At the start of 2022, Neel Somani was a quantitative research analyst working at Citadel, billionaire Ken Griffin’s powerhouse hedge-fund firm, which manages about $47bn. But at age 24, Somani quit in late February with dreams of getting rich in crypto and helping to build a more decentralised financial system.
Somani chose to focus on Terra, one of the most popular blockchain networks, and its related cryptocurrency, USDTerra, ranked among the 10 largest cryptocurrencies by market capitalisation at the time.
Last week, the $50bn Terra blockchain collapsed, handing investors massive losses and crushing Somani’s dream just weeks after he started to pursue it.
USDTerra, or UST, was known as a stablecoin, meaning it was structured to always trade one-to-one against the US dollar. But starting on 9 May, UST fell below $1. The fall rapidly accelerated, with the coin trading as low as 5 cents at one point. The price of its sister coin, Luna, which backed UST, plunged to close to zero in less than a week, from over $80 in early May. Along with the price drop, the Terra blockchain was halted twice on May 12, once to “prevent governance attack,” tweeted Terraform Labs, which backs the blockchain.
More than $48bn of market capitalisation in UST and Luna evaporated in a week, and the impact was felt broadly among cryptocurrencies. To defend UST’s peg, Luna Foundation Guard, which supports the stablecoin, spent over 80,000 bitcoin in its reserve, adding selling pressure to the cryptocurrency, which recently changed hands at levels some 56% lower than its all-time high.
As recently as early April, Somani was bullish on UST. He tweeted on 5 April, “I quit my job at Citadel to build a project in web3!” On 13 May, he retweeted the same post, with a touch of self-mockery, “I quit my job at Citadel to get wrecked in web3.”
Somani, who graduated from the University of California, Berkeley, with a triple major in computer science, mathematics and business administration, had worked at Citadel as a quantitative research analyst, with a focus on commodities. He also once worked at Airbnb as a software engineer, according to his LinkedIn page.
On a recent video call with MarketWatch, Somani, sitting in his apartment in Chicago, seemed calm, though he said he is now “essentially unemployed,” and had to scrap his newly started venture and lost about $20,000 from his investments in Luna.
Still, he doesn’t regret his decision. “I recognise this was a setback, but also it comes [with] the territory. I knew what I was getting myself into when I quit my job to do crypto,” Somani said in the interview.
When Somani left his job to start his own business, his parents hated the idea. “Because they saw how much I was making at Citadel, and they were confused why that was not enough,” he said. He got more support from his friends, who are mostly at his age. Some had become “very successful” in crypto.
Somani declined to reveal his salary at Citadel. The average annual compensation of a quantitative researcher at Citadel comes to over $400,000, according to Glassdoor, based on 176 salaries including 11 that reported cash bonuses.
Somani said he made the move to crypto because it was more interesting, and “there’s also the potential to make more money.”
“Crypto seemed like a good intersection between my interests and what I wanted to do with my career,” Somani said, adding that he has always wanted to become an entrepreneur. “I just saw this cycle of feast and famine, where people made a ton of money and lost everything. I wanted to take on more risk in my career, honestly. So I thought this is the right field.”
In fact, a few years ago, when Somani was still in high school and was given free fractions of bitcoin at hackathons, he threw them away. “I thought they were useless at that time,” he said.
During the past two years, especially after seeing how the cryptocurrency was used amid Russia’s invasion of Ukraine, he recognised that “we actually do need some kind of decentralisation to prevent big authorities or big corporations from basically blocking access to financial freedom,” Somani said.
Despite Terra’s collapse, Somani has not been deterred from trying to make it big in crypto. He is already drafting his next business idea, still in crypto, and is preparing to announce it in the coming weeks.
Of course, Somani is not the only true believer who has left a well-paying, stable job in more traditional finance, or tech, to work in crypto. Hiring has been fast, as the nascent industry evolves quickly. In 2021, crypto hires surged 73% from two years earlier, according to a study by LinkedIn Economic Graph published in April. In contrast, the number of hires in traditional finance declined by 1% over the same period.
Somani was not, however, oblivious to the risks he and others were taking. He understood the key issues around Terra. In a blog post published on April 5 where he outlined his business plan, Somani highlighted that “history will view algorithmic stablecoins as either a) a disaster as inevitable as the subprime mortgage crisis or b) the greatest recent innovation in financial history.”
He pointed out the risk of a “death spiral,” a concern that had been shared by some other critics of Terra.
Based on Terra’s design, investors are supposed to be able to exchange one UST for $1 in Luna, and vice versa. In relative stable market conditions, when UST is trading below $1, traders have an incentive to buy one UST and exchange it for $1 of Luna to make a profit. In theory, as UST is burned to mint Luna, the former’s supply would be reduced and its price pushed up back to $1. That’s how the stablecoin is supposed to maintain its peg to the US dollar.
However, the design had weaknesses. During broad selloffs, when a massive amount of UST is sold, the stablecoin could fall below $1. As arbitrageurs are incentivised to burn UST and mint Luna, the latter’s supply could rise and its price fall sharply. Such a development could, in turn, shake investors’ confidence in the ecosystem and further reduce demand for UST. This is what appears to have occurred last week, analysts noted.
“Algorithmic stablecoins are based on confidence and trust in the economic incentives of the stablecoin issuer’s underlying ecosystem. Once that trust and investor demand evaporates, they quickly fail in a death spiral,” Ryan Clements, a professor at the University of Calgary who has conducted research on algorithmic stablecoins, earlier told MarketWatch.
Somani argued in his post that lending protocol Anchor, which is based on Terra and pays interest of up to 20% to users on crypto deposits, was the key reason that drove up the demand for UST. But such an interest rate was unsustainable, as it was powered by the protocol’s reserves, which could eventually be drained, Somani noted.
“If I were a trader, at the time I wrote my post I would have immediately short UST or short Luna, but instead I am a builder, so I thought, let me try to fix it and create those use cases for UST to prevent it collapse,” Somani said.
He originally thought it would take a year for Terra to fall down and “we could potentially save it before that happened.”
Somani, at the time, proposed to create more use cases for UST through building an “Ethereum Virtual Machine” on Terra, which could help bring the top decentralised applications on Ethereum, the most popular smart contract blockchain, to Terra.
Somani was initially attracted to Terra by its vision to build a stablecoin that was decentralised, instead of being issued by centralised entities, such as in the case of Tether and USD Coin. Also appealing was its fast growth. Terra, which was created in January 2018, became the second largest blockchain in December 2021 for decentralised finance protocols in terms of total value locked, behind only Ethereum.
“I thought that the algorithmic stablecoin project was first very intellectually interesting that you could construct a lot of interesting financial derivatives using them. And also there’s a need for decentralised money,” Somani said.
In fact, Somani liked algorithmic stablecoin so much that he once asked Citadel if he could receive a paycheck in Bean, another stablecoin powered by algorithms. The request was rejected, as the Chicago-based hedge-fund firm said it could only issue paychecks in US dollars. In April, Beanstalk, a protocol that backs Bean, was exploited for $181m by attackers.
Though Somani still believes that there is a need for decentralised, algorithmic stablecoin, he is not sure if they will be viable and whether they could avoid the fate of UST and similar projects.
On 7 May, when UST first lost its peg, Somani got texts from some friends who had money in Anchor, asking him what was going on. “Oh, relax,” he said. Somani did not think the development was a big deal because Terra had fallen below $1 before and quickly restored the peg.
Some of his confidence stemmed from the fact that Terra had the backing of some big-name investors in crypto. “I just figured that some institutional investors would come [and] kind of save the day,” Somani said. Firms such as Galaxy Digital, Pantera Capital and Coinbase Ventures were among backers that once invested in Terraform Labs, which supports the blockchain. Jump Crypto and Three Arrows Capital participated in a $1 billion purchase of Luna in February.
Representatives at Galaxy, Pantera, Coinbase, Jump Crypto and Three Arrows did not respond to emails seeking comment. Terraform Labs did not respond to a request for comment.
UST’s peg was briefly restored on 8 May, but it fell below $1 again on 9 May, and the crash began in earnest.
On the morning of 9 May, Somani was still hoping the peg would recover in a day or two. However, by late afternoon, “I had some investors in my own project call me, and they said that I’m gonna have to pivot and that they don’t see a future for Terra,” Somani said.
By May 10, “I was thinking, OK, how can I modify my project to maybe make it work on other blockchains,” Somani said. Then, by 11 May, “I accepted that the reasons why I built my project to begin with were no longer true, so I just completely dropped the whole idea,” according to Somani.
Now he plans to open source the project, though some others in the community told him to hold off, as they are exploring a forking of Terra, which means to make a change to the blockchain protocol and split the chain.
On Tuesday, Do Kwon, founder of Terraform Labs, proposed a revival plan for the Terra ecosystem, suggesting the blockchain be forked into a new chain without the algorithmic stablecoin. The old chain would be be called Terra Classic, while the new chain is to keep the name Terra.
The proposal is up for a governance vote on Wednesday.
Somani now says that the past week has been kind of “nice” in some ways, allowing him to take a break from the fast-paced life he had embarked on when he quit his hedge-fund job. Working on the cryptocurrency project was much busier than his job at Citadel, he said. “I spent every waking moment thinking about my company, like every single weekend, every hour, when I’d go out with friends, I was still thinking about it. I was still responding to messages,” Somani said.
Over the past few days, “I can cancel all my meetings,” Somani said. “Obviously, none of my meetings were relevant anymore. So it freed up a lot of my schedule.”
For the past week, Somani got drinks with friends, went on a date, and kayaked with some people he’d met at the Terra hacker house, where builders on the blockchain attend workshops, work on their projects, and network with each other. He had been attending the event, which is held at the Chicago headquarters of the proprietary trading firm Jump Trading, for the past few weeks, and plans to continue to go there this week.
“It was a pretty sad hacker house last week,” Somani said.
“Many of the people who lost a lot of money are no longer there. They left. And some of the people were very young. They’re like college students. They didn’t quite know the impact — so they were like, oh, should we keep building? And I was like, no, you should not keep building. There is no Terra anymore.”
This article was first published in MarketWatch.