Turbocharging web3 with incentivised and decentralised RPC

Conor Svensson: So I'd love it Yair if you start by introducing yourself and telling us a bit more about how the Lava Network came to be.

Yair Cleper: Yeah, for sure. So can you hear me? Fine.

Conor Svensson: Yeah, audio is great.

Yair Cleper: Okay, great. So super excited to be here, you know, taking me back and telling the story of Lava Network. But just before diving right in, maybe a bit background about myself. I studied computer engineering, born and raised in Israel and started my first startup when I was 21 building a couple of very interesting projects. My first startup in 2009 was something similar to Wix, the website building. We ended up selling it after three years and then I founded something similar to Etoro, the trading platform, the social trading platform. Yeah, I jumped into different verticals from business intelligence, augmented reality and the last start was in supermarkets, supermarket technology. Three and a half years ago my co-founder Gill, who has a very strong techy background in cybersecurity, dragged me into web three. He started by developing a couple of bots 3 1/2 years ago. I believe it was the NFT summer and yeah, we jumped right into the rabbit hole. It was a super exciting time, all time highs and we were really interested about a few problems. The main problem was that infrastructure was shit. You know back when you were thinking about a handful of chains, starting with Ethereum and then Solana and more chains but we believed in a multi chain future and obviously we believe that infrastructure will serve as a very critical layer, data access layer for blockchains if they want to enable developers to interact with the chains themselves. And this is what brought us to think about Lava.

Conor Svensson: So it's interesting you had this kind of multi blockchain thesis a few years ago because obviously in the last year or so we've really seen everything go crazy with respect to first you had the layer 2s and then it's the app chains or L3 networks and there's just so many networks spinning up. And whilst there was certainly a few years back that narrative about there being more than one dominant blockchain network, what was it that led you to think there was going to be an absolute explosion in them versus just there being five or ten big ones or even a top three.

Yair Cleper: That's a great question. you know we didn't try to make any assumption for the future because we were newcomers to this industry right? So we are more trying to learn and listen. There were a lot of voices at one time. We were talking about consolidation of the different chains. Personally I believe that the chains, the ecosystem, the tokens, they represent different agendas, different communities and as you see in the history of humankind you don't see only one state, for the entire globe. Right? Everyone is a different characteristic, every community brings its own values, its own principles. And I think this is the proliferation of chains that we see today. It's the same way that Web 3 is going to develop into, right? You think about, read, write, own, how the Internet started to develop and in the .net boom in the 2000 you see so many websites kind of doing the same. So I think we're living in this era that it's not only a financial revolution but also a new Internet, enabling tools for different communities to express their ideas, their values and yes and that's what the future holds for Web3.

Conor Svensson: Decentralised Infrastructure has certainly been an area which has had I think an interesting growth because at one level you have some of the slightly more speculative deep in type projects where there's a view that this technology should potentially work but then ultimately they're trying to take on many of the large cloud providers and there can be challenges there just because of the economies of scale and these very large incumbents. But arguably the idea of our decentralised RPC's has been an area which has actually grown quite quickly and has managed to prove itself in terms of being aligned really with the ethos of Web3 in terms of being able to achieve decentralisation but also being able to come in at a price that's compelling for whether it's end users, businesses or protocols to actually work with this. So it would be great to unpack a bit more about you, why RPC works in this regard and the kind of incentive models that need to be used for it with Lava Network.

Yair Cleper: Yeah, exactly. I think you put it in exactly the right way. The bottom line is that I believe centralised solutions cannot keep up with web 3 and now AI. And we will see more and more decentralised protocols like Lava that are kind of incentivizing a scalable infrastructure with demand but without compromising on the reliability itself. So you know when today you're surfing the web you're typing www.google right? But you have no idea what is TCP, IP, HTTP, what version is that at the moment and if you're using a GCP AWS where the website is located. But unfortunately in web 3 because of the infrastructure, because of the architecture you need to know that DApps have to know that and obviously chains have to know that in the same way because everyone represents a different Internet. Right? So when we see the hundreds and thousands of different chains I think this creates the way that the traditional infrastructure provider works today. It creates a bottleneck for such a scale. And Lava is kind of solving that by creating a decentralised marketplace because it brings together data providers that ensuring fast scalable and permissionless access and can enable the DApps and the chains not to worry about the infrastructure, not to worry that when their user is coming to an opening, you know the wallet, they can see an up to date token price when they want to make a transaction. They're not going to get an error when they're minting a token, when they are claiming an airdrop. All of these have to make sure that the infrastructure is flawless. And we use decentralisation not only for the sake of decentralisation but we believe this is the right way to bring together infrastructure that simply works and can scale the industry.

Conor Svensson: And I'd love to dig a bit deeper when you talk about the infrastructure just working because it is something that genuinely can do that. But certainly, I'm sure, there's some people listening to this who are thinking, " how can it work” right? If you've got a large organisation who's got access to a tonne of servers surely they can spin up their own centralised farm of RPC's and undercut some of these other offers. But there's incentives at play for people to actually run these decentralised RPC providers. So I'd love to just get a bit more into the weeds about how that side works specifically on Lava.

Yair Cleper: So just to put the pin in what you just mentioned, the devs themselves, the developers, they are super skillful and they can develop this infrastructure, this redundancy but the question is do they need to? Right? It's simply like you asking the website owners to implement a version of the cloud or their home, right? A private version. It doesn't make sense. And this is why what Lava is doing is kind of rewarding the infrastructure provider based on the quality of service. Right? So imagine that you already run a node because you're a technical user or because you want to make sure you are contributing to the specific ecosystem. So anyway you're running a node and the Lava marketplace is inviting you permissionlessly to join and serve because you already can monetize your existing infrastructure. Same way if you think about Uber for nodes that's exactly what you do. At the end, you monetize your own vehicle, your own car to drive passengers from a place to each other. But according to the speed, uptime, accuracy, all of these parameters you get the score you're receiving as a service next time. So we believe and we already see that happening in more than 40 chains, we see that this kind of competitive marketplace is actually encouraging even smaller providers to thrive. I know that if I tell you the specific names of reputable providers, you're probably going to know them. But what we realised in the last year that sometimes you can receive a better service for the chain, for the DApps, from providers that don't know how to give you quotes but they know the technical standards and parameters behind the scenes. It's getting more and more important, especially when you see use cases in AI, right you have more niche chains and emerging AI that we see more and more use cases there. So the quality of the data is important and if we are able to answer and address this need of quality, of service metrics, I believe that this is going to make the infrastructure much more resilient and nobody needs to worry about infrastructure in the coming future.

Conor Svensson: And the quality of service point you made there is a very key one. How is it that you can evaluate quality of service? If we talk about the analogy, say with someone making use of a ride sharing service or maybe they are hiring out a car or something that they don't use. Then of course with these consumer facing type applications there's a whole process for people to review them and rate them and so on. But when you're talking about decentralised infrastructure you tend not to. It's not like someone's going to do an issue as they request over the API and rate it then and there based on how fast it was. But what are some of the mechanisms that you use there to achieve that?

Yair Cleper: Definitely. So the same way you finished a ride on Uber you need to, to report back if you want. Obviously how was the driver? Was it nice? Was the car clean? How was the driving? Was it safe? And different parameters. So the Lava network and the protocol itself is focusing on different parameters such as speed, uptime, reliability of the data, the latency and when this session between the DApp that’s consuming the data and the data provider is being made all those parameters are being aggregated. At the end of the session, automatically the DApp signing the transaction is signing the fact that all of this information is collected and is signing the transaction and this is being used for the data provider to come back on chain and claim the rewards for what he was serving. This is how the Lava protocol incentivizes honest behaviour and makes sure that all the important parameters are being used to score if you should receive service for next time. Behind the scenes there are other algorithms working like benchmarking the data for example, the latest block. If you query specific information like what is the gas price, and you know one of the providers is providing you with not up to date information, this can be used because of benchmarking to other providers to reduce the score or even slash. And the mechanism here lies at the bottom of the fact that providers have to stake Lava in order to participate and provide service on the chain.

Conor Svensson: So it's kind of a combination of staking incentives for the actual providers on the network but then also in effect sampling type metrics as well just to check the individual providers too that they're kind of performing as they should be, correct? I think that certainly, even if we take the example of Chainlens, our Explorer technology, we work with RPCs. They are the endpoints that we always use and so we typically find that there's providers or networks that provide these cheap RPC type services and I think it's quite astounding how affordable they can be. But then on the other side I think there's a lot of technical teams who are running RPC infrastructure whether it's your development teams or whether it's companies like ourselves as well. And it seems like there's a very big opportunity there to actually start embracing these types of decentralised RPC services both as a provider as well as potentially consumer.

Yair Cleper: Exactly. Because you know at the end of the day it doesn't matter what you have on the ROP on the package but it much more matters that you're going to have an uptime of 100% right? You're going to get the fastest response. And we started with the kind of decentralised RPC by going to different and serving dozens of DApps, from wallets to exchanges and so forth. But what happened exactly one year ago is that one of the chains changed its approach and said actually Lava is a public good. The chain had the same problem because they needed to provide a budget for those incentives and they ended up contracting different providers that have lock-in contracts with them and they don't know what they even get. So now they started to shift the budget. Instead of working with 1, 2, 3 providers they put these budgets on what are presented as incentive pools and those incentive pools are open for the public. It's open for the community to join stake Lava as an accountability tool and immediately receive service based on the performance. So I just had a very interesting conversation with one of the CMOs of the chains we work with and he said this budget, we're not looking at that as an infra budget anymore. This is actually a community marketing budget goes back to the community.

Conor Svensson: That's an absolutely fascinating point that you make there because there are, even now within web3, whilst at its heart many of the people building are focused on the decentralised decentralisation angle, there are these very real walled gardens where you have these incumbents that have managed to establish themselves, get significant kind of positions within the industry and when new networks launch they get a lot of traction. Then all of a sudden maybe they want to onboard this protocol or service and as you say there's the sort backroom deals that happen, which is no different from traditional kind of software in that regard, and having optionality for other approaches which can be more community driven is something that really we'd hope to see more of. And as you say there when you can have like a large layer two protocol that comes along and says well instead of paying these incumbents so to speak, we can actually ensure that the community helps with this side by providing the right incentives there, which is clearly win, win for everyone, 100%.

Yair Cleper: And I think you know that at the end of the day, you think about Filecon, you know, focus on decentralised storage, right? It's kind of a complementary to Lava’s mission to provide decentralised data access. And I think this kind of collaboration between decentralised storage and data access, this is ensuring that Web3 AI agents and the foundation is both secure and scalable.

Conor Svensson: Absolutely. So you've got some really impressive metrics as well. I mean this is from a post on X from just over a month ago saying you've got over 2.3 million monthly users across over 40 blockchains and over 80 billion RPC requests.

Yair Cleper: That's correct. I actually, I checked for one partner that we were working with and I checked last week, in the last seven days we reached almost 10 million unique users. And what's crazy about that is that they probably never heard about Lava and it's serving as the backbone to ensure the uptime of the service, to make sure that they can have reliability and accuracy of the data that they receive. And RPC is only the first stop, right? Think about all the different data primitives that are out there. We work together with N Axelr, Cosmos, Starnet, Filecoin and we see their decentralised infrastructure needs. On the other hand, you can think about Lava as a marketplace. So you start with RPC because this is the building block but afterwards, what about sub graphs, AI agents and many more data types that the DApps or the chain themselves they need to provide. I think this is only the first op RPC.

Conor Svensson: And it's definitely interesting to understand how this can evolve because RPC data is very standardised and there's a lot of processing of RPC data. From our own perspectives that you know, we have to be able to then generate subsequent data, such as token information for Ethereum compatible networks. You mentioned sub graphs and presume things like data availability types could come into this as well? But what are kind of the near term plans in terms of what other data you want to bring on chain?

Yair Cleper: Yeah, so I definitely think AI is a big narrative right now and I think that if you can create a marketplace of AI agents that is integrated to Web3 and it's ensured the same time, transparency, trust, resilience, we can see a lot of feasible apps that are developed in the next course of time. If you think about more interesting manipulation of the data, you know, sophisticated APIs, like checking what is the NFT or the meme coins that the specific wallets hold and then do some computation of the data itself, it’s just one of the million different use cases that you can do with the data. But I believe that no one today has answered, okay, this is a data access layer that we can really trust. I don't know many very successful DApps that are focused on a single chain. So today we see it multi-chain and the DTS themselves and they're moving between the chains when they're transitioning they need to do the same work on and on again. Right? Who are the providers providing RPC in this new chain, who I can trust. Then they're implementing you know their redundancy layer because they cannot really trust them. But if today, if tomorrow, and already today they can have a single line of code with Lava getting access to all these different blockchain. I think this is a game changer to the industry.

Conor Svensson: I guess I have to ask off the back of that though is how do you ensure the trust of the data that's being produced by the network? I’d be interested to get your thoughts on that.

Yair Cleper: So basically you know there is a predeterministic algorithm that checks and compares as you said before, the reliability of the data, without the providers themselves to know. So providers first of all don't know in advance who they're going to be giving service to. Right? This is how it really reduces the risk of manipulating the data. We see kind of DNS hijacking and so forth and the providers they are available to receive service and the chain pairs them to the DApps to provide this service but it doesn't ensure in advance who they're going to provide service to. At the same time every couple of requests is being benchmarked to a different provider and behind the scenes they are mechanism of conflict detection and then conflict resolution that implement that. This is the bread and butter of the Lava network and they kind of ensure data honesty.

Conor Svensson: I guess again it kind of reiterates the value of having a decentralised approach there because ultimately there's not really an incentive there for people to be bad actors on the network because all it does is they're going to lose their stake as a result of it and they don't know even who they would affect because there's no sort of determinist in terms of which provider service which requests.

Yair Cleper: You know it's kind of a validator approach right. They have to be up and running and they cannot not participate in securing the blockchain otherwise they're risking their their stake. So the same metric was not implemented until now in blockchain when regarding the data. So we want to change the paradigm and kind of bring back the trust to the access layer.

Conor Svensson: So we've got a minute left and just to kind of leave with some closing thoughts for teams that are running their own RPC infrastructure. What would be your ask to them?

Yair Cleper: My ask to them is to look at the board, the explorer, that Lava has invited them to the community to see such an amazing community and super super techy one answering different questions. At the end of the day all the feedback we receive is that it simply works and I invite DApps builders to try the gateway on different ecosystems too, to see how simple it is to implement the endpoints of Lava and also join as a provider. If you're running a node, even a simple RPC node you can start monetizing that and you see an RPC provider already receiving more than $300,000 in the last few months from the chain. The Lava foundation reported almost $2 million that the chain actually gave back to the community. So it's really early. Mainnet was just launched four months ago. You know I'm super excited to see this kind of bull run and how it's ending up being for Lava's awesome.

Conor Svensson: Yeah, thank you so much for your time. It's been really great to hear more about Lava Network and we're definitely going to be doing some digging, even ourselves into it, because we're already running some AR/PC infrastructure, so it seems like a no brainer to check it out in more detail.

Yair Cleper: Thank you so much Conor. I appreciate being here.

Conor Svensson: Thank you. Bye.