DeFi Utopia - TOKEN2049 Dubai 2024

hi everybody uh my name is Eugene I'm the co-founder of ellipses labs and we are all about building high performance defi I'm really excited to share our vision for the company with you guys all today so we are core contributors to Phoenix which is a limit order Book Spot decks on salana it's completely onchain so the order book you see here um is just updating live directly on the salana [Music] blockchain [Music] apologies uh so like I said Phoenix is fully onchain the entire order book State the entire matching engine lives on the blockchain uh we've done over $20 billion in cumulative spot volume actually that number today is over $24 billion and we are in the top five by daily spot volume across all uh across all blockchains and the thing that we find most important about Phoenix today is the active market makers on the exchange are profitable without external incentives so the exchange today is operating in a sustainable State there's no uh unsustainable incentives coming into play to juice the volume numbers um and so today what I want to talk about is our vision for the future of defi because we think Phoenix today is not good enough so I'm going to go into a little bit of a quick history lesson on the history of defi so early defi was really enabled by uh the launch of ethereum L1 which was the first general purpose computing decentralized Computing environment that was available and so that enabled the uh def Pioneers like Unis Swap and maker and a to deploy their initial programs and these were really proofs of concept that you can put a financial system on chain you can put these Financial Primitives and they inherit a lot of these properties from the L1 itself so all of these products are trustless they're auditable they're fully censorship resistant at least to the ENT that the underlying layer one is and Beyond being just these proofs of concept they also showed these net new use cases so Unis swap showed that you can actually have an exchange you can actually have liquidity without relying on a centralized exchange to list your token without relying on having a market maker to provide liquidity a and compound really showed that you can have lending uh fully on chain without any trust involved which is uh definitely a net new use cas case but again these Protocols are really proofs of concept this is really early Defi and they constrained by the limitations of ethereum which does around 20 transactions per second across the entire uh State machine and so these came out in the 2016 to 2017 era of Defi and there's been a lot of progress on blockchain infrastructure since that early era of defi but the core defi protocols themselves have not actually improved very much so we've seen some movement in defi in the 2020 21 cycle as well as some more in this current cycle but at the core protocol level these have really been marginal improvements so moving from XY equals K to concentrated liquidity most of the Innovation here is still happening on ethereum Main net and the protocols that are moving to the layer twos which are more performant blockchains they have more block space and lower feed uh lower latency most of the dominant def protocols on these layer twos are pretty much copy paste from the protocols in ethereum which in my mind kind of defeats the purpose of the better infra if you have better infra then you really should be able to enable a new set of applications that are uniquely enabled by the uh improved infrastructure and there's even this school of thought today uh called coin nihilism which is looking at the current state of the market where there really is just a lot of onchain gambling and in my mind it's this really really pessimistic view it says that there's no way defi is ever going to be better than traditional Finance we shouldn't even try that goal is way too ambitious the only net new thing that we're going to create with crypto are these shitcoins and the future of defi really is just this casino that is enabled by regulatory Arbitrage today and I think this is what we see in the market right now but it's an extremely pessimistic view of the future and so we really are at this Crossroads where on the one hand you have defi dystopia which is this world where there's not going to be any more innovation in crypto really the most important thing that is happening is we've built this decentralized casino which has some semblance of product Market fit very real product Market fit in these Ponzi games and these new uh meme coins but I think the future is a lot brighter and this defi Utopia we love to talk about at Ellipsis Labs is really this long-term view that we can actually build a decentralized financial system that is better than the traditional Financial system and there's a lot that has to happen along the way but I think the path is actually very tractable so I'm going to go into how we think this is going to happen so first we need to take a look at what some of the problems are in traditional Finance today and then where defi has strong advantages over traditional Finance so in trfi today we have this big system it's Global in scale but it's inefficient in many ways and it all comes down to the fact that traditional Finance has evolved in this iterative way over a hundred years where equity in companies is literally represented as slips of paper that people would hold in safety deposit boxes to do a trade uh the settlement time of very long because you don't even have any sort of electronic rails whatsoever and this is not the world we live in today but any system that's been built up over so long just has all of this scar tissue inside the system which we would call on the engineering side like technical debt any engineering system that's so big and has been around for so long and hasn't really been refactored or rebuilt from the ground up is just going to be really really slow and hard to change uh the system itself also is highly regulated there's all these entrenched middlemen and they have veto power over any changes so it really really blocks any sort of innovation or even attempts to innovate within the traditional Financial system and that means there's not much opportunity for new ideas where uh if someone has a good new idea or even a bad new idea on how we're going to change something in finance some experiment they want to run it's extremely difficult under the traditional system now defi has a bunch of advantages so the ones that folks generally like to talk about are uh defi offers permissionless access the protocols themselves can be trustless which means uh they're going to be auditable where the source code Terminus compiles down to this bite code that lives on chain and uh they have access to this composable shared State I think the most important piece is permissionless innovation where on these decentralized rails someone can deploy like a brand new exchange primitive like a single developer can deploy one within weeks and just see what happens with Phoenix we built the original prototype uh in 6 months and just deployed it to main net when we're ready just like we're going to see what happens um and this is something would have been totally impossible in the traditional world so I've spoken a lot about defi being a better system than traditional finance and so we need to talk about some of the metrics that I care about so so first is efficiency where for a trading protocol what this means is the prices that are get offered to the retail users are as good as possible so this means really tight spreads on trading pairs as well as uh deep liquidity and we want to get to a spot where the liquidity is better than what exists on centralized exchanges and then there's this component of sustainability which I touched on before it's not like the ESG sense of sustainability is that the protoc needs to be self-sustaining so the protocol can't depend on these external incentives forever which we see a lot in defi with these token incentives where the product looks like it has product Market fit it does a lot of volume it has a lot of tvl and then when the incentives dry up all of that goes away so there was never actually any real product Market fit and the key Insight here is we haven't really innovated too much on the protocol layer since this 2017 era of defi but the infro itself has actually improved a ton so there should be a lot of room to build better Primitives build more efficient sustainable Financial Primitives that are unlocked by this High throughput infrastructure so like I said before we really view Phoenix which is a limit order book on salana doing hundreds of millions of dollars in Daily volume completely unincentivized we really view this as a proof of concept just the first step on this long path to building a future for defi where we've proved that active liquidity can outperform passive liquidity on chain and again by outperformance here I mean we can offer Phoenix market makers can offer better prices than the amms while still being profitable while still being sustainable but clearly what we have right now is just not good enough to displace traditional Finance because it's the same primitive that exists on these centralized venues a limit order book and it just happens to be onchain so it's definitely going to be less efficient because the market maker are competing in a more difficult more expensive environment so we need to do better than Phoenix today so I've talked a little bit about infra so we have ethereum mayet today which does around 20 transactions per second we have salana which does around a thousand transactions per second uh then cost to land a transaction will be anywhere from less than one cent on salana up to dollars or double digit dollars on ethereum and compared to centralized infra this is around a thousand times worse uh which is pretty bad I don't think we'll ever be able to get this number down to 1x but we should be able to get it down to 10x to 100x and I think the path is relatively clear and the goals here are we need to be able to get to tens of thousands of TPS uh where you can really support active liquidity across many different assets across many different protocols and you need the block space to be so abundant that the cost transaction is under a tenth of a cent per transaction latency is also extremely important where you want the latency for pre-c confirmations to be as quick as possible ideally under 100 milliseconds ideally significantly lower than that and we still need to preserve all the advantages we talked about of building on top of decentralized infrastructure so it still needs to be sufficiently censorship resistant it still needs to be um auditable and trustless and so there's a couple different paths that we're seeing in the market today to scaling the infrastructure continuing to improve the infra the first is via more performant layer ones so this is the approach that uh blockchains like salana and monad are taking where you really push the boundaries on Hardware you really push the boundaries on software to get to like the final form of a performant proof of stake system and then the other way to scale is via layer twos where there definitely is a trade-off you're trading off short-term sensorship resistance uh for a ton of performance on a centralized or slowly rotating sequencer and the path to scaling the l2s is quite clear where we have uh these alt da Solutions like Celestia and igen da that are really scaling da beyond what ethereum main net can provide and on the uh on the execution side there's a a bunch of different things that are getting experimented with today we have these General execution environments where you're going to maximize composability and then you also have more applications specific uh execution environments where you'll get better performance and sacrifice some composability but there's very credible paths here to getting to tens of thousands of transactions per second so once the infra is built out and it's getting into a relatively mature mature stage today now we need to get into the meat of the D5 protocal design where we need these Primitives that actually derive more efficiency from better infrastructure where again I think the limit order book is just not going to be good enough and here we really need to lean into the properties of dii that make it unique to trafi because remember we're going to be eating this 10 to 100x um uh performance hit relative to traditional Finance so we need to make up for it in terms of better protocol design faster iteration and a couple things that I think make obvious sense for a trading primitive are programmable liquidity and more expressive order types these are the types of things that be very difficult to implement on a centralized exchange certainly wouldn't be able to move as quickly as on a decentralized venue on The Leverage side today in the market we mostly see clones of this a protocol which came out in 2016 there's been some minor improvements on that but with uh a really performant base layer you can have Oracle updates every you know half a second every tenth of a second uh you can provide much more efficient leverage you can have actual markets for credit that we don't see in defi today and that results in a lot of inefficiency in the current market so this is what we're working on here at Ellipsis Labs we're working on everything across the stack on the infro side as well as on the protocol design side uh if you are interested in building this future this optimistic future for defi you can get in touch with me here we're hiring across uh basically all business functions in our office in New York City thank you e