Interview with bloXroute: The path opened by Flashbots makes it difficult for DeFi to surpass CeFi

CN
链捕手
Follow
1 year ago

Guest: Uri Klarman, bloXroute CEO & Co-founder

Interview & Editing: Jack, BlockBeats

Translation & Compilation: Kaori, BlockBeats

When it comes to DeFi Summer, people often think of Uniswap, Aave, and when it comes to MEV, most people think of Flashbots. However, when we see the name bloXroute, most people may not have heard of it. But in fact, during DeFi Summer and even earlier, bloXroute's BDN (Blockchain Distribution Network) and RPC routing services have been favored by most DeFi participants. According to BlockBeats, during its peak, 90% of the DeFi transactions and MEV behaviors in the market used bloXroute's services. And all of this happened before Flashbots proposed the concept of MEV.

Recently, BlockBeats interviewed Uri Klarman, CEO and co-founder of bloXroute, to discuss the brief history of bloXroute's development, the current status of MEV development, and his views on the future development path of DeFi.

"The Most Primitive MEV"

Before entering the crypto world, Uri was a software engineer in Israel, and later, during his pursuit of a Ph.D. in computer science in the United States, he came into contact with cryptocurrencies. Alex Kuzmanovic, the co-founder of bloXroute, was Uri's former mentor.

Alex is a renowned professor in the field of computer networks. When the two met, they hit it off and together researched and built a system called "Webcoin" and published a paper. Unlike PoW, Webcoin did not require extensive calculations but instead crawled and indexed all web pages on the internet, creating an index similar to what Google owns.

Around 2016, the idea of Webcoin led the two to start thinking about blockchain and cryptocurrencies. They realized that they could build a system for rapidly disseminating information on the internet and named it bloXroute because it combines block and transaction routing. During this period, Bitcoin faced a lot of controversy, and their idea was that if they could send blocks faster and propagate more integrity on the internet, it might support larger blocks, reduce the time between blocks, and help the blockchain scale.

Fast forward to the summer of 2020, the DeFi frenzy swept the entire industry. Uri realized that if everyone could see transactions faster, they could all send transactions faster than the mining pools on Ethereum. Essentially, by quickly obtaining information and rapidly sending it, people could make better trades, which is essentially the lightning network of the DeFi space.

Suddenly, all DeFi traders became bloXroute's "customers." People used bloXroute to fetch, view, and execute their transactions, and even to send transactions without being seen. For this, they also chose bloXroute's private RPC and order flow. This was all before Flashbots emerged.

BlockBeats: Let's start with a brief introduction, such as what you did before, how you entered the crypto field, and why you decided to establish bloXroute?

Uri Klarman: Fast forward to the DeFi frenzy of the summer of 2020, we had some moments of panic. If everyone could see transactions faster, everyone could send transactions faster than the mining pools on Ethereum. Essentially, by quickly obtaining information and rapidly sending it, people could make better trades.

This is essentially the lightning network of the DeFi space. If you have a fiber optic connection between Chicago and New York, and you are one millisecond faster than others, you will leap in the sky of New York, be the first to reach Chicago, and then buy and profit before others because you are fast enough.

So suddenly, all DeFi traders became our customers. They wanted to use us to listen to blocks, view and execute other transactions, and even to send transactions without being seen. So they chose what we call our private RPC and private order flow. Looking back, what we mainly did was called protected transactions. We allowed users to send private transactions through their mining pools without being seen by others, and then Flashbots appeared, and suddenly, MEV became a thing. We run a Flashbots relay, and all searchers became our users and customers because they wanted to see other transactions.

At the beginning of 2020, we didn't have real paying customers, and then within five quarters, we made $10 million in annual recurring revenue, which is a very strong 0 to 1. Then Ethereum transitioned from PoW to PoS, and mev-boost emerged.

Now we have multiple roles, like everyone else, we are a block producer, we built a small portion of blocks on Ethereum, but on a very small scale, we won't heavily invest in this because it's competing with our customers. We are more of an MEV relay, taking it from the block producers, giving it to the proposers, and then sending it to everyone else.

We provide all our services to MEV searchers, our customer base ranges from Windermute and Jump to large trading firms, to an unnamed independent searcher in a basement in Taiwan, and anyone else from anywhere.

Going back to my journey, I started as a researcher, then founded a company, and I've been working in this industry for about 7 years, but deep down, I'm first a network person and then a crypto person.

BlockBeats: How did people discover bloXroute and its products during the DeFi Summer frenzy? Did you initially target institutions or retail?

Uri Klarman: At that time, we had 16 people, basically all developers. We had no sales, no customers, no sponsorships. MEV searchers and DeFi traders spread the word, telling each other that they were using bloXroute, and then we reached the peak of routing $1.5 billion in transaction volume daily during the DeFi bull market.

This is equivalent to 40% of the DeFi transaction volume on-chain being completed using the services we provided to users and customers. When the bull market ended and the bear market came, this data significantly decreased, and our current revenue is about half of the peak of the bull market, but we are still routing transactions for everyone.

BlockBeats: During the DeFi frenzy, how much money were you making per day?

Uri Klarman: Our annual recurring revenue (ARR) reached $10 million, which means we were making about $900,000 in revenue per month. Although our current revenue has decreased somewhat, our market position has actually improved. During the peak of the bull market, people didn't care, they would buy your service, feeling that everyone was rich. But when the bear market came, people started thinking about where they should spend money or where they shouldn't.

Many of our users said that they are now going out of business because they are losing money no matter what they do. We like your product, and we will continue to use it in the future. We are not stopping because it's not worth it, but because we have to stop all business.

So from 2021 to 2023, we are working with large developers, collaborating with DEXs, and having very interesting collaborations and discussions with many participants, many of which are related to our valuable infrastructure for DeFi transactions and MEV.

Not Following the Path of Flashbots, DeFi Needs to Compete with CeFi

BlockBeats: Interestingly, although bloXroute initiated the project that gave birth to the concept of MEV, Flashbots is the company that proposed this concept.

Uri Klarman: During the era when MEV started to become popular, it wasn't called MEV. All mining pools knew they shouldn't get involved, it was like a referee scoring a goal. Many mining pools said, "We know we can, but we won't do it because it would be bad, it's a taboo, and everyone would be upset with us."

Then the lightning network appeared, tearing down all the red lines. It was like saying, "It's okay, you should do this because if you don't, others will." And within six months, Ethereum developers completely changed their views and thought that MEV was now acceptable.

So, I believe that MEV is not a zero-sum game, but a negative-sum game, and that's really bad. If validators (formerly miners, now validators) maximize extracting more and more value from DeFi, then the pockets of DeFi users and LPs will have less corresponding value, and DeFi will become worse and worse.

As a result, DeFi will lose its competitiveness. If DeFi becomes worse and less useful, it will have less liquidity. For example, if I want to buy Ethereum or anything else on Uniswap or Coinbase as a retail user, I don't have five screens in front of me to track prices at all times.

But I just want to buy ETH at the current price, and if I go to Coinbase, I can buy at the current price. But if I buy on Uniswap, I will get the worst execution, which is the worst price I'm willing to accept. For me, there's not much choice.

So, compared to CeFi, DeFi is performing poorly. And if DeFi performs poorly and is not that useful, it will have less liquidity. In fact, this will result in less MEV for validators, which is very short-sighted. Don't get me wrong, I'm a super supporter of DeFi, and I think DeFi is very valuable. It's transparent, permissionless, global, always-on, 365 days x 24 hours, providing a unified financial infrastructure for the world that no one can manipulate, which is very valuable.

I remember at the 2019 or 2020 Pantera Summit, the head of the U.S. CFTC, who was also one of the regulators in the U.S., told the audience that when Bear Stearns collapsed in 2008, he was appointed by Obama. At that time, they had to decide whether to save it or let it collapse. They had enough money to save it, but if there were 20 Bear Stearns, they couldn't save them all.

So their decision at the time was to let Bear Stearns collapse, they didn't save it, and then the global financial crisis of 2008 happened. After understanding how much money each bank had and what they held, everyone realized that they couldn't see who held what. They could have saved the situation, just like Detroit today. If we could transparently understand what actually happened in the banking system, we could completely bypass it, which has become very apparent in recent events like the Silicon Valley Bank and some similar events.

What we are waiting for is whether DeFi has solved these problems. People don't understand how powerful and valuable this thing is, DeFi can precisely know what each protocol, each institution owns, how much, and what risks they are taking. So I think DeFi is crucial and should compete with centralized finance, not just be its co-pilot. But the current way of MEV is not leading us in that direction, it is making DeFi worse and hindering its competition with CeFi, so I am researching alternative solutions for MEV.

Validators are not just increasing their share in the pie, but making the pie bigger, like 10 times, 100 times, 1000 times. This is much better than capturing 50% or 90% of the share and making the pie smaller. So I am a super supporter of DeFi, but I don't think our current way of MEV supports this, and it's not good for DeFi. I am working with various partners to try to improve what we have.

BlockBeats: What is your view on Flashbots and its high valuation?

Uri Klarman: I really like some people in Flashbots, but liking them doesn't mean I agree with some of their views. Their early agenda was to minimize MEV as much as possible because MEV affects everything that happens. And the part that cannot be minimized should be democratized. This way, it's not just large validators who can extract it, but everyone can use it, creating centralized power.

As we discussed earlier, mev-boost, if everyone can access these third-party block producers, then everyone will get similar APY. But I think the result we see in reality is that we are maximizing MEV to the fullest, and from my perspective, the result is not good. That is, the result is not to reduce MEV, but to maximize MEV.

If all validators use the same API, perhaps maximizing MEV is valuable, theoretically it can avoid centralized power. What we observe is that block production is actually dominated by the top three participants, who account for about 70% of block production.

But going back to my previous point, what I care about is not just whether DeFi succeeds or not, and the issue of competing with CeFi. The current situation is that the funds that were previously extracted by Jump, Wintermute, and DeFi traders (such as HFT, DeFi arbitrage, etc.) are now flowing to validators. But this hasn't helped DeFi users, DeFi users are still losing money, and Uniswap LPs are also losing money. This money is not flowing to traders, but to validators, and then to ordinary ETH holders, which hasn't helped DeFi.

Perhaps in the future, we will have MEV burning, so that money will actually be destroyed. However, this still doesn't help DeFi users, they are still losing money, so I am more interested in solutions that focus on increasing the value of DeFi and have the ability to compete with CeFi, rather than focusing on finding ways for everyone to share MEV.

Some solutions, especially those like order flow auctions, are basically like this: "I make money from your trade, rather than taking $100 directly from your pocket." I operate behind your trade, and if you lose money, I will refund you a portion. Although this situation is better than being robbed directly, it is still not a good solution, especially if all order flows are now flowing to a very small number of participants.

Flashbots is working on something like Sauve, which is a different L1 that focuses on MEV and allows coordination of all MEV across multiple chains, among other things. This is a very complex process, it's not just a single transaction, but a meta-transaction, and it is handed over to searchers who will compete in an auction-like manner to see who can get this transaction, and then hand it over to the builders participating in this additional alternative L1.

Another thing is that there is actually no real cross-chain MEV, people are talking about cross-domain MEV, and the same goes for arbitrage between DeFi. Because CeFi is 50 to 100 times larger than DeFi, 60% of MEV is arbitrage between CeFi and DeFi. If you delve into it, you will know that 91% to 99% of the value actually belongs to validators, but when you are doing CeFi, DeFi arbitrage, about 55% of the value belongs to validators.

You trade here, you trade there, maybe you make money, but you need to take risks. If you don't make money, you won't do it. You also need a lot of liquidity, whether on-chain or on CEX, so it's not that competitive. So, the arbitrage between CeFi and DeFi is much larger than the arbitrage between Ethereum and Cosmos. As my main point, DeFi needs to compete with CeFi, and currently DeFi is not good enough, like a sidecar trailing behind CeFi, mainly used for capturing arbitrage.

Making MEV Auctions Disappear

BlockBeats: Will MEV and MEV auctions be a big part of bloXroute's business?

Uri Klarman: It's already the case. We are a profit-making company, but I want to make money in a way that makes DeFi valuable and doesn't lose value, and I can make even more money in that world. So, I hope DeFi can compete with CeFi, and I believe I will find a product that fits the market. If you gave me a magic wand today, I could magically change the MEV landscape so that there are no more auctions like we see today, but rather a first-come, first-served system, and some other things that allow DeFi to compete with CeFi.

Then I have to rethink everything we do in our business, and I want to make this party bigger. Currently, we are deploying our MEP (multichain event protocol) solution for BSC, and you can imagine that we are building a MEV solution that allows everyone to compete with us.

BlockBeats: Why do you say you can make more money when MEV becomes bigger, is there a mathematical principle behind this?

Uri Klarman: Many people who hold Ethereum hope for the success of Ethereum, including myself. For example, I have this token, that token, let me try being an LP on Uniswap, try this, try that. So there is quite a bit of activity with ETH and USDC and some other things related to ETH.

We, the people who hope for the success of Ethereum, are willing to put our money there. But for those who are not for Ethereum, they want to make money through trading, where do they trade? On CeFi, not on DeFi, because DeFi cannot provide them with the best prices.

And look at Bitcoin, many people are trading Bitcoin, the BTC trading volume on CeFi is twice that of Ethereum. And 99.5% of things happen on CeFi, with the remaining 0.5% happening on DeFi. Why don't they trade on DeFi? Because DeFi is not good enough. Again, whether you like Bitcoin or not, this is irrelevant. Is DeFi useful? Will people trade their Bitcoin on Ethereum? The answer now is no.

Once we make DeFi more valuable and compete with CeFi, it is possible that the volume of Bitcoin trading on DeFi will be similar to or greater than the volume on Ethereum, because on CeFi we see the trading volume is twice that of DeFi. But now, what percentage does DeFi hold? Can we make it 50%? Can we make it 500%? I hope DeFi wins, I don't want DeFi to be just a small part.

We know that CEX can go bankrupt, the recent one being FTX, and we can trace it back to the time of Mt.Gox and similar events. This is not the first time, nor the second, third, or fourth time we have seen this. Despite knowing that CEX may go bankrupt, people still put their money in CEX instead of DeFi, why?

We are not just talking about ordinary people like you and me, but professional risk investment companies and funds that understand the risks, and they choose to trade because CeFi is better. We need DeFi to become better to compete with CeFi, and if we can do that, the market share may expand 10 times, or even 100 times, and ultimately I believe we can expand it 1000 times. That's why I don't want to prove the current market share of DeFi, but rather think about how to occupy 30% or 50% of the market share in a market that is 100 times larger.

BlockBeats: Will this situation you just described cause a lot of reactions in the MEV community? Will the term MEV still exist in the future?

Uri Klarman: I think MEV will always exist. Flashbots wants to define success, they want decentralization, just like we do, but I care more about usability than others. I think if no one uses the most decentralized thing, it's useless, I don't want to build it, I want it to be decentralized and useful. We can't throw away usability, then what we are doing here is meaningless, or at least its meaning is a thousand times smaller.

So, there is always value that can be extracted, but you can make an option—everyone sends their trades to me, and I accept the one that pays me the most, and none of the others. Or you can extract a basis point from the trading volume and try to achieve real-time, first-come, first-served, just like a centralized trading platform, providing the same utility to get the current price, no one will cut in line, always get the right price, because arbitrageurs will trade, and the price will fluctuate in one direction.

Can we achieve pre-emption? How do we achieve first-come, first-served? What is fair? I'm not saying these questions are trivial, but my point is that this will actually make everyone make more money, which is more valuable for users and validators. So, it's not about making MEV disappear, but making MEV auctions disappear.

BlockBeats: So MEV itself is neutral, but MEV auctions are the part that everyone likes to participate in.

Uri Klarman: In my view, auctions are the worst part. Do you know why people hate high-frequency trading companies? We don't want decentralized finance (DeFi) to become centralized finance (TradFi), right? We don't want high-frequency traders to come in and play with their picoseconds.

Why do we hate them? Because they manipulate trades, whether through monitors or other means, they see your trade coming in, and you could have bought at the current correct price, but they bought before you, so you bought at a higher price because they saw your trade and acted on it.

Now, the MEV auction movement is here. If someone starts trading faster, they will get that price before others, that's okay, I'm not saying I don't have 10 different monitors and complex things in front of me, I just want to send a trade and get the current price, and no one can adjust their trade because they saw what I was about to do and trade in front of me.

In simple terms, if I initiate a trade, and someone else initiates a trade, regardless of the order, whoever arrives first wins, sometimes they win, sometimes I win, depending on who sends the trade first and when. My point is if I send my trade, others can only see it after I'm done, and they can't take action before that.

So I think MEV is not a neutral force, and there are no good MEV and bad MEV. But there is indeed bad MEV, called front-running, which means seeing someone's action before they take it and gaining value before that. This is different from arbitrage or back-running, back-running is just a fancy word for arbitrage, it means I will capture arbitrage opportunities after you, or immediately after I see your action.

This is actually a good thing, I like arbitrage. If the price on Uniswap is one value, the price on Balancer is another value, and the prices on Binance and Coinbase are the third and fourth values, I hope arbitrageurs and high-frequency traders will continuously correct the prices whenever there is a trade, so that the prices are always correct. So when I send my trade, I will get the current correct price, and this is what I hope to get from DeFi.

BlockBeats: Will this replace the role of crypto market makers?

Uri Klarman: No, these are actually the same thing. People trade, and if the price is not right, market makers will try to capture the difference. The main difference is that validators no longer say, "Listen, I'm about to create the next block, how much are you willing to pay?" and then the price fluctuates to a certain point.

Instead, let the validators have as much share as possible, but this is only a very small share because it is not very useful. Instead, you can imagine that they won't do that, but process trades in real-time in a first-come, first-served order, and then charge a basis point fee. Due to the many activities within these 12 seconds, with prices rising, falling, and rising again, by continuously correcting the prices, validators will earn a large amount of profit like all traders and DeFi arbitrageurs. They don't have the opportunity to extract funds from DeFi, instead, they charge fees to capture funds, which will make DeFi better and capable of competing with CeFi.

So capturing a basis point fee is equivalent to how much money we put in. But if we can expand it by 10 times, 100 times, it is definitely a better choice if we can do it. But the question is, should we do it? Should we achieve the goal of "real-time, first-come, first-served, fair ordering"?

BlockBeats: How likely do you think we are moving towards the direction that Flashbots and all MEVs are talking about now, which is the direction you don't like, and which direction we might be heading towards?

Uri Klarman: If there is a question that can replace this question, it's like asking how likely you are to succeed in pushing DeFi into a different realm. If you are a founder, you would say I want to change the world. But for a founder of a big company, trying to do big things is very difficult. Unless it's an individual like me, I think I will change the way DeFi operates as a whole, because I think I can do it, most people would say Uri, you can't do it, or I would say to myself, can I change everything? They would say no, and I think, yes, I can change everything. It's not so much about ability, but more of a personal trait, the belief that you can do it.

If you are in academia, it's like when your second paper is rejected, you adjust it and then send it to others, and then it's rejected, and rejected, and rejected. The first rejection hurts a lot, the second one hurts a lot, the third one stings a bit. After that, you just wait for the rejection letter, and then send it out again. But then you will have many papers, one will be accepted here, another will be accepted there.

Maybe you can't imagine a successful path. We will try that, but it doesn't work, we will do this, and it doesn't work either, and then we will do this, and eventually we will win. Going back to your original question, we are now on the development trajectory of MEV auctions and DeFi, which makes DeFi less useful and unable to compete with CeFi. But I think we have the opportunity to change this, we have the opportunity to move in a different direction and trajectory, to make DeFi compete with CeFi, which is the future I am striving for.

BlockBeats: What projects or teams is bloXroute currently working with?

Uri Klarman: Alex Mizrahi, who used to work at Flashbots (he used to work at Flashbots, but I think he has left now), made a very good comment in a chat, the gist of it was: "You can't imagine how many people in the Lightning Network really want to see a better future than now." I don't remember the specific details, but they all want to see a world where DeFi coexists with CeFi, instead of extracting all the value.

What we need to think about is how to achieve this goal, what methods to take, and the pros and cons of each direction. In the DeFi field, most of the people I have met think it is a good future, and few people do not want to see such a scenario.

Some say we cannot achieve such a future, some say I think it is not feasible, we cannot achieve fair ordering. But we can try to do this, start broadcasting from the other side of the DeFi world, and then gradually move closer to you, so that even if someone sees your trade, they cannot cut in line because the speed of light is helping you unpack first.

In short, if you send a trade from the farthest place to the nearest place, taking 100 milliseconds, then even if someone sees your trade and is a few seconds faster than you, they cannot surpass you within 100 milliseconds, because the speed of light is in your favor, we have a great advantage in this regard.

Can we do it? Can we overcome these huge technical challenges? This is the direction I am striving to explore, making DeFi compete with CeFi, and usability is very important. How do we achieve this goal? How do we achieve democratization? If we don't extract MEV, someone will always go and extract those MEVs, so we need to democratize it and try to make it fair for everyone to share. Maybe we will destroy it, or validators will get it.

BlockBeats: bloXroute also has a small part of the builder business, please talk about the market for "Block Builders" and "Block Producers".

Uri Klarman: If there is a searcher who finds an opportunity and wants to seize it, whether it's on-chain arbitrage or trying to back-run or pre-empt a trade after seeing a transaction, in reality, they don't know how much they should invest, and then they almost invest all their money.

These searchers don't know what other searchers are doing and how much they are betting, but basically they have to bet almost all the money. Block builders are mainly CeFi and DeFi arbitrageurs, they collect searcher bundles, and conduct large-scale ETH, USDC, BTC, ETH, or USDC, USDT, etc., CeFi, DeFi arbitrage based on the 12-second UniSwap price, because the UniSwap price updates every 12 seconds, they can flexibly adjust the actual fee paid at the last moment.

Block builders will build a block, and over time, they can store for longer periods to create better and better blocks, continuously improving and building better blocks. They start bidding around the last 200 milliseconds, trying to become the highest payer, while also comparing how their captured value compares to others.

If they earn $1000 in capturing CFA (such as USDC), they may propose $500, and when they see that the value of another person's block is $600, they may propose $610, and this change happens very quickly at the last moment, because only at that moment do they truly understand what the opportunity is. Prices change in real-time, and if you bet within five seconds, maybe the price will drop, maybe you will lose money. So you want to bet at the last moment.

An interesting dynamic can be seen, when the block value is low, before reaching the expected best value, many proposers will compete to bid, continuously proposing the best price, trying to get the block at that price. They will not release all the value, because it doesn't make sense for them.

Capturing CeFi-DeFi arbitrage carries risks, maybe the price has already changed when executing the trade, leading to losses, so compensation for risk is needed, while also needing to make a profit. If they cannot earn a profit of 1 or 2 basis points, they will not do it. If there is no profit, it is not worth it for their business. Therefore, we find that CeFi and DeFi arbitrageurs are more suitable for block building than anyone else, able to capture C-D arbitrage in reality, and adjust the actual amount paid or not paid at the last moment.

BlockBeats: It is said that there are only a few major players in the Builder market, and individuals cannot profit from it. What is your view on whether we should address this issue?

Uri Klarman: If you just want to be a searcher, you can create a bundle, trade with a million dollars, buy here, sell there, make $1, pay back the loan. For example, you make 99 cents, give 90 cents to the validators, and net 9 cents. If it doesn't work, the loan will be returned, there is no risk, and it doesn't belong to the block builders, but anyone can do this anywhere in the world.

If you want to be a block builder, you need the ability to capture CeFi and DeFi arbitrage opportunities. To do this, you need a lot of money, you need a lot of funds on centralized exchanges, and a lot of funds in DeFi. It's not just as simple as buying ETH here and selling it elsewhere to make a profit, for example, buying $1 million worth of ETH in DeFi and selling $1 million worth of ETH on Coinbase or elsewhere, in this case, the holdings generally remain the same.

You need to have millions here, millions there, and you actually need another million. So, you need a lot of capital and risk management. This is a very big operation, not only do you need to go through KYC on centralized exchanges, so not many participants are doing it, and even if they are, their tolerance for regulatory risk varies. Therefore, some people will trade tokens, while others won't, because some large high-frequency trading companies don't want regulators to be upset with them over this issue, while others may take a chance and do everything.

So you need different solutions to meet different CeFi and DeFi arbitrage needs, but they are all large and involve a lot of capital. Because of jurisdiction, trading in the US will be more difficult, but if it's in the Bahamas or similar places, regulatory issues may not be a concern.

We are currently working with partners to explore bridging this gap, imagine not needing to be a CeFi or DeFi entity, but a researcher, a large CeFi participant who gives you a quote to make a trade, allowing you to buy ETH at this price, but this price is not exactly the CeFi price, because they make a profit from it, but also provide CeFi pricing to DeFi participants.

So, small participants who do not have a large amount of funds in CeFi may suddenly become participants who have funds and provide quotes to others. But they will charge a fee for this, perhaps making the entire ecosystem more profitable, and this is what we are exploring.

BlockBeats: I think BSC is also very important to bloXroute, can you explain in detail the relationship between BSC and your company?

Uri Klarman: We have been playing an important role in the BSC field for a long time, there was a lot of trading activity on BSC at the end of 2020, so people like DeFi and cross-chain trading. DeFi happens on Ethereum, DeFi also happens on BSC, and others are an order of magnitude smaller than this. Arbitrum has been developing, you might want to add Arbitrum to the list four months ago. After Friend.Tech appeared, maybe we will see more activity on Base, but the real activity happens on Ethereum and BSC.

So Ethereum contains the development of the entire MEV ecosystem, with mapping and proof of work, and so on, we can see its development and changes. On BSC, people use our product to quickly get blocks, but we have only been working with validators in the past few months, trying to help them build better blocks.

Similar to the Flashbots argument on Ethereum, we hope that all validators can get roughly the same APY, if some do better and some do worse, then some will have a higher APY, so everyone will stick to using them, this is a centralizing force.

So we decided to build a method that allows validators to accept third-party blocks, not just from bloXroute, but from others as well. So we built this infrastructure, allowing others to compete with us, and we are confident in our competitive ability.

We allow each validator to build their own block and get a block from us, and then choose from it. We can continuously build better blocks because we can see transactions faster and see blocks faster.

So we are attracting MEV searchers, trying to capture arbitrage, and so on. It's a bit like a chicken and egg situation, in a growing network effect, we already have many validators, so we have more searchers. More and more searchers use us, creating more value for validators. So we find ourselves working with validators to help them increase APY, as well as working with DeFi participants.

Whether we have a partnership with PancakeSwap or not, basically anyone can send transactions to us, and we will protect them from manipulation. We will execute the transactions afterwards, if there is realizable value behind it, and give some to the validators and some to the searchers. Suppose you leave $100 on the table, but you are only trading here and not on all DEXs, then it's not a good trade, and we can provide some rebates. Most importantly, for front-running transactions, if you are about to make a big trade, you will definitely want to send the transaction through us.

We have good cooperation with the BSC Foundation, validators, and DeFi participants, to build something different from what we see on Ethereum, because BSC is different from Ethereum, it is between block builders and validators, there are very few validators here, so we will try to give a block to a validator, and if the validator wants to solve this problem, they can, so as to prevent sandwich attacks. We have a good working relationship with BSC and can see its gradual development over the past few months.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

OKX:注册即返20%,全网最高返佣,不薅白不薅!
Ad
Share To
APP

X

Telegram

Facebook

Reddit

CopyLink