Creamery. Like so many other famous Sili-con Valley venues, The Creamery doesn’t look like much: a low-slung, single-story wooden building with white letters above the doorframe; a small patio; some seat-yourself indoor tables; a menu of breakfast sandwiches, salads, and the usual assortment of cocktails and cappuccinos. It’s a modest place on a nondescript San Francisco street corner, yet its walls have heard billions of dollars’ worth of venture capital deals and countless startup pitches for massive successes and failures alike.
Maybe The Creamery is popular because it’s right near a freeway off-ramp and a Cal-Train station. Maybe it’s because patrons can walk right in and out, with no fuss. Or maybe it’s just because tech people have always met there. (Its numerous well-heeled customers could not help The Creamery survive the pandemic, however. The famous establishment closed in August of 2020.)
In Brian’s case, he chose The Creamery because it was right across the street from the makeshift office he had rented at 1 Bluxome Street. He had wrapped up at Y Combinator a few months before with a bulging list of contacts and potential investors, while the startup school—as it does with everyone who enrolls—took 7 percent of his company. Still, Brian was very much alone, professionally and personally, when Fred replied to one of his bitcoin threads on Reddit.
Fred had left Sunnyvale a few weeks before, where he had been bunking with old college friends, and was now living in San Francisco. When he met Brian, it was like one of those rare Tinder dates that actually clicks. “Something felt right in my gut about this. It just felt exciting,” Fred recalls. This emerging company called Coinbase felt like a rollicking video game he had never played before. But it was real.
The bromance between the mid-twenty-somethings was mutual. If Brian had had cold feet about a startup marriage with Ben, this time he was ready to jump in quickly. In Fred, he had found a cofounder, a friend, and a fellow fanatic. Together, they bashed their keyboards around the clock, often working sixteen-hour days as they struggled to compile the code that would let people do what hadn’t been done before: acquire bitcoin simply by providing a bank account number. No overseas wire transfers, no intimidating mathematical strings—just a basic website that felt like online banking.
It had been nearly four months since Brian had taken the stage at Y Combinator. Now, in November of 2012, it was time to see if Coinbase was for real. It was time to launch a feature to buy and sell bitcoin with one click. A whisper of San Francisco fog sat outside the window as Brian and Fred huddled anxiously over a laptop as the feature went live.
• • •
Success!
A trickle of customer orders came dribbling into the website. Weeks later, it was a stampede. Word got around about this new and easy way to buy bitcoin. Volume increased, and so did their workload as Brian and Fred struggled to keep the site up and running.
The first crisis came when a software bug skewed the appearance of customers’ bitcoin balance. On the Coinbase side, things were fine—the bitcoin was there—but for some customers, it looked like they had been wiped out. Coinbase’s crude customer service portal flashed with dozens, then hundreds, then more than two thousand frantic requests from panicked clients.
“Where the hell is my bitcoin?” “Is this a scam?” “Give me my money back!” The anxious, often abusive, invective kept pouring in. It was a critical moment for a fragile startup with an even more fragile reputation in an industry fraught with distrust. Brian and Fred worked around the clock, taking turns sleeping on the floor while the other beat back the cascade of customer requests and repaired the bug.
Finally, following hour after hour of exhaustive coding, the fire was out and the site was fixed. Coinbase’s credibility was restored. Brian, calm as ever, turned back to reading tech news. Fred, too frugal to take an Uber, stumbled toward home in San Francisco’s notorious Tenderloin district, whose streets jangled with broken glass and the screams of junkies. Fred passed through it all oblivious. At one point, he shuffled for two blocks behind a blind man who staggered pitifully down the wretched sidewalks.
Finally, Fred found his way into his bed. People outside were still stirring.