# The Onchain Underworld: Solvers, Market Makers, Routing Algorithms, and Where AI Fits In > Shoal Signal episode. Guest: Nikita (Barter). Host: Gabe Tramble. - Publisher: Shoal Research (https://shoal.xyz) - Published: 2026-04-20 - Video: https://www.youtube.com/watch?v=RicJ2_3eWak - Canonical: https://shoal.xyz/research/episodes/the-onchain-underworld/ - Keywords: crypto, DeFi, solvers, market making, MEV, infrastructure, Ethereum, Bitcoin, interviews ## Transcript **Gabe (00:00):** Hey everybody, this is Gabe from Shoal Research and this is the first episode of Shoal Signal and I'm excited to have Nikita from Barter talk about all things DeFi, routing, and what I like to call the DeFi underworld. So without further ado, Nikita, yeah, please introduce yourself and tell us about what you're thinking. **Nikita (00:20):** Thank you, Gabe. It's pleasure to be here. Hello, everyone. My name is Nikita. I've been working in crypto full time since 2017. I joined during the ICO era and I was fascinated with new tokens and new primitives that emerged. And in 2020, just before DeFi summer and pandemic, I joined Oneinch as a third employee after the two co-founders, Anton and Sergey. Anton and Sergey are both technical, so they needed help with business development, investor relations. integrations and I've been chief business development officer at Oneinch for two and a half years, helped scale the team from three people to 150. And yeah, in 2022, I moved to Chile and I had a baby here. So I decided to take some time off crypto, know, focus on family. And around that time, the concept of intents started to emerge. And I'm super bullish on DeFi in general, actually. One of the reasons why I've joined Oneinch is because I felt like 2019 centralized exchanges had so much power over like what's going on in crypto and maybe market makers as well because for any project whatever you're building whether it was like DeFi, GameFi, you SocialFi, you at some point you have to engage with like listing team and you know like distribution of your assets to people and centralized exchanges were extremely sharky and when DeFi started to emerge I instantly fell in love you know the fact that like Uniswap the way Uniswap worked you can just create pool and whoever wants can start buying or trading your assets was phenomenal and like super refreshing to me as a crypto user. yeah in 2022 I feel like the concept of intent started to pop up and it wasn't even better I like like ideas that people don't need to like commit to a certain path or transaction they can just specify an intent like I want to buy one is at the best possible price like let's say like around two thousand dollars And then solvers or other sophisticated parties would make sure you get the best rate possible and would fulfill your intent. I definitely... felt like this is the way to go because we want mass adoptions, we want people to use our services, our infrastructure, and crypto was way too complex at that time. Actually, I still think crypto is way too complex, but we'll talk about it later, I guess. So we've built a solver, and Cowswap was the leading and pioneering platform at that time that allowed solvers like ours to find this best route and provide the best rate to the user. **Gabe (02:40):** Thank **Nikita (02:50):** So we launched in 2023 around February, yeah, February, March. And within like three months, we became number one solver on Cowswap by volume and users and transactions. And to this date, we have facilitated over 35 billion on chain. And last year, we were third largest router on Ethereum mainnet behind Oneinch and Kyber network. And we are very proud that due to the fact that we exist, users of Cowswap and other venues have more than like $5 million of coins more on their wallet than they would otherwise had to if, you know, barter never happened as a project. Because, you know, if we didn't exist and we didn't provide rate, then the second best like solver would facilitate your transaction. And you would get like a little bit less ⁓ in terms of your tokens. And yeah, so. due to our existence and users in crypto have like 5 million more and we are proud of this metric. But I don't know, should we dive what solver is and how it works? ⁓ Yeah, what do you think? **Gabe (03:55):** Yeah, yeah, was just going to ask to get into that because for solvers, this is not like a traditional business. This is almost like a competition, like a head-to-head competition, if you will. So yeah, if you can kind of give a brief on what solvers are and kind of what the reason solvers are here for, that'd be great. **Nikita (04:13):** Right, I've suffered for like three years trying to explain what solvers are to everyone during crypto conferences like ECC, Denver and others. But a solver, like the way we see it, is an interesting, unique mix between aggregator, market maker and MEV bot. Just like an MEV bot, you have to be aware of every block and the market conditions and you need to find the most optimal liquidity that exists on chain. Just as an aggregator, you aggregate all available, like ideally, all available liquidities that exist with like AMMs, pools, wrappers, et cetera, et cetera. So you try to rely on on-chain liquidity and provide the most optimal path for the users, the best rate possible using this publicly available liquidity sources. And as market maker, Every now and then you can tap into your own liquidity, into your private pools, and essentially market make and earn fees or spread within transactions. Solvers are a mix of these three unique market participants that also exist. I guess the biggest difference with aggregators would be that aggregators are more user facing interfaces. you know, people go to zero X and they interact with the depth while we as a solver, we power like B2B platforms like cowswap, for example. Yeah. So solvers specialize on, you know, optimizing, you know, inclusion in the blocks, optimizing, you know, gas that you spend on with every like swap and action. optimizing, analyzing the state of on-chain liquidity and try to give you the best possible rate and earn like a small fee on top of it. And a lot of people actually think being solvers is very profitable and like kind of cool. I strongly strongly disagree with that. Maybe it was profitable, well, not maybe it was relatively profitable in like 2023, maybe 2024, but the margins have been going down. The competition intensified dramatically. The users are getting more sophisticated and they are using... Well, the quotes have to be extremely tight in order to fulfill the intents. And we are facing tremendous competition from market makers. market makers are... pushing us to innovate and find ways to compete with their private liquidity. And this is probably one of the most fascinating events that are happening in our landscape right now. **Gabe (06:43):** Wow, yeah, so can you take us a little bit deeper, what, between maybe you and let's say five other solvers, what are the mechanisms or ways that separates you and how do you go about thinking of differentiating the product to bring the best prices at the end of the day and win these auctions? **Nikita (07:01):** So I would say there are two prime type of solvers on Caoswap. Some of them are market makers like Wintermute that operates resolver or others like TokaLabs for example. The other types of solvers are more algorithmic one like ours that rely on on-chain liquidity and that build it to unite existing DeFi primitive. So market makers goal is fairly simple. They analyzing, you know, the liquidity that's available off chain liquidity on centralized exchanges. And they are figuring out if they can improve, you know, the quotes that exist on chain with their own inventory. So what winter Mute does and what is actually like very close to equilibrium, very hard for solvers like ours to compete is that they do not have this block time constraints in terms of How long do they have to wait? so let me put it this way. They have private liquidity in their wallet and they can price it the way they see fit according to a lot of information that they have access to, including things that are happening off chain. And it's super cheap and efficient in terms of gas because they're not... combining like Uniswap, Curve Balancer liquidity into one complex route and trade. They just use a hot wallet to provide you an excellent quote on like Ease price or like Bitcoin price. And they just do the swap using their own private liquidity. So what we do and what a lot of other solvers does is we rebuild all the DeFi primitives that exist on chain into an off chain like trading engine. So for example, we rebuilding like existing AMMs and their mass and all the logic inside our environment. And we use Rust because we need to emulate like millions of different ways how we can split your liquidity. Like if you want to trade like 100 E's, we can split it into so many different ways. Top into incredible amount of pools. across all the DEXs that exist on chain. And this is fairly complex. It requires a lot of computational power. It requires different algorithms. There are different approaches from mass perspective, how you can work on routing and bringing these different DEXs together. A very popular and ⁓ maybe one of the most mainstream approaches is convex optimization. Not the best one, but a lot of people are using that. So I think we have market maker solvers that use their hot wallets and their own liquidity to fulfill transactions. And there are like algorithmic solvers that rebuild all the DeFi primitives from scratch. yeah, like other people can use it. So yeah, other people can use it and it creates like problems like slippage, for example, if we, you know, calculate it. **Gabe (09:51):** Yeah. **Nikita (09:53):** like rate on like using uni v4 pools and then someone swaps before us then the rate is going to be different and the user would probably get less and we don't know who is gonna like do what because the blocks are being built in real time and we don't have the information we only know like the end state of the block when when the block was built in the previous iteration also for our listeners if anyone wants to dive deeper into it i think there is a pretty good open source project called propeller heads that run Tyco where they've built an example or like the logic of the implementation. And if anyone wants to try build a router algorithm, I think starting with Tyco and with propeller heads like GitHub repository might be the best way to go and get you familiar with in general, how it works. But most of the solvers like ours, we rebuild these primitives by ourselves. And we are using different tips and tricks to optimize the way we use liquidity. And we are also connected directly to like block builders, for example. So they can guarantee us inclusion. We are also using pre-confirmations, projects like pre-mef, et cetera. to land on top of the block or very high in the block with our routes. So yeah. **Gabe (11:12):** Got it, got it. So, so. And help me if we're understanding correctly. Basically, you there's two types of players, the market makers and then the solvers. And then the market makers, it seems like they have maybe a quote unquote easier game because they're pulling from known liquidity. They kind of know how to price it. They have additional external information where you guys are running almost this like simulation of of the entire ecosystem and then predicting kind of how a trade would land. And then that's how you **Nikita (11:20):** Yes. Yes. **Gabe (11:42):** move forward or make decisions. Is that kind of a correct illustration? **Nikita (11:45):** It is correct. I just add that I think being market maker, good market maker is extremely tough and pricing your assets is an art. And, you know, some, some projects like winter mute, for example, they excel at this game and they do a very good job. Well, for new participants, it's a little bit tough, but ⁓ yeah, in general, like what you described is exactly what's happening from our point of view. It's our job is extremely important because if projects like ours do not exist and we are squeezed out, then there are only market makers left and the DeFi primitives like AMMs receive toxic flow. And essentially like now market makers provide very good rates, very tight spreads, but as if they become like monopolistic, they would improve and the rate like, like the rates for end users would get worse and the DeFi will start mimic trade fi. So it's a **Gabe (12:38):** Got it, got it. **Nikita (12:39):** ideological battle. **Gabe (12:41):** And do you think, like what is the state right now? Do you think that is the case? Do you think it's not the case? Where are we kind of trending in terms of this decentralization component versus off-chain liquidity? **Nikita (12:57):** So yeah, in 2023, we were like the largest solver on Cowswap. In 2024, another solver called Copium Capital was leader in terms of transactions and volume. In 2025, last year, we were second behind Wintermute. So market makers now are the most dominant players, and they are extremely important in order to provide the best rates for every platform. And even like new... new intent systems like Cowswap and Merge, they're not targeting to onboard algorithmic solvers like ours. They want market makers to provide quotes and to be present on their platforms. the market trend is a little bit grim. It's going towards centralization. Moreover, we now have prop AMMs and prop AMMs are kind of playing the same kind of game, trying to provide tighter spreads within the block. And yeah, it looks like it might get even worse. **Gabe (13:56):** Okay, And so for someone who is like new and is like, hey, you know, I really enjoy this. Is there space for kind of like new entrant players to, you know, come up with their own order routing system or algorithm, et cetera? Or do you think this is kind of like a sailing ship where, you know, the floor just raises so fast that it's not necessarily feasible for a new entrance to kind of pop in? **Nikita (14:22):** Well, let me put it this way. I feel like our algorithm and system was historically one of the best to interact with AMMs. But then RFQ liquidity, liquidity from Hashflow, Bebop, liquidity that is provided by other market makers have become incredibly competitive. And it's kind of a different type of game. I would say within routing an algorithmic perspective, I don't think there is much innovation that can be built. Moreover, it's extremely complex and not very profitable. So I don't envision like new players emerging or innovation happening. Actually, if you look at the past couple of years, I would say the number of solvers and algorithmic projects have been shrinking. Like Copium guys, have pretty much left the market. They stopped solving and we acquired their code base. There was a big announcement back in September this year, last year. But I would say on the AMM design, on the way how we interact with liquidity, we definitely see some innovations happening. **Gabe (15:16):** Mm-hmm. **Nikita (15:29):** One of those innovations is superposition, what we've built in a way, trying to compete with market makers and offer a different path to users, how they can like earn yield and how can they empower systems like cowswap or uniswap, one-inch fusion and others. **Gabe (15:47):** Can you tell us little bit more about the mechanism there and the challenge that basically trying to tackle with that? **Nikita (15:53):** Right. So honestly, I am super, super excited about Superposition. I feel like this is one of the most interesting primitives in DeFi. And we've showcased it to people initially at DEFCON in Argentina this year and received a lot of like tremendous feedback. But we've been competing with market makers and market makers use their hot wallet to settle transactions. And we've been thinking, how do we compete with them? Because in terms of like gas, they are... Perfect, you it's just a transfer from the wallet, you know, it's ⁓ very optimized and you price liquidity using off-chain data. So how do we compete with that without becoming a market maker ourselves? And we've decided that why don't we use like other people, community and like users, their liquidity to fulfill trades that we receive from cowswap and earn yield natively. So what superposition does and how it works is, You as a user give an allowance to our smart contract to interact with your wallet and you give an allowance on a token level. So you can choose like the most basic example we have a stable coin. So you can choose like three assets, USDT, USDC, DAI, and you say I'm comfortable holding those assets and I'm happy to provide them like to stream them as liquidity to DeFi as long as I earn yield and I receive more tokens in return. So... Yeah, you've chosen those assets and let's say you have a thousand USDC, then we use your liquidity to facilitate swap and call swap and give you in return one thousand one die. And then we use your liquidity again for another swap and then you receive one thousand two USDT. So your liquidity like always grows. It stays inside your wallet, every block. You can pause it at any time. You can stop it at any time. You can like we have... Tinkering with some fancy use cases like imagine you're going to sleep and you know for eight hours You're not gonna use any of your stable coins and then yeah We start like routing trading you wake up and you have like more tokens that you went to sleep with and it's like first self-custodial yield product like right now if you want to earn yield you always need to deposit funds into a vault into a pool the vault narrative is like popping up and we are thinking like well, yeah, you're your wallet is essentially like an ideal vault. And if we can add a couple of constraints and a couple of like use cases like bridging, for example, or like not just swaps or even maybe lending, you can earn yield while still having custody of your assets. in a way we transform, you know, your wallet from a spending account into a saving account. And I think this is really cool for like a lot of users. We've just launched in November last year and we've been live for four months. We have close to 500,000 TVL. We've connected over 100 wallets and the DAP is live. You know, it's safe. We have a security audit. Do not use your prime wallets to interact with it as the concept is very new and you can never be too sure. But you know, if... Like we plan to build more and more features on top of it. We plan to improve and, know, battle test security. And we feel like this might be the end game of, you know, DeFi in general. **Gabe (19:07):** Okay, and this is kind of what we discussing before, like a solution to competing with these, you know, the market makers who have access to this large bucket of liquidity. So you're kind of expecting, you know, this on-chain liquidity that you can tap into is kind of like a comparable solution. **Nikita (19:25):** Exactly. mean, market makers, you know, they're getting more and more dominant and they're earning yield while toxic flow and, you know, goes to AMMs and to users essentially as well. So in our opinion, it's super to build this, you know, layer and also to find new ways earn yield. A lot of yield on is not originating from on chain elements, like for example, stable coins. Stable coins primarily earn yield from like treasuries or depositing collateral. if ⁓ your stable coin is backed by USDC, then USDC is ⁓ deposited somewhere and earns yield. While we unlock yield from DeFi trading, from swaps and purely on-chain yield, yeah, like feels kind of new. At the same time, you know, if you look at the market, we've analyzed, you know, user wallets and there is like so many stable coins and other assets that are just lying idle in user wallets, not doing anything, which is kind of like economically inefficient. And we feel like we want to offer them this unlock with superposition and start generating like a sustainable yield, which is Currently, we are targeting like 7 or 8 % on stables, which is like bigger than what do liquidity providers get when depositing funds into like Balancer, Uniswap, Curve, or even lending protocols like Aave. **Gabe (20:54):** Wow, wow, yeah, yeah, that's pretty comparable. you know, just to transition a little bit, I'm curious for, on urine, since a lot of this is very much like almost high frequency trading background, you know, algorithm chops, probably a lot of. curious how you're seeing the deployment of agents and AI for this type of routing and not necessarily AI agents running around and trading with each other, but more so where do you see the actual tangible applications of AI and even with like ⁓ mythos that just launched, right? A lot of this, I think kind of like the underlying parameters that we're seeing is you can have agents reason autonomously very powerfully, right? So does this unlock new ways of routing liquidity and how does that look? Or is it just kind of like, yeah, AI's been great and we're coding 10 times faster, right? So yeah, I'm curious how, for you specifically, how is this impactful and do you see things happening on chain yet or what should we expect for? **Nikita (21:56):** Right, so. Okay, let me begin with like how we use AI. And I think every project is now going through transition or should be already at the end of transition where you move from like previous workflow towards like AI agentic workflow. We've had 16 people working full time at peak ⁓ in barter and we've downsized to 12 and we might downsize a little bit more. It's just because some of the people both on a business end and on, you know, engineering and didn't pick up AI tools fast enough and didn't use them in their workflow. Also, if you check our recent announcement, I think we've built a lot of small prototypes, niches, product. updates, different interfaces or analytical like micro tools, I would even call them, using AI that just improve user experience and overall UI of our products. So yeah, using AI is essential. You cannot get things done now or you would do it like... so much slower than it would not matter by the time you finish. So that's one side of the story. Another one is how do, what does it mean for our product for like routing and swaps and liquidity in general? And we, we are bullish on AI and agentic, not just payments, but operations on chain. We feel like agents will interact with crypto agents will be doing things on chain and they need tools for it. So we've packaged our swap API into an MCP and the same we've done for superposition so that the agents can deploy liquidity or perform swaps and also analyze every block they can get a state of ⁓ on-chain liquidity and find arbitrage opportunities or ⁓ doing... doing anything they can. Like, I don't think people are using it right now. Also, it feels a little bit risky to give access to an agent with your liquidity, with your wallet, because who knows what agents might do with it. Yeah, but this is definitely the future. ⁓ Agents will transact. Yeah, we have a... I personally don't have a trade fire experience, but some people on our team do, and they sincerely believe that the banks and institutionals will never allow agents to interact with them at such pace and speed and capacity as they can interact with blockchains. It's just like they will be limited with banks APIs and on chain they can do a lot more interesting cool stuff. So we are, we are bullish on agents. are building for them. Yeah, we are definitely building for them. **Gabe (24:44):** Got it, got it. So agents are not necessarily changing the structure of your product directly. As you're doing computation between, I don't know, like Aave or Uniswap, right? You probably have some system that is robust, and you remember refining for the last few years. And I guess on those pieces specifically, are you guys implementing like... **Nikita (24:57):** Right. **Gabe (25:09):** agent logic or any like autonomous decision making at all or do you think that doesn't even, is that not even required for your infrastructure or even kind of like the nature of solving, et cetera? **Nikita (25:21):** We've been debating it for a while and for now it looks like, know, mass and optimization, all it takes to, you know, build an algorithm and start routing in DeFi and, you know, utilizing liquidity on chain effectively. So agents will not improve much. However, there is one use case. where we find where agents can help us. And that's when like volatility spikes happen. So in the past, when, you know, like suddenly the market would move like drop dramatically or would start to rise in dramatically, the block space would become extremely competitive. The AMM pools would be like heavily abused by MEV bots. **Gabe (25:46):** in **Nikita (26:05):** And, you know, routing for us becomes a little bit more complex because we have to account for higher slippage and, higher gas costs. And that's where AI can do like brilliant things by predicting the market conditions based on the previous cases where things have happened. can also, yeah. So this is one use case where we've adapted AI for our algorithms and logic, but they do not. affect day-to-day operations. It's just for, you know, this volatility cycles. **Gabe (26:35):** Got it, got it. And do you see this kind of trending in that direction? Or do you think, you know, for your product, it still is too early? Yeah, I'm curious on that piece. **Nikita (26:45):** Excellent question. I don't know for sure. ⁓ I think it feels like mass does the work. I would say that probably AI lowers the barrier of entry because for example, one of our co-founders algorithmically, he have been building a shipment software for over nine years and the software is used by giants like IKEA, Philip Morris, Pepsi to... bring the goods global from different factories into different countries. And he told us the same mathematical problem, how you optimize shipment of assets and how you optimize DeFi routing. And it's funny that I was just at ECC Cannes and I was speaking with trade five people, institutionals, and there were a few engineers from Fidelity. who instantly said, wait, like, like, routing sounds so much similar to shipment. And yeah, you know, it was the first time someone told me that outside of our team, but this was our alpha. And this is what like this experience, his experience helped us build a very good system that was extremely competitive, like in 23, 24. Now I think anyone can kind of do it, relying on AI, relying on like, you know, public knowledge and even my be able to try like different concepts and stuff. **Gabe (28:05):** So you guys are looking at solving as almost as logistics essentially. **Nikita (28:10):** Yes, this is how our algorithm is built. **Gabe (28:14):** Interesting. yeah, without spilling that, the alpha, how do you frame shift between kind of thinking of them as, you know, trading versus supply chain, **Nikita (28:24):** Yeah, well, our vision is that, you know, AMMs are being abused and not earning enough yield to be competitive. Moreover, know, another thing we have in mind is that DeFi is still extremely complex. let's look at the, you know, what happened with the infamous, notorious, know, ⁓ swap on Aave when the user lost like $50 million. That happened like a couple of weeks ago. The swap went through Cowswap and one of the solvers for some reason didn't have like the best security constraints and executed swap at a terrible, terrible rate for a user. Nobody knows and nobody yet fully understands what happened and who was this user and what happened with the funds that he was trading. However, we provided a couple of feedback to different media news agencies, for example, DL News and Blockworks. we've seen the swap was so notorious that it was covered even in media like Forbes. But a lot of data and information that was inaccurate in terms of who profited, what actually happened with the funds, and where is the issue coming from? So we think one of the issues is that DeFi is still really complex. imagine you're given approval first, you wait for ⁓ a block time, so like 12 seconds, until you approve assets to be used in a certain protocol. Then you perform this swap in action and you wait for it to settle. are incurring like slippage. So DeFi is still very, very clunky. That's why the volumes are not there. I've spoken with a lot of different founders and we all agree that DeFi haven't happened yet. It's still small, it's still niche. It's by far the best use case crypto has, but it hasn't like fulfilled all its promises. And we need to make it user friendly. And this is what we do with barter, with superposition because user doesn't need to know what's happening. Like they can check. They can verify the logic, but users are not monitoring day to day what's happening with their liquidity. Moreover, if they've given approval to us next time when they will want to swap those assets, they have like one less action they would need to do on chain because we already have an approval. So, yeah, this is our vision on what's happening and what's not working and how do we bring DeFi to the next level. **Gabe (30:52):** Yeah, well, thanks for that. And I'm curious for you, what are you kind of seeing now, especially during this quiet, I know you said before we got on the call, Ken was really good. What are you seeing as the things that you're pretty excited for, kind of where people should be looking at? we wanna see where the alpha's at, right? or the signal essentially. And maybe it's not like, buy this coin, but more so like in what direction should we be looking when it comes to your domain superposition? I think you're saying one of them, yeah, with that in mind, like how are you looking at that? **Nikita (31:30):** So let me compare the market sentiment I felt during Consensus Hong Kong and East Denver, which happened in February, and East Cannes. So in February, during those conferences, there were much less users, and people were a little bit confused in terms of what's actually going on. And, you know, like all the market trends were kind of going down, you know, the token prices, the volumes, the active users on chain, blah, blah, blah. So, yeah, everyone was quite pessimistic. However, what I've seen in Cannes is that we have a lot of real builders. We have quite a good amount of theoretical use cases and they're not yet, they're not even theoretical. They're already... There are like MVP, you know, proof on chains that they work. They just need to scale operations and get people and they, you know, grow it. And, you know, this feels like the flywheel is starting to get back on track. And they were like real builders in Denver. Everyone was, you know, like directly related. There were like less random people, less traders, maybe less people doing BD. I'm a BD person myself, but it's kind of good. good to see a lot of technical talent, engineers, people who fill the product and update it on a daily basis. about what's exciting, I feel like, imagine I have like a thousand USDC and you have a thousand USDC. And if I deposit it into Uniswap, I get one sort of yield. If you deposit it into Ava, you get like a different sort of yield. And... **Gabe (33:10):** Any. **Nikita (33:11):** our yields may vary by various numbers. Like I would get like 3%, usually like Uniswap LPs earn like 2 to 3%. And the other, it would be higher, it would be like 4, 4.5, maybe like 5%. So essentially the same, know, assets power different primitives, earn different yields, someone earns less, so someone is like less aware, and essentially will move towards where he can earn more. But I feel like there is a lot of convergence happening between those DeFi Legos. And everyone was building different parts of the stack. And now people are building everything under one hood. So what I'm trying to say is that projects like Fluid, for example, their liquidity is deposited into them. is used for both AMM liquidity, like for swaps and for lending. So it powers like essentially those two primitives. And I think this is nice. I think we need like a same kind of shared liquidity layer that would power bridging, that would power swaps, that would power like lending. Our superposition is one way of doing it. There is a one inches building towards a similar use case. Their platform is called Aqua. It's not yet live, but they've built a developer SDK. I also think there are like a lot more things that are happening with stable coins and real world assets. ⁓ It's ⁓ surprising that a lot of like DeFi people, actually want to trade real world assets on chain. And I think we should allow them to do it. And we've had some big names, pioneers like Ondo, but Ondo reminds me a lot of Bencore or Kyber network, would say like early DeFi 2018. And those platforms were gated. So for example, I couldn't have built a pool on top of Kyber network. need to do like QIC, QIB. need to provide like legal opinion letters that the token I'm bringing there is not a security and stuff like that. Uniswap didn't have this constraints and we see new real world assets that offer kind of similar. user experience without like gated entrance. yeah, people are loving it. People are experimenting. People want to buy like oil and like trade it. I would say a lot of use cases are still kind of, you know, not gambling, but I would say trading and, you know, get rich fast oriented. However, **Gabe (35:29):** Mm-hmm. Yeah. **Nikita (35:43):** Yeah, yeah, yeah. In the long run, it always bring innovation and bring like sustainable good use cases that will benefit everyone in the industry. So this is, this is my take. **Gabe (35:53):** Okay, so basically this like collapse of the fragmentation is this big piece and then some extended assets that can be traded whether it's like RWA, know, tokenized stocks or commodities, etc. **Nikita (36:07):** Yes, I also want to add that I feel like Ethereum Foundation started to play a much bigger role than it used to be because historically, know, Ethereum was always very decentralized and it was kind of hard to like engage with the core team and the foundation. And now they are helping, you know, bootstrap daps and they've hired a lot more people to coordinate. Sounds very refreshing and I think it will help smaller teams pioneer new use cases. also the fact that L2s are trying to specialize and a lot of activity is going back to Ethereum Mainnet. It also feels nice and lowers liquidity fragmentation. **Gabe (36:50):** Cool man. Yeah, this has been super great. I think a lot of people don't quite understand how much thought goes into kind of under a swap and it's not just like, you know, concentrate liquidity or even... you know, like full range, these two mechanisms that people don't even really know about themselves. But there's like a whole underworld network of money movements and routes and ordering. yeah, so yeah, appreciate you coming on. Yeah, if there's anything we didn't cover, like happy to have you share a little bit more. **Nikita (37:21):** Yeah, no, I'm super excited about the future of Ethereum, DeFi, crypto in general. I think there are a lot of synergies with AI and of course, AI has a faster adoption pace and solves a lot more use cases. However, crypto is solving finance, it's solving transparency, this is what we need. yeah, I also feel like building in crypto is incredibly tough and... for all this like DeFi primitives, one simple mistake can lead to like catastrophic losses by users, liquidity providers, like other protocols as well. We've seen a couple of nasty exploits recently like with Resolve and Rift. And yeah, I feel like people should just be aware that it's such a complex part of the stack, but it's very essential and it's when it succeeds. It's gonna be huge for everyone and I'm bullish it will happen sometime, you know, next year or so. **Gabe (38:18):** Yeah. Cool, man. Well, thanks. Yeah, Nikita, thanks for coming on and sharing the ins and outs and hope to have you soon. **Nikita (38:27):** Thank you so much, Gabe. You know, really appreciate everything you guys do at Shoal Research. I think you guys provide some of the best insightful pieces on the market. Our team reads all of them. Yeah, thank you. **Gabe (38:38):** Appreciate it. Yeah. Take care. **Nikita (38:40):** Take care, bye bye.