Hyperliquid founder Jeff talks about three sources of inspiration for starting a business.

Author:TPBN,Translator: Wu Shuo Blockchain

This issue features an interview with Jeff Yan, founder of Hyperliquid, by TPBN. Jeff shares his experience transitioning from trader to entrepreneur, and the opportunity to rethink the decentralized spirit of the crypto industry after the FTX crash. Jeff recounts the birth of Hyperliquid, its technological philosophy, and its development vision: to create a fully decentralized, on-chain native financial system, becoming the "underlying network" for all future financial infrastructure.

He delves into why they rejected traditional venture capital, how they adopted the spirit of Satoshi Nakamoto to maintain protocol neutrality, and Hyperliquid's unique growth path at the brand and community levels. The interview also touches on the mechanism and innovative significance of perpetual contracts, team building principles, and a decentralized community-driven brand philosophy.

The entire conversation not only revealed Hyperliquid's technological thinking but also showcased the balance between idealism and pragmatism among DeFi entrepreneurs. John: Welcome Jeff to the show, and thank you for taking the time. Our audience is mostly familiar with technology but may know less about cryptocurrencies. Could you briefly introduce yourself, your experience, and your business? We can then explore some different directions in more depth. Jeff: Sure, it might be appropriate to start with my story. In May 2022, we were a small team focused on cryptocurrency trading, primarily DeFi and SCI-Fi. At the time, we realized we wanted to do something in the DeFi space because DeFi products were still in their very early stages. Basically, none of the products were very good, and we, as traders, felt we could do better. Later, the FTX crash became the driving force behind our decision to fully commit and build Hyperliquid. At that time, we suddenly realized that discussions about decentralization and self-custody in the crypto space, though written about in the past, seemed to have been largely ignored. But suddenly, these issues became extremely important. People began to realize that "if it's not your key, it's not your coin." We deviated from the original spirit of Satoshi Nakamoto and Bitcoin. So, we believed the world was ready to trade cryptocurrencies in a decentralized, self-custodial manner. Back to the present, Hyperliquid is now a primary on-chain price discovery venue, a fully on-chain financial system. Our goal has always been to build a platform that can ultimately support all financial activities; although today it primarily functions as an exchange, it is more than just an exchange. Most notably, it's a trading platform that supports on-chain perpetual contracts, generating over a billion dollars in revenue annually, and pioneering in many ways. John: It's interesting to hear you say that. Most founders say "we," but it's clear you're more like a guardian of this project, responsible for building and releasing it. Is that how you see yourself? Is this your way of thinking about the project? Jeff: Yes, we definitely drew a lot of inspiration from Satoshi Nakamoto. I think the founders of Bitcoin are very unique, whether it's him or them, it's uncertain who it is, but if it's Satoshi Nakamoto, Bitcoin was certainly first in many ways. Its uniqueness lies in the fact that it's not a traditional product. As you said, it's not a top-down company, but a product. I think the DeFi and cryptocurrency industry has also drawn a lot of inspiration from Bitcoin; after all, everything originated from Bitcoin, and there will never be another project as big as Bitcoin. Many crypto projects still operate from the top down; centralized exchanges are a good example. While they are good business models, their core operations don't align with the original spirit of the crypto world. Therefore, Hyperliquid is based on this mindset, striving to build a truly neutral protocol that will ultimately become the infrastructure for upgrading all financial systems. John: So you don't have venture capital? Can you elaborate? Is it because you're financially efficient and don't need funding, or do you have some particular philosophy about the role of venture capital in the crypto industry? Jeff: This does indeed go back to Satoshi Nakamoto. If Bitcoin had raised Series A funding earlier, it might not be the Bitcoin we know today. Even the world's best investors, with the most powerful investment desks, couldn't change that. So I think venture capital actually provides a very important service to the world; it effectively allocates capital and helps many projects grow. While they aren't always perfect, overall, I think their contribution outweighs their negative impact. However, in this case, I think when you're building a neutral protocol, there are likely some crucial elements involved in the flow of funds and transactions. Neutrality becomes more important than anything else in this situation. I think there's a path dependency problem in this case. If insiders dominate from the beginning, no matter how much supply, ideas, talent, etc., are distributed, it's impossible to avoid that "big bang" moment, which will leave its mark on the protocol's history. So, for any project being built, this way of early capital entry will have an impact. I think funding and rapid expansion are viable strategies, but when you're building a project that needs to remain neutral in the long term, the influence of history is very important. We prefer to take it slow and ensure everything is done correctly.

Venture Capital, Market Competition, and the Original Intentions of Hyperliquid

John: I want to continue discussing this issue. You can't stop venture capitalists from building positions on the open market, right? Suppose a major investment firm says, "We want 20% of the shares, and we will buy tokens and build positions on the open market," and it will perform like any other asset in our portfolio. Has this ever happened? Is it irrational to do so? Are there legal structural restrictions that prevent them from doing so?

Jeff: You can make the same argument for Bitcoin. For example, many venture capitalists are large holders of Bitcoin. I know several venture capitalists who were involved in Bitcoin very early on. So the issue isn't who owns it, but its origin. It's more of a matter of principle.

If you can't claim on one hand that the platform is neutral, that anyone can build it, and on the other hand say, "These people had the opportunity to build this platform first," then it's not perfect. While practically speaking, any project born after Bitcoin needs to launch in a highly competitive market, and innovation must be funded, we still want to get as close as possible to Satoshi Nakamoto's ideal, and that's still worth pursuing. Jordi: So, did you know what you were going to do from the beginning and just execute the plan, or did you iterate along the way until you achieved what you have today? Jeff: I don't think we were entirely clear about what we wanted to do initially. We just wanted to do things we could do well, and even doing one thing well is difficult. Initially, we focused on finding a big opportunity, and we found a space in cryptocurrency that might desperately need a completely permissionless platform, and perpetual contract trading was the obvious choice. At the time, these kinds of transactions probably accounted for more than 50% of the crypto market's revenue. That's where we started. Importantly, we were unwilling to compromise on the principles of how we operated from the beginning. Which companies have particularly inspired Hyperliquid? Jordi: You learned a lot from FTX, avoided those mistakes, and were inspired by Satoshi Nakamoto. So, are there any off-chain companies that have particularly inspired you? Jeff: There are many. Actually, every big company, especially tech companies, has been very inspiring to me. I grew up in the Bay Area, so it's hard not to be influenced by it. I think Amazon is a huge source of inspiration, especially because they start from first principles, are both driven and very pragmatic. I think it was Bezos, and probably others, who realized that they had built most of the internet stack, and it would be a waste if these technologies were only used in retail. Why didn't they abstract it, create suitable APIs, and allow anyone to leverage this powerful infrastructure they had built? That's the story of AWS and cloud computing, and I think it's a truly remarkable story. This also reminds me of Hyperliquid, which was initially an optimized blockchain specifically for on-chain perpetual contract trading because no other infrastructure at the time could do that. Later, we gradually realized that many other areas in finance, and ultimately the entire financial system, could benefit from this high-performance decentralized ledger. So you can see it as a way to provide infrastructure for liquidity. John: What do you think of Bezos's background at D.E. Shaw, where he was a trader? This may have influenced Amazon's form to some extent. How do you compare it to companies like Google, which originated from academic research? What were you doing before this? Do you feel like you entered the entrepreneurial path with a trader's mindset? Jeff: Yes, I think so. While I don't fully understand what kind of trading Bezos does, at least the approach of automated trading is a lot like doing physics. You do a lot of approximations because you can't be 100% accurate, because the market is inherently full of randomness, with noise outweighing signals, which is precisely the beauty of the market. It's like sifting through noise for signals. It's completely different from supervised learning settings in AI, which have almost unlimited data and very high quality. I think building Hyperliquid felt similar. Actually, we don't rely on massive amounts of data to drive decisions; we rely more on intuition. We just think very seriously about what the world should be like and try our best to do it the best we can. Of course, if something obviously bad happens, we'll adapt and adjust, but we won't deliberately create data, especially when the data is unreasonable. Therefore, we almost never use methods like A/B testing. John: You've mentioned perpetual contracts (perps) a few times. Could you provide a clear definition for our audience and explain what perpetual contracts actually are and why they're so appealing? Jeff: Okay, if you want to know what people are trading today, you can actually trade a specific asset, like Amazon stock. If you want some kind of leverage, that is, to make more money with less money, there are two ways to do that. The first is trading futures, which usually trade indices, like the S&P 500. The basic concept of futures is that two parties agree to a contract, similar to a long and short position. If the S&P 500 price moves by $1, the long position earns $20, while the short position loses $20. Futures contracts are usually settled with an underlying asset; for example, you might be trading the price of the S&P 500 over the next three months, or taking delivery of a cow at expiration. John: I see. If it's a trade like corn futures, if I let the contract expire, I end up having to physically take delivery of all the corn. But most Wall Street traders already know how to avoid that. You've also heard of strange situations, like oil prices going negative, resulting in someone buying oil at a negative price and ultimately having to accept all the oil. Of course, in a purely financial context, this is clearly not the expected outcome. So, do perpetual contracts unlock higher-frequency trading, higher leverage, or a different type of trader, with more quantitative or algorithmic trading? Who are the customers for these products? Why are they interested in them? Jeff: Yes, they are interested mainly because, as you said, futures trading has many problems, such as the possibility of settlement. Sometimes you need to settle, but most traders avoid settlement by constantly rolling over their positions. Take Robinhood as an example; futures are far less popular than its options trading. For retail users, options are very attractive because it's similar to buying a lottery ticket. You can buy a ticket, which feels great because the maximum loss is limited. However, options pricing is very complex, especially for retail users, particularly those using apps that don't provide enough information. You might get "fleeced" because of the complex structure of options. It looks simple, with only a strike price and an expiration date, but it's actually very difficult to price. Perpetual contracts combine the assets of futures and options. You want to trade only the price of the underlying asset, want leverage, but don't want to worry about expiration—that's the significance of perpetual contracts. Take Bitcoin as an example: almost every exchange has a highly liquid Bitcoin perpetual contract, the only liquid asset that never expires. All price discovery happens on these perpetual contracts, with billions of dollars traded within them. It effectively drives the price of the underlying Bitcoin asset. This is extremely useful for professional traders because they can trade it, and it's the most liquid Bitcoin trading instrument. For retail users, it provides a clear and understandable price, making it less likely to be scammed because there's only one market, it's highly liquid, and the bid-ask spread is very small. Therefore, it's a win-win situation for many participants. Of course, there will always be some people who prefer options, or even traditional futures contracts, but overall, perpetual contracts are undoubtedly more attractive. John: So, what are the biggest bottlenecks to scaling up a network like this? If the scale is very large, I'm more concerned about whether there are some tokenomics structures that would motivate people to build entire data centers? Are there any ASIC miners currently being built to run this network? I know that Bitcoin was initially mined with personal computers, then evolved into data centers, and finally, in pursuit of the cheapest energy, ASIC miners were developed because the algorithm is very stable. So, how will a network like Hyperliquid scale over time? What stage of scaling are you currently in? Jeff: Yes, Bitcoin is a bit of a special case; it's probably the only mainstream network still using Proof-of-Work (PoW). The ASIC miners you mentioned, all that crazy stuff like mining Bitcoin with volcanoes, are actually participating in the consensus mechanism. All other high-performance blockchains today (at least as far as I know) use Proof-of-Stake (PoS), which is a more energy-efficient method. Its security is based on an economic model, not computational power. So, on Hyperliquid, there isn't much innovation in how the network stays secure. The problem is relatively simple; basically, everyone stakes their native token. In Hyperliquid, this token is called "Hype." People stake it and indicate they endorse the network. It's similar to voting for a legislator on your behalf; you're essentially treating the validators you choose as objects of your trust. This way, if a validator does something wrong, the tokens staked to them are at risk. From an economic perspective, the system ensures network security through certain mathematical models. If most of the staking in the network is honest, then the system can function properly. If you want to obtain enough staking to do bad things (such as double-spending—spending the same dollar twice), you need to control a large amount of staking. Simply put, Hyperliquid's security does not rely on creating crazy ASIC miners, but is based on economic models. Brand and Community: The Power of Decentralized Marketing Jordi: How important is brand to Hyperliquid's success? Because I feel that you haven't focused on brand building like traditional companies, such as hiring advertising agencies and developing strategic plans. I believe many of your competitors do these things, but you have one of the most powerful brands in the world. It's amazing how much attention and excitement you get just by typing "Hyperliquid" anywhere in the world and posting it. But I'm curious how much of a role this brand effect plays in your success, compared to scale and other product decisions. Jeff: Yes, I think we're very lucky; the power of the community is incredible. I don't know how to describe this community—it's probably close-knit, intense, but also very inspiring. Our team is actually quite introverted and small, only 11 people, and we don't even have a dedicated marketing person. If we did, I don't think we'd do very well. It's not just the product itself, but more about the product and the community, the whole ecosystem. It's made up of many different parts. For example, some people just post things on Twitter, which I think is really cool, I really like it. And other people build products on the platform, which I think is also a kind of viral marketing. They build products on top of the protocol, and these products synergize with Hyperliquid itself, offering something new or expanding upon it in certain ways. The financial backbone of Hyperliquid is largely built by community members, and we hope this trend continues. Anything that can be built by the community will be done by the community. I think this decentralized marketing approach is the embodiment of decentralization, not only on a technical level but also on a social level, which is very important to us. This is the Hyperliquid brand—if you have to interpret it negatively, some might think we're a bit cold, or that we're too focused on technology and neglect marketing. But from this perspective, the final result is something much more powerful: people feel a sense of ownership within this network, a feeling that Web2 companies can't achieve. Competition and Focus: Hyperliquid's Strategic Advantages Jordi: This is why the companies trying to take your market share are often large, well-funded financial institutions, but they seem to have difficulty competing. Is it because there's a large community that wants Hyperliquid to win? How do you see this advantage? Besides that, what other reasons make your reverse positioning make competition difficult? Jeff: To be honest, we don't focus too much on day-to-day competition. I think there's too much to do right now. You're right, many companies do want to do the same thing as Hyperliquid, or want to gain market share, but we don't think about that too much. There's too much to do. If I succeed, it will mean doing something that no one in the world has ever done before. I think that sometimes I get anxious because of competitors trying to undermine Hyperliquid's established advantages, but that also makes us lose sight of the bigger picture. After all, we still have a long way to go to reach our real goal.

Hyperliquid's Vision: The Future of Finance and the Rise of DeFi

Jordi: How much potential does Hyperliquid have? What are your ambitions?

Jeff: I see it as, if everything goes well, eventually encompassing the entire financial sector. You could say it's the coordination of finance, which is the coordination of human behavior, not just finance itself.

Jeff: I see it as, if everything goes well, it will eventually encompass the entire financial sector. Jordi: You're basically saying that even if you don't cover 100% of the total global financial market (TAM), you'll still have a huge impact on it? Jeff: Yes, you asked how big? I think that's correct, but I don't think it will directly change everything like it replaced finance. My thinking is more like the impact of the internet on finance, or like the rise of electronic trading in the early 2000s. That's been going on for over 20 years; it's time to update the technology stack. I think DeFi is that technology update.

Quick Q&A Session

The Future and Market Expansion of Hyperliquid

Jordi: I have a quick Q&A session. I asked some friends in the crypto space to see what questions they wanted to ask you. I'll ask a few questions quickly, and I hope you can answer them briefly, as some questions are quite specific.

First, when will we see the first centralized exchange shut down their perpetual contract trading and just run a frontend on top of Hyperliquid? When will they surrender?

Jeff: Within a year. What are your thoughts on the US market? Jordi: What are your views on the US market structure? Under what circumstances would you consider bringing Hyperliquid to the US? Jeff: That's a good question. We consider the US a very important market; it's obviously a global financial center, and the US dollar is the reserve currency for capital. We very much hope, of course, that US regulation will develop and truly embrace DeFi. I think we're already taking steps in that direction. I can't comment specifically, but I feel I don't know enough about this issue because things change very quickly. To be honest, I think there are many very smart people doing this. I've heard about initiatives like granting exemptions for decentralized front-ends, which I think are really cool. Why do some centralized exchanges say Hyperliquid doesn't want to be listed? Jordi: Why do some centralized exchanges say Hyperliquid doesn't want to be listed? Jeff: Have they said that? I don't know. I don't really care about that. We don't want anything; I think we're more focused on building. Some exchanges have listed Hype, which I think is cool. And some other exchanges might feel it's not in their priorities, which I also think is cool. We don't really deal with these institutions much. Are there any plans to support native multi-asset margin? Jordi: Are there any plans to support native multi-asset margin? Jeff: I think it will definitely happen, but I don't know the specific implementation path. The concept of native multi-asset margin isn't clearly defined yet, but I think it will happen. If we're going to accommodate all finance, we should definitely be able to trade using different types of collateral. When will perpetual contracts for commodities and U.S. stocks be launched? Jeff: There are already some perpetual contracts, such as perpetual contracts for gold. As for when perpetual contracts for other commodities and stocks will be launched, I expect it will probably be next year, because HIP 3 will unlock a lot of features, basically allowing anyone to deploy their own perpetual contracts. What are the most common misconceptions about Hyperliquid? Jordi: What do you think are the most common misconceptions about Hyperliquid today? Jeff: There are many misconceptions. The most common misconception is that many people think Hyperliquid is just a perpetual contract exchange, or that it is a centralized thing. Actually, it's not. It's a blockchain network where validators are permissionless. Currently, there are 24 validators, and anyone can join. The top 24 nodes by staking amount form the validator set. Each validator executes all transactions, including all orders. In other words, Hyperliquid's underlying layer is decentralized, a point often overlooked. People tend to focus more on its product, which may be a good thing, as it means the technology is abstracted for the user.

Jeff, do you miss the Bay Area?

Jordi: Do you miss the Bay Area? What do you miss?

Jeff: I miss Chick-fil-A (laughs). Actually, I miss the mountains more.

Although not mountainous within the Bay Area, Singapore is very flat. It's a great place to live, with a wonderful environment and good work conditions, but I do miss the natural scenery like Yosemite. John: When I was in Singapore, many people said that Singapore lacked a strong local culture, perhaps because it doesn't have an underground culture like punk. Do you feel you've found a great community in Singapore? Do you like the culture here? Do you like the future direction of this country? Jeff: To be honest, I'm too busy right now to delve into the culture. Singapore is a great place—safe, modern, with good food and great air conditioning. Overall, it's a great place to live. Frankly, at this stage of my career, I don't pay much attention to culture. If I were in New York now, I probably wouldn't care much either. What are the standards for team building and talent recruitment? Jeff: To be honest, we're not really good at recruiting. I have to say, we're always recruiting, always looking for talent. I know a lot of people listening to this podcast are very smart. If anyone is interested in systems engineering and high-performance computing, and wants to build the infrastructure for the future of finance, I genuinely believe this is the best place to work, and we'd love to have them join our team. Our team is very small, only 11 people, and the requirements are very high. I think our standards are probably among the highest in the industry. I think we're one of the most efficient engineering teams. But we're trying to expand, and there's definitely a lot to do. We maintain very high standards in many aspects, including competence and integrity, which we value highly. If any listeners meet these standards, we warmly welcome you to join us. Are you hiring for positions that support remote work? Jeff: For new members, we currently only support in-person work. Initially, we supported remote work, but we found that this approach didn't quite suit our work style. What are your thoughts on the 996 work culture? Jeff: I think the most important thing is the quality of the work. I believe everyone's workload is different, and everyone feels tired at different times, which is also the most important thing. Personally, I work even more hours than the 996 work schedule, but that's because I feel I have no upper limit; I don't feel restricted. However, everyone's situation is different.