Why I left the most desirable team at Goldman Sachs


If there’s a team in Goldman Sachs’ investment banking division that combines both cool and the kudos of working for an investment bank that likes to refer to itself as “the firm”, it’s Goldman’s tech team in San Francisco. Gaurav Budhrani worked in the team for six and a half years before leaving voluntarily in September 2021.

Budhrani is now CEO of PrimeBlock, a bitcoin mining company. Leaving GS wasn’t an easy decision, says Budhrani, but it was one that came to feel inevitable. “It was 2016 when I first went down the bitcoin and crypto rabbit hole,” he reflects. “I spent a long time talking to Goldman colleagues and partners about its potential. My belief has always been that crypto and digital assets will be a massive industry and that we need to understand the emerging clients and the implications of this technology.”

Ultimately, though, Budhrani was drawn to bitcoin mining. When Primeblock invited him to become CEO, he says it seemed the inevitable culmination of his career up to that point.  “If you were a priori designing a sequence of work experiences to prepare for the bitcoin mining and infrastructure industry, it would look a lot like my career,” he reflects. 

Budhrani started out in computer engineering after completing a degree at the Indian Institute of Technology. He initially worked in network engineering for Cisco in India. But Cisco’s Indian operation wasn’t the place for the “young, ambitious, driven person,” that Budhrani says he was. “I wanted a more dynamic working environment,” he explains. He studied an MBA and joined Goldman Sachs.

“I started in the natural resources group at Goldman and focused on sectors like energy, commodities, and renewables,” says Budhrani. “When I moved to tech investment banking in San Francisco, my focus was on deep technology verticals, including semiconductors and next-gen computing technologies from high performance to quantum computing.” 

It’s this combination of computing expertise, energy knowledge, and a deep understanding of computing hardware and supply chains, that Budhrani says equips him to become a bitcoin mining specialist. 

PrimeBlock’s co-founders include Ryan Fang, a former summer analyst in Credit Suisses’s technology team. For the moment, it’s been bootstrapped by the founders, but it’s reportedly in talks to merge with a SPAC that will take it public. Vertical mergers may yet follow. “Bitcoin mining is operationally intensive,” says Budhrani. “The key inputs are capital, infrastructure, and energy. As a result, the companies that can scale faster than anyone else will have a competitive edge in the near term. Still, as the mining industry matures, the differentiation will come from innovations across the stack – upstream and downstream.” 

In the bitcoin mining context, upstream differentiation means improvements to the chips used for mining. Downstream differentiation means improvements in chips’ energy consumption.  “We have already seen significant advancement in bitcoin mining chips,” says Budhrani. “Each successive generation of chips is becoming more efficient. As a result, you are adding more capacity and security to the network without growing the energy footprint in the same proportion.” 

The bitcoin mining industry faces various questions as it evolves. They include location. Following China’s mining ban, the U.S. has emerged as the preferred place for many miners and Budhrani says this is unlikely to change.  “The U.S. is the biggest energy market in the world and has the best infrastructure,” he says. “It also has political and regulatory stability. – I also expect the U.S. energy complex to become an advocate for bitcoin mining in the future.” 

There’s also the move from ‘proof of work’ to ‘proof of stake’, with the latter eliminating the need for miners. Solana runs a proof of stake model and ethereum is moving towards this, but Budhrani is unperturbed.  “The security that underpins bitcoin will always be proof of work,” “The best guarantor of the integrity and durability of a network representing a trillion dollar of value is the aggregate compute capacity on the bitcoin network. There are arguments for less compute intensive ways to protect the integrity of a blockchain network such as proof-of-stake, but proof of work will remain the most important consensus mechanism for a truly decentralized public blockchain.”  

Ultimately, crypto also has the potential to be undone by quantum computing, but Budhrani says this too is unlikely.  “Quantum computing won’t impact crypto assets in the next 10 years,” he says. “You’d need a quantum computer with well in excess of 5,000 logical qubits and the market leaders are currently at 20-25 logical qubits. We’re a long way from a situation where quantum computers can break public-key cryptography and hopefully by that time you will have quantum resistant encryption algorithms.”

Contact: sbutcher@efinancialcareers.com in the first instance. Whatsapp/Signal/Telegram also available (Telegram: @SarahButcher)

Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)

Photo by Nana Dua on Unsplash


Leave a Reply

Your email address will not be published. Required fields are marked *