The Breakdown - The Founders of Chainlink and Synthetix on DeFi, Derivatives and 25 New Decentralized Price Feeds

Episode Date: January 31, 2020

Yesterday Chainlink released price reference data for 25 of its decentralized oracle networks which, together, power more than $100m in DeFi.  On this special interview episode, @nlw talks with Sergey Nazarov and Kain Warwick, the founders of Chainlink and Synthetix respectively about:  The evolution and goals of Synthetix, a novel type of derivatives exchange where users can interact with any asset with a price feed.  The challenge Synthetix faced around spinning up their own oracles around price feeds The history of their collaboration and how Synthetix came to work with Chainlink Chainlink approach to building decentralized oracles for data such as price feeds Chainlink’s announcement yesterday about the new published price reference data for 25 oracle networks The state of the idea of decentralization, and how what was previously a concept is becoming operationalized One thing that gives them pause or scares them about DeFi and crypto and one thing that makes them excited for the future 

Welcome back to The Breakdown, an everyday analysis breaking down the most important stories in Bitcoin, crypto, and beyond, with your host, NLW. The Breakdown is distributed by CoinDisc. Welcome back to The Breakdown. It's Friday, January 31st, and today we have something special and a little bit different. Chainlink is a decentralized Oracle project. What that means, in short, is that they take real-world data and make it available for use for. for smart contracts. D5, for example, needs basically 24-7 access to real-time market data like prices for those

applications to be able to function. Yesterday, ChainLink published price reference data for 25 Oracle networks. And the whole idea here is that ChainLink is taking a decentralized approach to sourcing price data so that the Defy applications that use these price oracles aren't relying on a single centralized source. Chainlink applies the same sort of decentralized, node-powered logic that blockchains do to sourcing price data. And according to Chainlink, these Oracle networks collectively secure more than 100 million in value in defy applications. So today, on the breakdown, we're going to be speaking with Chainlink founder Sergei Nazaroff. But to make this even more real and put it in

more context, Kane Warwick, who is the founder and CEO of Synthetics, will also be joining. Synthetics is a DFI platform that enables decentralized derivatives exchange. So basically allowing people to short and long a variety of crypto and real world assets, but it does it through synthetic tokens. So what does that mean? In the way that maker users stake ETH and now other assets to mint die, synthetic users stake over-collateralized positions, but to mint a huge array of different what they call synths, which are basically any asset with a price feed. So that could be crypto, Forex, commodities, or even new baskets such as a bundle of centralized exchange tokens. And when they've minted those synths, they can take short or long positions on them.

The reason I wanted to have these two on together is that they actually have been collaborating for more than a year now. And it was the importance of price feeds that brought synthetics together with chain link. Going back more than a year, these projects have been collaborating to help synthetics move away from a centralized oracle that they spun up to these decentralized price feed oracles that are built by chain link. So in this conversation, Kane and Sergey and I talk about, one, the evolution and goals of synthetics, two, the challenges synthetics faced around spinning up their own oracles for price feeds. Three, the history of the collaboration between synthetics and chain link and how they came to work

together. Four, Chainlink's approach to building decentralized oracles for this type of data and other types of data that are relevant for smart contracts. Five, Chainlink's announcement yesterday about the newly published price reference data for 25 Oracle networks. Six, going a little bit broader, we talk about the state of the state of the this idea of decentralization and how it has evolved and what was previously a concept is becoming operationalized.

Seven, finally, we talk about one thing that gives them pause or scares them about defy and crypto and one thing that makes them excited for the future. Now, one more quick note before we dive into the interview, because we wanted to get this out as quickly as possible and as close to the news that Chainlink had yesterday, this is less edited than we normally would. It's much more raw and uncut and just the conversation as we actually had it. So let's dive in. All right, guys, we are here with Kane Warwick from Synthetics and Sergey Nazaroff from Chainlink. Guys, thank you so much for joining. Thank you. Thanks for having us. Yeah, thanks for having us. Today is going to be really fun because rather than kind of a normal,

hey, you know, tell me about your project, how'd you get into crypto kind of interview. We get to have something that I think is a little bit more interesting and more dynamic where two projects that are really pushing the boundaries of defy and crypto more broadly that have worked together are almost getting to share like a case study of how a collaboration has worked, how it's enabled something that wouldn't have been possible in exactly the same way without either party. And I also, I think it's a chance to get a broad, view on just the state of defy and decentralization in general. So I'm really excited for this conversation. And where I wanted to start, I think, is Cain, if you wanted to tell us just a little bit about

synthetics, what you're building, what you're trying to do, and kind of where things are in the market right now for you. So we have a live product. You know, it's been on Ethereum for over a year now, 18 months or so. And what that enables people to do is get exposure to different asset classes within Ethereum, within the Ethereum ecosystem. So we've got a number of different assets that you can hold in the form of a token. So a gold tracking token, a silver tracking token, a bunch of different Forex tokens. We also have different crypto assets that you may not be able to get exposure directly to on

Ethereum, so things like B&B, for example. What the project allows you to do is hold those tokens and convert them from one to the other using an exchange via repricing. So we essentially allow you to reprice an asset that you're holding into this new asset. So if you're holding a bunch of Bitcoin right now and you want exposure to gold instead, you can go to our exchange, which is a decentralized exchange. You don't need to go to a centralized service. And you turn up with that token and you can convert that token at the current price. And this is kind of the point of this conversation. Where do these prices come from?

You can turn up and say, okay, you know, the conversion price between gold and Bitcoin is, you know, X. And so I'm going to convert my $100 with gold into $100 with the Bitcoin. And what that means is that within the Ethereum ecosystem, you can get price exposure to a whole range of different assets. And obviously that, you know, that's going to grow over time. as we add different asset classes, things like equities, you know, potentially ETFs, indexes, et cetera.

So, you know, that's the intent of synthetics and that's what we're building. And obviously it's something that is still, you know, in progress. We're still working on and still, you know, kind of growing the project. But, you know, even today you can go to synthetics exchange and, you know, immediately get access to, you know, over 20 different assets. So let's actually zoom it back a little bit just because I think it's incredibly fascinating. and it's one of these projects that the more I dig into, the more interesting it gets. What was the spark of motivation for it, right?

Like, what is the interest that you have? And maybe this is kind of commentary on the larger world of just derivatives and the importance of derivatives in emerging markets like this. There's always an access component to this, right? You know, Ethereum is this sort of universal computer that people can access. and the idea that if you have every single asset class on Ethereum, then anyone in the world who has access to an internet connection, you know, can access this financial infrastructure.

So I think that that is as a really critical component of it. But like most startups, that wasn't necessarily the end goal that we started with. We actually started building a stable coin back in 2017, which turned out to, you know, not necessarily be a great fit. for the market. And so, you know, over kind of a series of several pivots, we've moved away from this idea of a stable coin to, you know, synthetic assets generally and now to this idea of, you know, a synthetic asset exchange where people could come and trade synthetic assets. So it's been somewhat of a journey over the last couple of years to kind of get to this point. But, you know,

we think we have something now where, you know, there's genuine product market fit and there's demand for it. A friend of mine actually a couple weeks ago was just describing he was raving about synthetics. And his description was synthetics is maker, if you could mint anything as well as die. Synthetics exchange is Bitmex decentralized, K-Y-C-free, on-chain with Ethereum, no slippage, no counterparty risk, of an exchange shutting down, never give up your assets like you do on a centralized exchange, which I thought was interesting shorthand.

So I don't know if that resonates, but I think that's pretty amazing. Is your friend looking for a job in marketing body chance? because that's a pretty app description. I keep telling him he needs to do a podcast specifically on defy and describing new products, new protocols. So I guess then the next question is, how has the reception been? What have you seen as you actually put this out in a market?

Because I think another part of this that's great for this conversation is that this is not just a theoretical thing, right? This is something that's live in market. You're getting feedback from users, from investors. So what are you actually seeing in the market? Yeah, so it is, as I said, a project that, you know, is still being built out. You know, even today in beta for the next version of synthetics exchange.

So there's a lot of functionality that's still missing. A lot of things is being built. And, you know, we're actually halfway through our transition, our price speeds, right, from, you know, our own centralized article over the chain link. So even that process is still, you know, somewhat in flux. But I think the overall reception in the market is people are excited. They're excited about this idea. that these pulled collateral models,

and without going too deep into that, this idea that someone can turn up and they don't need to find a counterparty to be able to trade. So, you know, we've seen things like Uniswap, obviously employing these structures. But the idea of being able to turn up and trade directly with a contract on Ethereum as opposed to finding a counterparty to be matched with,

I think it's quite powerful. And that's one of the reasons why people are so excited about, you know, the potential for the project. The idea that you can have a decentralized bidmex that runs on Ethereum and is genuinely permissionless, which again is something that we're working towards is pretty exciting. This may be getting a little bit in the weeds, but just for people who are interested, are you seeing consistent trends around the volume, right? Like what people are interested in investing in, right?

What pairs they're trading? Or is it all over the place? It is a little bit variable. And there's some other confounding variables in there. as well, which is that, you know, being a missionless project on Ethereum, we also have an attack surface. So there are people who are always trying to attack the project, you know, front running, Oracle attacks, things like that. You know, there is this ongoing kind of arms race as we try and deal with some of the people that are trying to attack the project, which does add some

noise into the tracking of what is the activity that's happening on the exchange. But there's a pretty decent trend and growth of trading between some of the assets like Maker, for example. And we obviously have link on the exchange as well. So we're seeing some of the, I guess, smaller assets, not necessarily Bitcoin or Ethereum generate more interest over the last, say, two to three months. Makes sense. Okay, so now let's transition a little bit. You kind of intimated this as you were talking about what's in development,

but this is obviously the type of project that requires particular infrastructure. And so you brought up the issue of oracles. And so maybe this is a chance for you to talk a little bit about, A, the challenge that you started to run into around oracles and price data, and B, how you got to know chain link and how that relationship evolved. back in 2017, you know, in 2018 when we were building the project, there wasn't a, you know, decentralized Oracle network, right? There wasn't even a centralized Oracle network that you could kind of consume. There are a few places where there are price fees on Ethereum, but, you know,

was never going to be able to support our use case. So we were kind of forced into this, you know, not ideal situation of rolling our own Oracle, you know, and in crypto, when you roll your own almost anything, it generally tends to be, you know, quite a challenge and probably suboptimal. We kind of knowingly did this, and then, you know, we sort of hoped that we'd be able to build something that was robust enough that, you know, it wouldn't break down. And then we had an incident in one of the price feeds for the Korean want that created this, you know, attack surface, and there was this huge incident that needed to be, you know, essentially rolled back with the attacker, you know, and I think that that was one of those moments.

where it became very clear that this thing that was background risk that we'd kind of accepted from the start when we'd built out the project had become kind of an existential threat to the project. But even before that, we were talking to the chain link team, I think it goes back maybe almost even two years that we started talking to Sergei and the team and we made a decision that we would transition and they've obviously worked really closely with us

to understand, I guess, specifically what our needs were and how we were consuming price speed. So it's been a long process to get here, but I think we've started the transition and we're close to finalizing it and being able to shut down one of the major risks in the project, which is the centralized Oracle service that we run.

So this is a perfect transition, I think, Sergey, to bring you in and have you explain for those who don't know a little bit more about chain link and what chain link is trying to do and the type of infrastructure is trying to provide for projects like synthetics. Yeah, yeah, absolutely. I mean, even before that I just, I just want to say that I think what Kane and Justin are building at synthetics in terms of making smart contracts, go to the next stage of what they can actually accomplish is the thing that actually got me involved in this whole space and is really the future. So before I even jump into what we do, I think it's extremely exciting what they're building and it's kind of the next iteration of what's going to come to define blockchain.

use cases and what blockchain-based infrastructure is used for in general. What Oracles do is they basically provide data into defy applications like Cane Synthetics. Now, data is quite important for financial products because the financial products are essentially written around the data. So there's something like a settlement price or there's the price at which a trade happens or all these kind of financial products function on the basis of they're being going to agree upon price on which we exchange things or upon which we settle the contract. And so the accuracy of that price is very important because it essentially determines what the financial product results in. Now, the financial product in the case of synthetics, in all the exciting ways that your

friend Nathan mentioned earlier, runs with all these great features of, you know, there's no custody and it's decentralized and it has all these great guarantees that the traditional financial products simply don't and won't be able to unless they use blockchain-based infrastructure. But they have all those guarantees because they use an infrastructure like Ethereum. That infrastructure allows the actual contract logic to function and to define the conditions of the contract. Now, as you build more advanced, you know, more kind of useful, contracts like the ones that Kane and Synthetics are building, you come to a point where you need to know what's going on in terms of data. So you need to have market prices to settle a

contract. You need to have market prices to exchange two different tokens of value, as King described. And because the smart contracts, key capabilities are that it provides all of these security and reliability guarantees, you actually need this expanded version of a smart contract that Kane and Synthetics are working on to provide that guarantee both at the level of the logic that in this case lives in something like Ethereum and in the case of data delivery that effectively triggers the contract. So you need guarantees for both the data triggering the contract and for the logic of the contract. Now, Ethereum does a great job at processing that logic and making it accessible to a huge amount of private keyholders that can use it. And the problem that Kane described

was that providing end-to-end security such that the data triggering the contract is also secure is a large challenge that, as Kane put it, there's no real need or benefit to having to take it on. And the thing that we do is we build a piece of software that securely solves this Oracle problem, this data delivery problem, such that data is delivered at the same level of reliability as the contract code that's being executed. And so now you have an end-to-end assurance that the contract will function correctly. So you've expanded what the contract does. You've maintained the security of it, and this avoids the types of situations that Cain described. That's kind of what we focus on is providing this infrastructure and enabling exciting kind of next generation smart contracts

and derivatives smart contracts like what Cain and synthetics are building. So it's been a pleasure to work with them. And I think from their point of view, it's logical decision because they no longer need to expend an extremely large amount of resources, both building an Oracle mechanism and securing an Oracle mechanism, which is a large undertaking equivalent to building a piece of infrastructure like a blockchain and in some ways, you know, with its own very specific levels of complexity. And I think the other good news is that we've been able to do this so successfully so far that we've arrived at a recent release of over 25 of these reference data networks. And we've recently listed them all in one place on a feeds listing page called feeds.

chain.link. And one of the things we highlight there is actually how, in addition to being able to solve the security problem of how do you successfully deliver data to a very efficient, high-quality smart contract like synthetics, you also need to solve the economic problem of how are Oracle networks supported? And we're very glad to say that in working with synthetics and other great DFI DAPs, they are able to essentially pay a fraction

of what it would cost them to put that same data on chain from a single centralized note. So I think there's actually two wins here. One win is the ability for DFI DAPs, like synthetics to have high quality infrastructure that takes an infrastructure and a security problem off of their hands and allows them to focus on launching more great markets

and more great products. And on the other hand, they get to pay a fraction of what they would have to pay in order to broadcast the same data themselves while getting anywhere from seven to 20 plus times the security that they would get otherwise. So really, I think it's a kind of a win-win situation, both for the defy DAP developer using the infrastructure

and for the larger kind of ecosystem, the larger decentralized infrastructure that's out there to enable DFI. Okay, so this is great. And I want to dig a little bit more into this. But so for people who are listening, who are just kind of trying to wrap their heads around this for the first time, and really peel back the inner workings.

What you're talking about, Sergey, is effectively allowing these defy DAPs that rely on prices to trust that those prices are accurate, right, that they're not tampered with, that they are, and to do so in a way that doesn't require one centralized source, right? It's one centralized point of trust. So I guess just, you know, peeling it back even a little bit further from what you've just released today to just in general how you guys focus on this problem, how do your decentralized price feeds work, right? How do they change the paradigm from just grabbing a price feed from one source to actually making it something that people don't have to trust a single source to acquire that data? Yeah, of course.

I think one of the most important points is to understand what the outcome we're seeking to achieve. And the outcome that we're seeking to achieve is that token contracts can be written on a network like Ethereum and they can be entirely secured by that network. But when you write a contract that needs to interact with external data, systems like Ethereum do not have the capacity to give that data to a contract. So this is the Oracle problem, and this is when people are forced to build a system or use a system that provides that data. But that data is a critical part of this contract. So you could either manipulate the contract code or you could manipulate the contract code or you

could manipulate the data. In many cases, they lead to equivalent outcomes and they're almost equivalent as attack factors. So I think that the first nuance point to understand is that there's a more advanced class of contracts than just token transfer. That more advanced class of contracts wants to be more advanced, but it also wants security. And in the process of getting the data that it needs to be more advanced. It needs to retain the security that the contract code itself has, but that the data being delivered doesn't. And then the logical question becomes, how do I apply the security model that secures the contract code to the data delivery, to provide the same assurance about the data that's essentially controlling that contract?

And the answer is actually very intuitive. You basically create a form of decentralized computation, which means you have fully independent node operators redundantly doing a similar computation and coming to consensus about the accuracy and the outcome of that computation. And this is what an Oracle network is. An Oracle network is a collection of fully independent node operators,

which essentially go get data, validated across the group of fully independent node operators to the point that it's sufficiently accurate and sufficiently validated that it can be considered reliable enough to trigger a high-value smart contract, which is guaranteed to then act on the data that it's given. And what we do is we make a piece of software that node operators, fully independent node operators run by leading dev-op and security teams in the blockchain space, some of which secure, you know, over 70 million in USD for various staking protocols.

Right now, the networks that we have, that we have live, the Oracle networks we have live, I think at this point they secure over 150 million in defy on Ethereum alone. And then, you know, there's other amounts in other environments. And I think the answer to your question is that we're taking decentralized competition, as a security model, we're applying it to the retrieval and validation of data such that

a group of fully independent node operators guarantees the correctness of data to a sufficiently high degree that it can be accepted as secure by something as secure as Keynes Synthetics Smart Contract Code. And once you give people the ability to have both the ability to write the code and the ability to interact with all the external data that they need for the code to, you know, know, know about significant market events or IoT events or any collection of external data, I think what the combination of those two capabilities does is it greatly expands what people can build. And that's kind of our goal is to enable great teams like Kane and Keynes and synthetics to really build something that that isn't just about token movement. It's about creating

a contract that is secure end to end, but also interacts with market prices in the real world in a more meaningful, more useful way. And I think that's being, I'm very, I'm thrilled to see that being proven out in the case of synthetics through all the usage they have. And I, you know, I think it makes complete sense to me. just to jump in here. You know, I think we, our project comes from this kind of pre-history phase of of Ethereum in a sense that like you couldn't consume this data unless you built your own infrastructure to get it on the blockchain, right?

And I talk to a lot of teams, you know, small teams that are starting to build things today. I think we are seeing a shift now where there is just this expectation. that this data exists, right? And, you know, that it will be there, that it will be available, and that they will be able to just consume it. And I think that that's really exciting, you know, to sort of see because we didn't have that. You know, we're only talking, you know, months or years in terms of, you know,

when we started this process. So it is really exciting to see this. And I think we're just going to see a proliferation of different, you know, ways that people are going to consume and implement this common of computation and data access. So it's really exciting, obviously, for us as a project to have access to this now and to be making this transition and remove this attack vector and risk. But I think, you know, as an Ethereum user myself, you know, I'm really excited to see

all of the new stuff that's already being built and starting to emerge. It's really cool. Yeah, I mean, I think in some ways, and I don't know if you guys would agree with this characterization, but I feel like one of the more important shifts, over the last couple years is moving from speaking the language of decentralization to actually operationalizing decentralization, which I think is a lot of what you guys are talking about, what you spent so much time on. And I guess that brings up for me a question of just maybe zooming out even farther from your projects and the collaboration that you've had.

How do you see the state of decentralization? And maybe just we can limit it to defy right now. You know, it's something that is obviously one of the most important areas of development, one of the most focused on areas of development and building and experimentation in the crypto markets. But, you know, how far along are we? How early are we? Just for, again, this is assuming that not everyone is as thick in the weeds of defy as some of us are. Yeah, I mean, I'm happy to start because, you know, we get this question a lot, right? you know, how decentralized this synthetics.

And I think one of the challenges with DFI is that, you know, a project or a system is only, you know, decentralized, or is only as decentralized as the most centralized component, right? And so I kind of alluded to this that it doesn't matter what the attack factor is. If the ultimate, you know, end state is that, you know, funds are stolen or compromised, then, you know, that it's kind of irrelevant, right? And so, you know, if the attack vector is that there's some centralized component that the team controls, right? And this actually came up yesterday where someone reached out to me and said, you know,

is there a list of kind of functions that the team has, you know, in terms of what it can do, what it can control within smart contracts? And my answer was kind of no, because we actually have total control, essentially. to redeploy the contracts and redeploy the logic. So so long as that is the case, the system is extremely centralized. But the interesting thing about that is that we've kind of accepted that so long as we were controlling the Oracle, having that centralized control of the contract logic is not really an additional attack vector. The end result is the same.

And so one of the really important things I think that's coming up for us is we will reach this point where we have actually handed over, you know, the control of the data feeds to the chain link node operators. And just as an aside, it's been really crazy in the last month. I've had four or five different conversations where I'm talking to someone who's building a really interesting project and they actually throw out there. Oh, by the way, I'm actually operating one of the nodes that's feeding. some of the synthetics data feeds. And these are some of the smartest people in the space that I've been speaking to. And it's been super interesting to see that there's all these people in the background that are actually operating these nodes.

But I think the end result is we will get to a point in the next, say, month, where we will need to make a decision, you know, as a project, as a community, to start to kind of hand back some of this control, particularly around the logic, given that, you know, we've now solved this kind of centralized Oracle control problem. And so I think it's going to be a challenge that a lot of projects are going to, you know, need to grapple with is, you know, how much control should we have, how decentralized should we be, what is the process of slowly decentralizing? Because if we had started off as a, you know, in quotes, I guess, fully decentralized project, whatever that means, we wouldn't have been able to make the pivots that we've made to kind of get to this point of product market fit. So there is a trade-off there between your ability to respond to what the market actually wants you to build and to kind of protect the decentralization of the project. And so I think this is something that's still so early that we don't even have a proper language to sort of describe it or even discuss it.

And I think that that's slowly emerging, but we're still very early. and it's going to take, you know, a bit more time before we can, you know, have this shared framework to even discuss the areas of centralization, what the tradeoffs are, and, you know, have really good disclosures about what those tradeoffs are in any given project. One of the most, I think, resonant critiques of almost the fetishization of decentralization is that it is in and of itself a word that describes a state of being, not an outcome or a goal, Right? So people say, like, when we talk about decentralization, we need to talk about what the end goal is. Is it to reduce these attack vectors where you could have economic capture or some other type of problematic situation? Is it to ensure some type of censorship resistance, right? And that really where we need to be working back from is not this monolithic notion of decentralization, but whatever the end goal of decentralization is. And that might be different in different contexts, in different protocols, and different systems. And again, that to me seems like an evolution of how we're thinking about this as well,

even just from a language and self-understanding perspective. Yeah, Nathaniel, I think that makes perfect sense. You know, of the trends that I'm seeing in general in the DFI space, I think there's two significant trends that I think are going to have a large impact in the coming months in this year. I think one of the first things that I'm seeing that I'm very, very hopeful about is I think we finally reached a certain critical mass of private key holders that create a user base of people that can use products like synthetics and that can use all kinds of other different smart contract,

you know, DFI products. And I think what that really results in is it results in an environment where a high-quality team like Keynes can show up and build a successful defy application and succeed at at the same at a similar level to web startups. And I think that pattern of success is going to become more and more common because even if you if you go back three or four two years ago, you didn't have as many people with private keys that could use a defy application like like the one that synthetics has made. And so there was just a timing issue with with being that forward looking and building something that futuristic and that great. And I think now that that's going away and that

we'll see a pattern of success because there's a user base to consume DFI products, I think that's going to greatly, greatly accelerate what the space can be about because at the end of the day, you need economic activity and you need users to define the success of any application and web-based or decentralized. And so I think that's a huge change that has just recently happened. I think the second thing that that's very exciting is kind of what Kane mentioned before about him seeing people building all kinds of applications that combine data and smart contract logic. And I think that Kane's point of view and his story about how when they started out, they were forced to build a piece of infrastructure. It's similar to the story of some of the very early early days of even before Ethereum and when people were building smart contracts in like 2014. 14 and 15 and stuff, people would go ahead and build a blockchain infrastructure because they

wanted to build some kind of use case, some kind of smart contract use case. And I think the reality is that if there's a data rich environment with a lot of inputs and outputs, similar to the environment that web developers quickly build web applications and quickly iterate on those web applications, and if the decentralized kind of application, and the smart contract environment where developers build all these defyed apps can be a place where people can very quickly both write the logic of the contract and get all the inputs and outputs. So all the data and all the payment capabilities, I think what we'll see, you know,

and putting that together with the user base that could consume whatever they make, I think what you'll see is you'll see a much greater speed of iteration and a lot of people launching more expansive and just all kinds of different products quicker, which once again, when coupled with that new user base that can consume those products in a sufficient large quantity to make them successful, I think it's the type of thing that can come together to help redefine our space into decentralized finance. There's successful decentralized finance companies. They built a secure application that is secure end to end in a way that no other application, no centralized application can be, it can provide guarantees that no centralized financial product can provide.

And guess what? You know, they built all that and they became extremely successful because there were enough people with private keys that wanted to consume that application. And if that starts happening, then the story of sales. success of our space switches from being about tokenization and moving tokens around. And it goes on to people building all these next generation, truly censorship resistant, truly decentralized, truly kind of fair and kind of fraud-proof smart contract applications that we're all excited about. And that's, I mean, that's the really exciting thing for my point of view is that the space could quickly become about that type of economic activity,

about those types of applications between these two dynamics, which is what I'm very excited to see. So I wanted this, I think this conversation has been super interesting. We could talk for a lot longer on so many ins and outs of the particular projects that you guys have, but also the industry as a whole. But where I wanted to leave this maybe is just put you guys on the spot a little bit. So I'm really interested in two things. First is something that that keeps you up at night when you think about, you know, the, what's still to be built and the challenges that either your particular project or just the defy space in general are facing. But then second, I'd love to hear something that fills you with optimism that makes you excited. And I, you know, Sergey, I feel free to cheat

and kind of go into a little bit of what you were just saying because I think you almost answered this and kind of inspired the question. But what's one thing that is really stressful and maybe something that we just need to be really diligent about being level-headed and clear-eyed about? And what's one thing that gives you a lot of optimism for the future of defy or other parts of crypto? Yeah, I'm happy to jump in here.

I think something that concerns me, and I know, you know, in Nathaniel, you touch on this quite often, you know, in a number of different places. is the potential for the impact or the response, I guess, if, you know, what Soga was describing starts to happen, right? You know, if we start to get this inflection point where there's all these users flooding in, there's all these different use cases, all these great new products emerging, you know, people are, you know, regulators, whoever are going to step in and say,

hang on in a second, we want to have a bit more control here. And so I think one of the sort of pattern that has been, you know, implemented over the last four or five years is this idea of having foundations that operate these networks, right? And, you know, we have a similar pattern. And it's a pattern that I think exposes a project to, you know, regulatory capture in a way that is, is not ideal. And so we're actually going through this process of unwinding this and moving away from this idea of having, you know, a single entity that maybe governs the system and moving more towards like a Dow-like framework. And so that's something that definitely worries me is, you know, that we have, I think, a window of opportunity, you know, over the next maybe six to 12 months

where we can get the governance structures right for these types of projects to ensure that they are sufficiently decentralized. And then I think the thing that makes me optimistic, is that, you know, I talk to a lot of, you know, small teams of, you know, two, three, four people that are working on incredible projects that I think are really going to drive some of this, you know, user demand. And, you know, there is this latent demand there. But I think as, you know, more products emerge, you know, and people start talking about all of these cool new things that they're using, it will actually bring in a flood of new users. And I think as we solve each of these problems, whether it's the Oracle problem or, you know, as we have

viable solutions for teams to hook into within Ethereum, it's going to, you know, create this platform that is, you know, as easy to build on as, you know, web to or, you know, or whatever. And I think that that is going to have a compounding return that we really can't kind of anticipate right now. You know, we're still super early. And so for me, the really exciting thing is, you know, every day people are solving these genuinely hard problems. And each time they solve those problems that solve for everyone, you know, because the system is open and people have access to it and people are talking about it. And people can see that that solution now exists and is out there. And so I think that we just have the ability to, to kind of consume all of this effort

that we're all putting in a way that creates an output that is far larger than the individual contributions. And so for me, that's the thing that's the most exciting is that, you know, over the next two, three, four, five years, we're going to see a total transformation in the way that products are built and how users consume them. Yeah, you know, I think it's interesting. One of the things that I've shared occasionally on Twitter and on this podcast, which I don't know if it's a contrary opinion or not, I guess it depends.

on who you ask is that I actually think that the state of new user adoption is in an awesome place as it relates to defy, right, in the sense that we're not out actively recruiting tons of new people into the space, but rather building these things which were all dogfooding and that are early adopter driven from people who have a proper sense of the risk involved and the technical complexity involved, to, as you pointed out, sequentially work kind of one by one through these issues. To me, that creates the opportunity that we actually make this space safe kind of on our own and without the stick, right, of regulatory threats or anything like that,

before you start to see more and more people come in. And I think that that's actually really, really healthy. So I tend to agree with a lot of the points that you're just making. Yeah, I agree. I think that that is something that makes us a little bit safer. But, you know, if you're around, which I know I know you were, and I know a lot of people listening were for 2017, that can work until it doesn't. And we could see a flood of new users come in maybe before we're 100% ready. So that is something that I think, you know, there is an onus on all of us building right now to really try to kind of put as much structure around the stuff as we possibly can. But, you know, it's also exciting to see,

you know, how much is being built. Absolutely. All right, Sergey, you're up. One thing that makes you excited and one thing that makes you nervous, I guess, in whatever order you'd like to tackle them. Yeah, so I think the thing that makes me a little bit nervous or the thing that I'm that I'm very eager to see kind of is that there's on-chain privacy. So I've been, you know, looking at on-chain privacy, whether it's through homomorphic encryption or zero knowledge proofs or any number of combined approaches or even trusted execution environments, where in the real world, in like the traditional world, contract privacy for even though all these really complex contracts is extremely important. There's certain contracts that they have to stay private by law because

they could affect some kind of trading volume or something in, you know, that's very common in global shipping. Insurance contracts need to stay private. A whole bunch of financial product contract needs to stay private. So I think that the things that people are working on right now are kind of scalability is a big topic because you have people in gaming

in certain use cases that are hitting the real limits of existing systems. So scalability is a big topic. We're very focused on the Oracle problem because it expands what smart contracts can do while retaining their security, which kind of goes to redefine the space towards all these use cases like essentialized finance, insurance, shipping, all these use cases. But I think the thing that I'm also very worried about and the thing that I think also needs to fall into place

is the ability for contract terms and contract outcomes to remain private. Because that is just a fundamental requirement that many of the digital agreements, that we're trying to replace half. Now, I've seen a number of things happen there, and I've seen some progress with zero knowledge proofs, and I've seen some interesting things with trusted execution environments, and we actually have some work that we've done where you can do some computation in an Oracle,

and that can help create some privacy, and, you know, these types of things. But I think, generally speaking, on-chain privacy is a big, big piece to this puzzle for true large-scale mainstream adoption, which I think is going to happen, whether it's through zero knowledge or homomorphic encryption or any collection of other approaches. Now, the thing that I am very excited about is kind of echoing what Kane and you have been saying is that I'm, you know, I've been building smart contracts in this space for many, many, many years. Kane has been building them for many years.

A number of people have been building smart contracts that aren't. about token generation. They're about some kind of financial product or an insurance product or shipping product. They're about a contract that doesn't just generate a token and track its ownership. They're about something that reacts to real world conditions the way that 90% of the digital agreements that we're seeking to replace as an industry do. And I think the really exciting thing is I'm like Kane has, starting to see the infrastructure that's necessary for people to quickly build and iterate, similarly to how web developers are used to quickly building and iterating on products. I'm starting to see high-quality teams be able to combine smart

contract logic with highly reliable data from something like an Oracle network together with a connection to some kind of payments output that their user would want to receive, whether that's an on-chain token or whether that's a connection to some kind of payment system, once again, through an Oracle. I think I'm starting to see more and more people, more and more quickly build higher and higher quality decentralized applications at a speed that I haven't seen ever before. And then if that happens, if that happens, the only real question is, is there enough demand to make them successful, to make the ones that build a good application successful? And I think we're just passing that threshold now.

So I think I'm really excited about these two dynamics really coming together. The high-quality teams that are building truly decentralized applications because they have the infrastructure like Ethereum or like ChainLink or whatever other collection of infrastructure that they compose to use there to use to make their application. the other side of it is that there being enough high quality, there being enough private key holders to make them successful in the building of that application. So I think those two dynamics coming together show me a world where not only did somebody build a great defy application, but they also achieved a level of success that justifies other people building more defy applications, which is how I think we go to a world where we're building smart contracts that are about all these useful types of interactions with the real world

the way that digital agreements are today. And I think that's when we start to really, as an industry, take hold of a relevant amount of ownership of basically contract activity out there in the real world. Love it. All right, guys. Cain, Sergey, thank you so much for your time today. Like I said, I think we could talk about this for hours, but I really appreciate you guys joining. And I'm excited to see what you guys build both independently and through continued collaborations. Thank you for having us on the show. I really appreciate it. Yeah, it was good fun. Thanks a lot. So there you have it. Regular listeners of the breakdown will know that my interest

tend to be on the highest macro levels of just about anything we talk about. So I was excited in that context to hear about what both Kane and Sergey think around this concept of where decentralization is today and what makes them nervous and optimistic about the future of decentralization and defy and crypto more broadly. But I also think that when it comes to defy, a lot of the really important stuff to be paying attention to is in fact in the details. The word decentralization is right there in the name, right? But it doesn't matter if some big part of the system like the price feeds is centralized. So I really appreciated for that reason being able to go a little bit deeper into the weeds with

Kane and Sergei. These projects represent some of the most on the edge leading indicators of where Defi might go in the future, as well as some of the core. infrastructure that is being built to enable those and other types of futures. So I hope that you enjoyed this conversation. Let me know how you like the episode. Let me know how you like this multi-guest format. Hit me up at NLW on Twitter, nlw.substack.com for the newsletter where you can get the breakdown every single day to your inbox or subscribe on iTunes or Spotify or wherever you listen. And as always, guys, thank you for listening. And hope you have an awesome weekend. I will chat with you again on Monday. Peace.