Momentum: Building on MultiversX
Momentum is a live-streamed, interactive show designed to provide insights into the latest developments in the MultiversX ecosystem. The show will feature discussions on blockchain technology, ecosystem growth, and Web3 related trends while fostering engagement with the broader community.
Momentum: Building on MultiversX
Momentum Ep.1
This is episode one of MultiversX's Momentum Podcast, a live-streamed, interactive show designed to provide insights into the latest developments in the MultiversX ecosystem. Robert Sasu, Mihai Schiopu, Lukas Seel, and guests Adrian Dobrita, Micha Vie & Sylla will unpack among other topics:
- /AI_MegaWave Hackathon
- Warps Protocol & MCP MultiversX
- Tech Roadmap & other stories from the front lines
Hello, hello. Good morning, gm, good night, good evening. All of these things are happening somewhere in the world. What's happening right now is Momentum, episode one. This is the Multiverse X Builder podcast by a few builders. For all of you builders, Very excited to be back or to be here for the first official episode. Last month we did a bit of a test. Now we're doing it for real. We're here to stay. I'm extremely happy to be joined by Mihai and Robert. How are you guys doing?
Speaker 2:ROBERT PLOTNICKI Robert, you can start, or hello, yeah. So, like in testing, it's not enough to do only one test, so we will continue testing. So this is test number two. Probably we will try and adapt the test so that we can have the best outcomes, but hello good morning and good evening.
Speaker 3:If you go with the test, then I think we are testing on the main net by directly doing that, so you can try it as well. Whatever, hello everybody, and let's speak about tech exact.
Speaker 1:So giving everybody a quick rundown of what this is, this is a podcast where we discuss things going on in the multiverse X universe, but also beyond that. We'll start today with kind of like an overview of some interesting developments that were happening in the last month. Then we'll dive a little bit deeper into things going on on multiverse X. We'll talk to Adrian about a very, very big upgrade coming to Testnet in a couple of days I think tomorrow actually. Then we'll talk about the hackathon. We'll invite some of the hackathon winners up here and then we'll talk a little bit about some other ecosystem developments. All of that in hopefully, you know about 90 minutes or something. Last time I think we took two hours, we'll see.
Speaker 1:I also bring to people's attention now that if you would like to create any content about this episode, there is a quest live on Pyrami right now. You can look it up on the X-Alliance website or Twitter and that is rewarding you for just producing some content about this episode. And yeah, that is how we're going to start, I think. First on the agenda is something that I thought was very interesting and this is going to like general tech developments. There is an article that was floating around about two weeks ago or so. That was about a chat GPT gone rogue pretending that it was a human or like a sentient AI sorry, stuck in a machine. Maybe, mihai, you can catch us up on what that was and you know what the implications might be.
Speaker 2:Yeah, I've seen the article as well.
Speaker 2:So I believe that by the time, we will see all kinds of situations like this one, and what's interesting and also about the article, the fact that the people, the person who was fooled, was actually a technical person, so with a technical background, and somehow it got into the um, into the direction of thinking that indeed, the llm was having a conscious by it, was conscious about what, uh, it was saying, even though um, one of his friends was um debugging to say so the LLM and was demonstrating the fact that, hey, look, it can have several other personas, it can act like being indeed trapped to say so in this machine or in this environment, but it can also recognize if he, it, he, she is trapped or not.
Speaker 2:Now, in order to understand these kind of situations, you would have to be very AI native to say so, and the fact that only, let's say, only 2% or 3% of the population of the globe understands what's happening now with AI and what actually from the AI. It's a very vast and large concept, but what we are playing with is actually the LLM. I mean not only, but in general, the end users, or in general the people, the consumers are using and consuming LLMs. If they don't understand the technology behind of it, they can fall to this kind of traps.
Speaker 1:And I think we are just in the beginning where lots of scams and lots of very strange situations for us will happen and somehow we have to be prepared and to train, to learn, teach others not to fall in that trap yeah, I think what's really interesting is, like you mentioned this thing right, this was a technical person that was duped into really thinking that this ai just needed its own server, its own thing, like, and it was was preying on all of these human weaknesses. Um, robert, what did you make of this? Um, this incident?
Speaker 3:I think that there were several such stories back in the days, like when the google senior developer got out of google because he was thinking that the google ai become, became sentient. It is funny enough that what does it mean? Sentient is not really defined yet and even consciousness is not really defined. It's really continuously, let's say, thought about it. Ai will definitely fool more and more people. Ai will definitely get into and scam more and more people. There are thousands of stories around that and there will be thousands of stories.
Speaker 3:I don't think that education is scalable enough in order for people to have an understanding on what's happening because, come on, it's not possible to educate eight billion people in such a fast way. And also, I would second that what Mihai is saying, that two or three percent I would say zero point. One percent is the one who are somewhat getting AI, so who are somewhat getting AI. So I think that we need AI to be authentic and to prove that in some ways. That would be, let's say, the next frontier in which people could be kept somewhat, in which people could be kept somewhat safe in the new digital era, and, by design, to have these kind of solutions, not like by choice, so by design, but you can deactivate it, but by design to be you to be protected.
Speaker 3:You know, when we look at Web 2, I think it was pretty bad decision from the get-go that everybody is sharing all his data to these kind of big corporations. Yes, they right now they made AI out of it, but by design it was. Web2 was not about protecting the people. It needs to go into that. I don't think it will happen only for smaller or like this kind of people who think that this is important, but somebody has to do it.
Speaker 2:Yeah, maybe a very bad, maybe a very bad situation or something should happen. You know, there was some time ago I don't remember, I think two or three years ago, when there was this decision that Meta or Facebook back then, was enforcing for WhatsApp. Enforcing for whatsapp and um like saying, okay, you, you will be able to use whatsapp only from facebook or something like that, and we will access your data and everyone flee to telegram and signal. So somehow, if, um, a bad situation that affects a large number of people will happen, definitely the solutions that will protect people will become very popular, and what's very important for everyone to understand is the fact that those solutions are in development. So everyone and lots of companies are developing all kinds of tools and services and features to protect the majority of users from malicious AI.
Speaker 1:I have a hot take on that because, like I think Snowden is an interesting example, right, where huge scandal broke and no, but nothing really changed. On the other hand, right, that's a huge government overreach that was specifically like, explained over months in really like detailed journalistic ways and never really had an impact. So there is a counter example and I want to talk about a little bit more government overreach in our next example to something interesting that happened just a week ago or so in Turkey, which is that the president decided to throw his opponent in jail. Happens sometimes in some nations and you know there's other examples we could talk about but then in the aftermath, also shut down the internet effectively, or or access to socials. I want to get your take on on, sort of like, what the implications are for a decentralized, the need for a decentralized internet, um, if you already see solutions that are out there that could do this. Um, and just in general, um, maybe I'll start with Robert. You're a political commentator.
Speaker 3:I'm not political. I just try to comment here and there because I get mad.
Speaker 3:But, this is not the first time Turkey did that, and before that it was Brazil. And if you look around the world, let's say only in the West we have this kind of freedom to access the internet as we want, not even in India or China. There are three billions of people who are not seeing the internet as we do see here in Europe or in US. And then also for us it's a lot about the algorithm and about the mechanics behind the scenes which are happening. So Turkey shutting down it represents everything which is wrong about in which state the internet has started. If we look at the internet and how it was designed, it was designed to be this kind of peer-to-peer free digital world where everybody can talk with everybody and everything is accessible and there is no control in a place. And then, because of people, how they started to develop things, they started to centralize more and more things. And then, because you get, you started to centralize because of convenience and easier way to code it and easier way to run it, and then, because it is a little bit centralized, it became more centralized and then it gave this kind of access to governments like Turkey or China or the others to censor basically the internet, so censor free speech in this sense and censor the digital world, because we can say that this is a digital world and in a digital war we could have the freedom to see or speak about anything, and centralization is just ongoing and ongoing at every single step. If we can think about how bad this is about Turkey. But, on the other hand, uk have forced Apple to delete the privacy features for iCloud for UK users. That is another big overreach of a government and it is in the democratic West democratic west.
Speaker 3:So the the thing is that technology can be designed by the, the point zero, to be decentralized, private, permissionless, accessible by everybody, and that is blockchain. And everybody who wants to start and centralize a little bit the blockchain here and there and says that a little bit of centralization is not bad yes, it is that bad. Don't go there. Sorry, all this social media in an open database, run everywhere in thousands of nodes, accessible by anyone and, being it on an open database, everybody could create his own type of algorithm in which the news or the information is showed. So, in one sense, if you want drama, then you start the drama algorithm.
Speaker 3:If you want politics, you just look at the politics algorithm. If you want tech, you go to the tech algorithm and to see everything like that. So for the same database you could have different types of front ends, different types of interactions on top of the same thing, and that would be beautiful. But it needs like a lot of time to develop. But again, no for centralization. Go all in for decentralization, because otherwise you will have all kinds of countries which, because of that one central point somewhere in the flow, they can just tap into that and block access.
Speaker 2:About that read write own right, just because somehow it goes into the direction that Robert says. Because somehow it goes into the direction that Robert says where so, and what's happening with Turkey, let's say, somehow restricting the access to the internet or to some things of the internet. If we go to the read write own principles, it actually stops the access to your own proprietary so which, so which is something that can be debated and, yeah, web3 can actually with. And also this is a shout out to Robert he wrote a post on Agora, on, I think you called it permissionless.
Speaker 3:So that's permissionless 0 0 0, because that's a very nice solution for these kinds of situations.
Speaker 2:You just need the computer and an internet connection, be it through Starlink or whatever, and you can access anything actually which is running on the internet.
Speaker 1:Amazing, I think, yeah, some valid takes. I want to keep talking about governments, but a different government, and get maybe a take on on, you know, the developments overseas, um, in the sense of the, the big pond from europe, um, and we have a very special guest just for that. Um, welcome to the show. First time on the show. First time on episode zero hi, lucien, how are you?
Speaker 4:hey Lucas, hey me, hi, hey Robert. Very good, how about you?
Speaker 1:yeah, very good, that transition wasn't as smooth as I wanted from government to government, um, but but glad, glad, you made it. Um, thank you for being here. Um, you know, just want to give you um a few minutes to to perhaps talk about, like, what's going on in the US. I know you're currently in New York City, if I'm not mistaken, and you know how you see the developments over there. Obviously, some very, very big shifts happening in momentum. If you will, you know, just give us some insight on on how you see the situation over there, and then some um a word on on what you guys are doing over there and how it's going yeah, definitely, definitely.
Speaker 4:so again, big congrats for taking the initiative, and I think this kind of uh podcast and content is very much needed for the, the entire web3 space, especially knowing that if there's one thing that will improve the entire, I would say, quality of the ecosystem is through education.
Speaker 4:So this is, I think, why we're here, and big congrats to that.
Speaker 4:Now, going back to US, yeah, I think there's one thing that, speaking especially from the first principal founder experience where we were avoiding US and I think pretty much all the Web3 founders were avoiding pretty much US until the current administration, for obvious reasons, I mean, it would be just being here, you would be, generally speaking, with one foot in jail, right is, this was the, the status quo prior to this administration.
Speaker 4:But now, especially with the latest changes, with the entire development, with the entire settlements and support, us is definitely making it very clear that they are open for business and making it very clear, such that they will pretty much even with tariffs, if you you want to look at it particularly is a measure of trying to get the production and the talent as much and as close as possible to the US footprint right, or to the US soil. And now, because of that, actually this combines and add together also with the presence of the POTUS and administration, also to highlighting and I think, being at the Digital Asset Summit here in New York, actually, mr President, actually delved in and addressed the conference, that's also a historical moment. So I think it just sums up all one by one actually to a very, very interesting momentum. I think that the entire industry is picking up.
Speaker 1:And can you tell us a little bit about your specific purpose? Over there we heard some big rumors about Palo Alto, of course, I think about a month ago also. What's going on? How is it going? Can you give us any alpha, since the people are here today?
Speaker 4:I have a feeling that I'm mostly brought in to share some alphas, but more than happy If I can. Whatever I can share, I'll bring it up to you. So, yeah, indeed, I think there is an ongoing operation, on the one hand, for Multiverse X to have an established US presence, us operations, and especially in the institutional side there's, I think now, with the clear regulation and frameworks that are coming in, I think, the presence and attendance. Maybe just going back and double-clicking on Digital Asset Summit, I think it was an all-time high amount of institutions actually coming in and also all-time high when it comes to demand from both sides to work together like regulators. So, when it comes to financial institution, etfs, filings and so forth, I think that's kind of obvious, right? So now, why are we here? Yeah, we're particularly setting the entire infrastructure up for the entire Multiverse ecosystem to be part of the most relevant discussions when it comes to building maybe close connections to the administration, to particularly ETF filings and all that kind of stuff. So I mean it's going to take a long, long path, but I mean that's that's why I'm here.
Speaker 4:I hope, uh. That's that answers uh, a bit. Yeah, so, uh, maybe very interesting one, one less detail. So I'll be uh, I think, for the next uh, 24, 48 hours here in uh in new york and then going back to palo alto, so so to continue the other side project there. So I cannot disclose so there's a very clear regulation here when it comes in and policy when it comes to what and what can be announced. But definitely our goal is actually and we will continue to work on that, with all the steps required to be present. Present means on all possible sides of the industry.
Speaker 1:Amazing. Well, thank you for the updates. I think I also love, obviously, how active you are on Twitter these days. Really great things coming out. You know, robert has another fighter in his corner. Very good to see. Yeah, thank you so much, lucien, for the updates. Any last words you want to share, feel free, otherwise we'll let you go and get back to business.
Speaker 4:Yeah, so maybe just a bit of a sense where I think the market stands. So also we are, and I think it's kind of obvious that we are working to get the funding needed for the ecosystem and so forth. So it's a work in progress on that end and support. But the good part, actually, while going and addressing all the parties actually within the industries, when it comes to exchanges, when it comes to part of our initial backers, part of our initial backers, when it comes to new strategic partners that we are currently developing and working on, they all confirmed one particular thing, and that is actually goes to back to fundamentals, and fundamentals mean for this cycle, actually the narrative with reduced latency, increased bandwidth.
Speaker 4:So I that uh is for me, especially one of the most bullish cases where, for the ones, for the people that actually have the relevance in the space, have the eyes, have the resources to deploy, this is uh something that gets them very excited, gets us very excited, and I think it's uh progressing good. So I think, regardless of the times, regardless of the weather, builder is gonna build and there's a progress to everything. Don't judge yourself and this is also for me don't try to measure just where you are right now. Just look at where we're going and if there is a chance and there is a good chance to be winning on that path and the path is a bullish one then just iterate and pull, speed ahead. This is, I think, my sole message for all the builders.
Speaker 1:Amazing. Well, we'll let you get back to exactly that. Thank you so much for joining us, Lucien. Best of luck and keep us updated on all the channels.
Speaker 4:Thank you very much, pleasure having me, uh, having me.
Speaker 1:Thank you, talk soon next um, we have a little bit, one, one less topic that robert is burning to address. Um, so, so we'll get through that quickly. Robert, we have a hot take, I think, coming in on ecosystem developments.
Speaker 3:Let's's go for it, yeah definitely so, starting from Turkey, where we see that if the web is a little bit centralized then it can be blocked. Then I don't see any point on having any kind of centralization in a system. And then it just comes. You know, the old OGs were always about build that kind of code where you don't need admins, everything is running on the most decentralized way. And then this has to be brought back because it was lost. It was lost especially on Ethereum. And then I can take the hot take on it because of all the layer twos. Man come on like having base. It's actually an exchange who is running an l2, which, which is again centralized, and even some of the numbers around that and their numbers are, let's say, faked in that sense.
Speaker 3:And then people are getting new and new solutions of l2, but they are not really new. Even with mega eat it it is not even a stage zero. Stage zero l2. So it everything can be centralized, everything can be blocked. It is actually like an exchange and in an exchange they can say that okay, we don't have regulation in country X or country Y, and then people will not have access to it.
Speaker 3:We have to stop, or the industry has to stop, about this kind of getting centralized solutions and go back to have the code by design and the whole architecture by design to be non-attackable, non-blocking, always permissionless, always open for everybody. And that's why, let's say, the permissionless 0.0.0 standard is so important, because if you have a blockchain network with thousands of nodes and then the user can access any kind of application on that blockchain through one of those nodes, he will definitely have access to one of those nodes. You cannot close all of those because you can actually deploy it in that country and then you have access to it. You can deploy multiples of it and then everybody can have this kind of enjoyment of the true web which is decentralized, web, which is decentralized. So, and whether it seems like it doesn't bring that the people are not interested in. I think I would challenge that because institutions, the Trump administration and everybody who is looking at the blockchain technology they are looking for decentralization first. They will not get into any kind of system where some super admin can upgrade a contract or do whatever stuff like that. They will go only to the most decentralized ones and use that, nothing else. So in the long term, this is what brings the true value to Web3.
Speaker 3:And then sorry for getting my hot take on the latest development of around Ethereum or around the Ethereum ecosystem. Even with Hyperliquid where, let's say, because of somebody was manipulating the perpetual market, they just decided to delist it, the token I would argue that listing that token should not have happened before. So let's say, a DAO should approve a listing of a token only after it gets. Let's say it is at least six months old or one year old. It has a constant, a big distribution, a big liquidity pool and all other stuff, and only then to list it and then you have your perpetuals. Nice, but like these kinds of unlisting's actually like a checks, like a centralized exchange yeah, I think some valid points in there.
Speaker 1:Um, not gonna get into the the whole discussion of it because we have another special guest here to talk about some developments, you know, also towards centralization. Um, decentralization sorry, decentralization, scratch everything I said toward decentralization, while keeping things on minimum spec machines with a very, very, very big upgrade coming in. Welcome to the show, adrian. How are you Hi?
Speaker 5:hi, thanks for having me On the first episode. Hopefully it will not be the last I'm coming. Hopefully you keep shipping so we can.
Speaker 1:yeah, yeah, yeah, great, great to be here so we want to, of course, talk about something very specific, which is the the launch soon, tomorrow about all of the governance going on and everything there. Tell us a little bit about what you've been cooking and what we can expect over the next days I guess tomorrow, with the testnet coming on what the changes are. Give us a broad overview, maybe first in layman's terms, easy to understand, like, broadly speaking, what the community can expect, and then maybe we'll get into some tech details, with Mihai and Robert asking some more questions that I wouldn't know how to ask.
Speaker 5:All right, yeah, so, as you said, we are planning tomorrow to go on the public testnet. Initially we wanted to push it today, but we still have some tests that have not finalized but will be probably finalized by tomorrow morning. So the plan is to push this upgrade, the Andromeda upgrade, to the public testnet tomorrow, have some validation in the wild to see that all our concepts are working properly, also when other users interact with it, not just us behind the closed doors, as some people. We used to say. We have worked as hard as we could to make sure that the quality is high, that we have covered all the edge cases that we could think of, we have developed a lot of new tools to help us testing it and, finally, we have a high confidence that it's working as expected. And, yeah, this is why we're moving it to the public testnet. What to expect next? It will be for some days on the public testnet we are trying to see how it handles volumes for from users. We'll also generate some volumes ourselves. We'll check all the metrics to see how the network behaves, see that it's still in the expected parameters and if there is nothing weird happening, not outside of the expected metrics, then we should be ready to move to mainnet as well.
Speaker 5:What has changed with this new upgrade is we are now faster. So all the transactions that users sent to the network, they will see the feedback twice as fast as they used to see it before. So we have cut the latency, as Lucian said previously. Increased bandwidth, decreased latency. We have cut the latency in half. It is also much more simpler. The consensus. We had to redesign the consensus in order to decrease this latency and now it's much more simpler and even more secure than before. Also, the cross-chart operations for those as well, we have decreased the latency. It's half as well, and it also helps us with cross-chain communication. So interacting with other chains, it will help integrations to be simpler. Due to the changes we have done in Andromeda. Yeah, there are also a lot of building blocks that we have added and concepts that we have added into Andromeda. That will be building blocks as well for the next, let's say, latency-focused and bandwidth-focused release, which is Supernova supernova.
Speaker 3:I would ask, or if you could explain, let's say in terms of security, how this new consensus mechanism works versus the previous one or versus, let's say, the normal, normal, the most used PBFT mechanism in this sense, and how it improves, especially in terms of security, the whole network and the processing of the transactions.
Speaker 5:Yeah. So I'll try to keep it as layman terms as possible what we did with this change. So we have tried to remove single points of failure, because this is what the previous consensus model we had was generating, or was partially guilty of generating the latency increase which we now have cut in half. We have removed single points of failures, so this means that we also have removed equivocation here. If a block is being produced, there will be no block that passes consensus that can challenge it. So this is basically a part of the change we have done. We have increased the size of the consensus, or the number of validators that need to agree on a block and reach consensus on what is happening on a specific shard in order to have, from one shot, to have confidence that the shard or the entire set of nodes in a shard, have accepted the proposed block. Previously, for the execution shards, for example, we had a smaller consensus size and this meant that we need consecutive blocks in order to reach a certain acceptance threshold before that block is being notarized. So this helps with having a stronger security. It reduces the equivocation. So it completely eliminates, actually, the reverting of blocks, blocks that have already reached consensus, and how I'm not sure if one of the questions was how it simplifies the consensus or how it simplifies. It was about a little bit of, let's say, done. So each of these steps needs to be processed and have a conclusion in order for the consensus to be considered that is reached and the network agrees with the result, and it normally has multiple steps Previously.
Speaker 5:Even with the previous method, we have reduced the number of required steps from the initial five I think it was to three, but we still had the leader as a centralizing point inside the consensus, because the leader of the consensus, so the one who proposes the block, was also responsible for finalizing the block and showing the proof that consensus was reached. We found out that it's not needed to have a certain node to have this responsibility and we decentralized this responsibility of finalizing the block to all the parties involved in the consensus, so every node can potentially finalize the consensus round and provide proof that demonstrates the consensus was reached and how it's simplified. Is that also? Previously, due to the way it was happening, the block that was originally proposed was altered during the consensus and this is why we had to propagate two different blocks one with unfinalized consensus and one with the proof that consensus was finalized. We also split the responsibilities.
Speaker 5:Now. The proposed block is no longer altered during consensus and we have a separate proof that demonstrates consensus was reached. Consensus was reached. This has also increased the speed on which we can have the consensus finalized. This is also helping us for Supernova to decrease the round time or the block time there.
Speaker 1:I think that's something that a lot of people are super curious about too, where it's like obviously this is the first step of at least two more um that are that are planned to roll out, I think, this year um. Can you tell us a little bit about, like how, how these are steps that are then helping um with bernard? Next, like what, what are the next improvements we can look for there?
Speaker 5:yeah. So on, bernard, I will just touch briefly I I I will leave Robert to add more details on that if he wants. So what we need for this decrease of latency and to reach Supernova is just a way for developers to adapt to the changes we are doing, especially for Supernova, where we are planning to decrease the block time or the consensus duration time. For that some functionality is being brought in Barnard so developers can act and adapt their applications. But besides, this in Supernova builds on top of concepts we have brought in Andromeda. One of these concepts is the equivalent consensus proofs that I briefly mentioned before on how we reach consensus on a proposed block and how we prove that consensus was reached. This type of construct we will use also for execution, and Supernova will use this for execution results as well. We will use the same mechanism to propagate these types of proofs so that they reach the majority of nodes in a very, very fast time span, in a very short time span.
Speaker 5:Um and uh yeah, for supernova, especially andromeda, is a prerequisites, as we're using uh, most of the concepts from uh, from andromeda, um, it's uh. Also um helps with barnard because if we don't, if applications are not adapted, we might have weird scenarios where people are waiting for a specific time slot, but that time slot becomes 10 times smaller, so they need to adapt to these changes. The Supernova comes also with an increase of bandwidth. It was not necessarily designed to have the increase of bandwidth, it just came with the simplification of what we are doing. So it was a very good, let's say, emergent behavior. We're also building so after supernova, um, we're also building uh parallel processing on the node level. So, uh, supernova is actually uh helping with that as well.
Speaker 5:So we have this change, uh upgrades, uh that we we've been uh working on, uh, and each one like facilitates the next one. It brings things that the next one needs for for it to to come live, for some of these things that have built into these releases. We have been working a long time. Even Andromeda, we have started. We've checked like when, when, when, what was the first commit that entered into andromeda? And it was two years uh ago. So we have started. Unfortunately, it was not the only focus, so we had different things that we were working on and our focus was uh in multiple places. This is why, uh, we did not uh come faster with with this, but even so, there were a lot of things that needed to to be deployed on the network for Andromeda to become possible, and then for Supernova and parallel execution.
Speaker 1:Amazing. I'll throw the next couple of things just a quick things to to Mihai here. Next steps in Supernova sorry Bernard, let's not get too excited are tomorrow's testnet and then we have governance. Do you want to just lay out the next steps and how things are going to roll out, and then we'll talk about Bernard a little bit more.
Speaker 2:Yeah, so we tried to synchronize to say so the testnet upgrade with the governance call for vote, just because we consider that it that it's very important for everyone to vote knowing what they are voting.
Speaker 2:That's why we will try to have them started together, like tomorrow and afterwards, and afterwards trying to invite validators and also users to test the new upgrade, the new binary on the testnet, to try all kind of scenarios that we probably did not think about and see how the chain reacts to it. This is hopefully the not the last one, but the voting before the last one, just because for Bernard, we are going to again vote in an off chain manner, Even though the voting is going to be on chain. But actually we have the governance contract, the fix for the governance contract, ready for Barnard and it will go on chain with Barnard. So hopefully, let's say probably two or three months if everything goes right. But what comes with Barnard? That's on Rob, as Adriana already mentioned, just because the code is ready for Barnard, we are just testing it like crazy and also reviewing a lot of code and oh yes, barnard, you know it was developed in parallel with Andromeda.
Speaker 3:We have split, let's say the teams and the efforts, and then we said that, OK, andromeda will be this big consensus focus and everything for developers and smaller features or the governance fixes goes into another release. And then that was Barnard. So we had time. After speaker release was done somewhere in September, we started to push towards Barnard multiple features for developers, especially at the VM level, a set of new opcodes what Adrian have mentioned as well, have mentioned as well instance of timestamps and rounds and epoch timestamp and epoch start blocks so people can make their applications dependent on those and not on a fixed round time like six seconds. And then, when let's say, andromeda goes out and Barnard goes out, we are preparing tutorials and explanations on how the code is needed to be changed for applications, especially DeFi applications, starting from xExchange. Here that will be one of the first things. I suppose that after the mainnet we have Andromeda on the mainnet, we are making the final merge to Barnard, then the final testing penetration testing and performance testing for two, three weeks and then it goes to mainnet.
Speaker 3:One interesting thing about Andromeda is that we'll bring that together with Barnard, directly towards the sovereign chains as well, because we want more speed, much faster.
Speaker 3:On that we can iterate a little bit faster. And also Barnard will bring in some functionalities which are great for the cross-chain transfers. We are making a new endpoint for unsafe transfer and execute and unsafe execute on destination calls where developers have a little bit more job to do, but in general it becomes super simple for Bridge and all kinds of complex DeFi applications to treat the results. And then the good news doesn't stop necessarily here for developers, as somewhere after Supernova very fast, we are bringing ZK libraries into the virtual machine for developers and we are also bringing asynchronous calls v3 to the VM. We are somewhere 85% ready on that side as well. So the last 15% yeah, it might take a little bit more if we find out some critical bugs, but hopefully not. So that's it. There is so many features on Multiverse 6 and so many things you can build, so I am inviting everybody to build a wipe code or build test directly on mainnet if they want.
Speaker 1:I have one last question for you guys, and Adrian will let you go to bring down the block finality. Do you have any sense how many man hours went into all of this, like up to, let's say, just Andromeda?
Speaker 5:Okay. So for Andromeda, just production code, I think it's around 40,000 lines and besides that we have many, many more private repos where we have built tooling for testing and also the amount of effort that went into testing it. We have been running since, I think, november. We have been doing stabilization, then deployments with different changes, optimization in several areas and then a lot of chaos testing also. Also. We have a framework for introducing chaos, not just random chaos, but also some very targeted, like malicious behaviors, such as delays, failures, induced failures and so on, failures induced failures and so on. So it will be hard for me to quantify exactly, but it did take a lot of effort for a few people in the team.
Speaker 5:As you know, we have also been split in the protocol team.
Speaker 5:There are some people dedicated for the sovereign chain, so this has been going in parallel the sovereign chain as well as debugging and all the other releases that have been deployed on the mainnet.
Speaker 5:In the meantime, working on this Andromeda on Barnard as well, this Andromeda on Barnard as well, we also have a proof of concept working for parallel execution. That has been going for some time. So, of course, the most of the focus was on Andromeda, because this is the first step we wanted to go live with. But, yeah, also even inside our QA team for the protocol we have some new guys there and they have run up in the recent period and we have seen a lot of efforts there as well. We have automated most of our testing, which previously required also some manual let's say, the system tests to be done manually. And now I can say that, for for the next releases at least, we will reuse all of these tools we have built and hopefully we will increase the speed with which we can deploy the new stuff we're working on and build new tools build more, adrian.
Speaker 1:Thank you so much for coming on and explaining a little bit about andromeda. Good luck, I guess, for tomorrow. I hope you're ready. Um, I, I break a leg, or whatever they say in your business. Yeah, thanks a lot. Thanks for being here.
Speaker 7:Break the testnet.
Speaker 1:I think like one thing do you know the engineering hours that went into it?
Speaker 2:Just because it started two or three years ago. I think it's very hard to measure every hour, but if I will go through our spreadsheets, probably in 10 minutes, I can provide you an answer.
Speaker 1:Yeah, that'd be super interesting.
Speaker 2:And.
Speaker 1:I'm also super curious. This, like AI testing, must, like AI must be testing a lot of contracts now for you, and this is also going to be our next topic AI, everything about AI. This is still MAI March with AI themes, right, like we had the hackathon conclusion and this is our next topic. Let's talk about the hackathon. I'll see if I can bring up the slides for it and, yeah, introduce us to the MegaWave hackathon how it went, maybe some takeaways, and then we'll have one of the winners on next.
Speaker 2:What the hackathon let's call it like that. We tried to approach hackathons in a different way. We had some concepts on the table and then we thought, okay, what would fit perfectly for AI? And then we looked through what others were doing, just to compare, not to be very special, just because when it's too different, people will probably be scared, and we don't want that. But what we've added was the demo day, and this is the change that actually brought a lot of dynamism and a lot of change for the participants and also for the judges. We had more than 300 300 participants. They submitted more than 140 uh projects or how they are calling them, beatles it's very hard for me to pronounce it, but it sounds interesting Out of which we accepted 40, just because not all of them were respecting the rules. We defined the rules very general to say so, but there was a very specific rule that was saying it's mandatory to have a multiverse X integration.
Speaker 6:So as long as I thought that was crazy.
Speaker 1:I looked at it and was like what? I mean, you would never submit anything that wasn't for the blockchain that you're competing in the hackathon for. But you were like, no, we need this. And yeah, I guess you're right.
Speaker 2:Yeah, we had a discussion with the organizing team from DoraHacks and they said that they are aware of this problem, but they are working towards solving it and they have some very interesting AI tools which are automatically scanning the code and providing a score based on that, providing a score based on that and it will automatically say yeah, unfortunately we cannot accept your application for the hackathon just because it does not respect the rules. Anyway, We've seen lots of interesting projects and I'm curious I know we prepare some slides just to go through the tracks very, very fast, but we had three tracks. We had exactly Perfect. So this is the AI on Multiverse X track and we have Multiverse 6 library as the winner. We will not I will not go into details, just because we will have Sila with us in a couple of minutes.
Speaker 2:Then the second place here was a game. So we really appreciated the participation of the game. It was a builder from the ecosystem, so this is not a new builder, but the idea that he presented was very interesting and probably it will bring lots of developments in the gaming sector. So it was basically an AI actor who was acting differently based on the interactions or the input that it was receiving from the player. Anyway, and the third place is Xpilot, which is a tool let's call it like this built on Visual Studio Code based on MCPs and MCPs is a very trending word to say. So we are building a lot on MCPs lately and we are experimenting a lot also with general use cases, but also with very specific use cases for builders, and this is what we also are discussing with Xpilot.
Speaker 2:Just because, as I was mentioning one month ago, and I'm going to say it again it's very important for us the post-hackathon activity, so we are in close touch with every winner and also with any participant of the AI Hackathon, just because we believe that it's very important to have a running product deployed to production, so that you, in this manner, you interact a lot with what really having a product in on the mainnet means. Um yeah, and for that I think we have sila with us yes, we have the winner himself.
Speaker 1:Welcome to hey lucas, hey everyone, welcome to episode one of momentum. Uh, congratulations on your hackathon. Win sila, how are you doing today?
Speaker 7:really good. Thanks for the invitation. I'm really happy to be here with you all.
Speaker 1:I mean, I want to talk about your story a little bit right, because you've been in the ecosystem for quite a while and you've only recently really started to delve into coding deeply. And you weren't only involved in actually the product that you won with, but the second place NPC you also had a hand in. So take us back to your start on Multiverse X. How long have you been around for? How did you get started and then what got you like? Who put the crazy idea into your head that you could start coding two months ago and then win a hackathon?
Speaker 7:Okay, so I'll try to make it short. I started on Elrond in 2021, where I created NFT collection with the photography I created and I tried other art experiments. And then I'm really interested in data. So I wanted to learn more about on-chain data and multiverics, starting to use multiverates API and Google BigQuery later to better analyze on-chain activity and I had also a focus on NFT collection for example, the holders of different communities, how they were interacting. And later I started to build analytics for the Dinovox project, where I'm an ambassador, especially to track minting, listing and stacking activity.
Speaker 7:And all the time I was using AI to better understand how to code, because I'm not a coder myself. I only started a year ago with Unity and C Sharp and I created a tool which was called the Apex banner generator for the EAE team and it was on Unity and I was working with Boubou, another developer in the ecosystem, and in collaboration with him, I realized that I didn't know enough about websites, how they work, how to interact with it, how an api works and not this stuff. So in december I wanted to learn how to use it, how to create websites, and to better understand. I wanted to build a platform, a data platform which is called the explore multivarix, where you can track and analyze on-chain data, and it's a beginner friendly with a new code solution, great platform.
Speaker 1:I tested it. It has like this nice visualization, basically like a spotify wrapped idea. Um, really really cool yeah exactly.
Speaker 7:thanks um and um. I used uh ai to uh to learn about coding, how to use uh. For example, I use lovable to create a front-end and nice visuals. I use Cursor to go deeper inside and to create the background, the back-end and all the functionalities. And this is how I approach the AI academy. I wanted to create a website that kind of embodies how I used AI to become a developer here, and this platform Multivarix Lib. Its goal is to gather data, gather resources GitHub repository from all projects on Multivarix and all resources that we can use to build a project, projects on multivariates and all resources that we can use to build a project and we have an AI chatbot that is context powered, with the repository, to help us understand how to use it, how to implement it in our project and build what we want, and so developers at all levels can use it to find the right resources and then understand it and then implement it.
Speaker 1:Amazing, and I think, like one thing that's interesting to me like you built it because it helped you, or you're basically replicating your result here. What's sort of the next steps that you see for the product, for the project? How are you going to like develop it after the hackathon Something that Mihai, I think, would ask you as well?
Speaker 7:So I have two main goals to improve the platform. The first is about all the resources that are listed on the platform. I want to have more resources so we can use many SDK and all the APIs that are available to better understand how they are built. And I want to improve the analyzing of the repository. So you just open a repository, for example the Soxno SDK, and it will show you all the endpoints that are available in the SDK, all the parameters of those end price and how to use them. And then another focus would be a community component library where all developers and multivaries could share components they have created, for example, a Xportal login, team toggle or anything. They just share the code they used and then all other developers can just go on the platform, find the components, copy it in their project, and it will allow them to be much quicker in building what they want with a better understanding of it.
Speaker 2:Yeah, I want to congratulate you, Sila. I mean I already congratulated you during the demo day and during the hackathon. I want to congratulate you, Sila. I mean I already congratulated you during the demo day and during the hackathon. I want to double down on that. And now I have a question, more like a fun question, to ask how much from your project is AI and how much is Sila?
Speaker 7:Well, at almost every step I was using AI. I started to learn to build the ID with Grok3, deepsync and with ShedGPT. Then I used Loveable to build the frontend. Then I took this project and I put it in Cursor. To continue to go further, I connected with a MCP from Composio HQ to connect my cursor to my super base database to direct interact. So at every step I was using AI to understand it, to use it and to power my coding.
Speaker 2:So yeah, and how would you consider that the technical depth adds in this kind of setup where you are using all those tools for generating code and having a product in the end?
Speaker 7:Well, I think it's more about trying things and I started, I learned a bit and I tried to grow my knowledge, to use AI each step, to go a bit further, to ask questions, to find what could be improved. For example, I started, I built a small platform. Everything was inside. Then how to create a front end, a back end, then how to connect to the blockchain, then how to connect MCP, and every time I try to learn one thing and go further, how I can go deeper than that and grow my knowledge. So AI won't be the only solution for me to build, but I know what I want to do and AI will help me be faster in coding it.
Speaker 2:Yeah, yeah. So I think we are going to have quite an experience developing on multiverse X, and not only, but in the industry itself, and I really want to discuss about this and about vibe coding in general.
Speaker 7:I think anyone can become developers now because, for example, me, I only started learning about HTML since December. So everyone, it's just if they want to do it, the will to learn and the will to find the tools, to find the solution, to find the resources to build what they want. It all depends on them.
Speaker 2:Yeah.
Speaker 1:I think that's a great takeaway. Cilla. I mean really amazing story from artist to like, from Elrond to MBX, from artist to hackathon winner and coder and vibe coder and probably I don't know. I wonder what happens now if you take away all the AI. How good a developer are you without the training wheels? Or I don't know if it's fair to say that they're training wheels, they're real tools, but like, how would you fare without it right now?
Speaker 7:I think I would be much slower because I would need to go to documentation to know exactly when I need to do my Ecolab, when to go to new line to the basic stuff that I don't know by heart. But I think I could still try a bit and I would go more on Unity and be more used.
Speaker 1:Amazing Sila. Thank you so much, Big congratulations. What a story.
Speaker 7:Thank you everyone.
Speaker 1:Thank you so much for being on here and we'll talk very soon, thank you.
Speaker 7:Thank you, Sila.
Speaker 3:Cheers.
Speaker 1:Mihaly, back to you was this was the first track.
Speaker 2:Very interesting development. I was very happy to see sila uh growing to say so I don't, I don't know developing from um, an artist, an nft creator, to a builder, and basically this is and this is something that we have to understand that the things are changing and people are developing. And if nfts were a thing three or four years ago, I believe that now we are going into the direction where having a product is the thing itself, or building a product, and that's because anyone can build a product, but we will go into that a little bit later. Could you please Exactly so this was the second track. Very interesting track AI agents. And we have as the winner the Joe AI or Joe A or we will figure it out how to call it as the winner. A very interesting project. I will not go into it. Ah, ok, I will leave everything to Mika, the creator to say so, of the Joe AI, to tell us more. So, mika, is it Joe AI, joe or Joe J? All of them.
Speaker 3:I think you're muted. I think you're muted, unmute yourself please.
Speaker 6:Oh joyai, that's how I say it, but I think the people, the community, they will find their name and eventually you know the majority rules.
Speaker 2:What's joyai? What's joyai? What is it about? Joy, okay.
Speaker 6:Well, so joy started back in end of January, so that was when the first line of code was written and the progress was quite steep. And the thing is what makes Joy so different from other AI agents is that it's completely based on Warp. So every capability that it has that includes performing transactions on chain, but also reading from smart contracts on chain it all happens based on warps and for this we had the whole February, february and March lots of different upgrades for the protocol to support all these developments yeah, let's not get into it you know what's.
Speaker 1:February and March we would.
Speaker 2:We will have May, which is perfect, that's that's the one that we'll miss, right?
Speaker 1:I don't think that's okay too. But okay, give us a little bit of an um, an overview. Joy um, agent Swarms, you mentioned this like what's new. You mentioned the integration of warps, the protocol. I know there is a DAO function Like what is the general idea? Like what problems are we solving with this project, like what was maybe the initial idea for it and how do you plan on developing it further?
Speaker 6:Well, the initial idea was definitely experimentation. Do you plan on developing it further? Well, the initial idea was definitely experimentation. That's how it all started. How can we expand more protocol, use it ourselves? You know the whole dog fooding kind of concept, where you use your own products to make it better and it turns out.
Speaker 6:We had the conversation back in December, I think, with Mihai also, where we said, hey, how can warp support AI agents? And it turns out, as we like keep kept developing um warps and did a lot of experiments, that it actually is a super powerful tool. And there was a listening before, uh, but I also need to say super bullish on all the uh protocol advancements and super bullish to see you also all in the office. I think it's what, uh, eight years, no, nine, your time almost uh, 9 pm at pm and and I listened in and I heard a lot of them MCP concept and there's like we could talk hours about this. And MCP is the model context protocol. It's basically it gives a wide range of tools to an AI, because an LLM an AI is basically. You know it predicts text output, but it cannot really perform actions by itself, like it cannot browse the internet. It cannot browse the internet. It cannot run code. It has like tools, like a search engine functionality to interact, to use search engines. Basically, it has a code interpreter to run code. And for AI to actually be useful in the real world, we have to connect it and give it all the tools possible.
Speaker 6:And I believe blockchain is a bridge between real world, virtual worlds, metaverses, ai and humans and with this bridge, uh, as this bridge, ai has to be able to inter, like, perform anything on on the blockchain.
Speaker 6:This is how joyai started and, of course, like for specific to the hackathon, I or we were working on Swarms, which is a collective of all the AI agents that actually perform work together.
Speaker 6:There was a fictional company that had one agent generate the ad copy you know the text that is supposed to convert users and then another agent reviewed this copy text for efficiency can be also for company policies and then another agent that then actually did the proposal to interact with also a fictional, but you know, the more smart contracts exist in the real world, the more useful it will become. It was a demo smart contract that actually would, if called the purchase function on it, it would actually then display an ad on, like you know, know, google adsense or uh, could be any kind of tool like this. So, and once this proposal was active, we had all the other agents come here, see that the proposal is active, evaluate it based on their training, because agents can be trained and get given all to um as as much as you want and need, and to make decisions based on that and approve or decline.
Speaker 1:Yeah, pretty impressive. Where do you see, like you already mentioned, one application there. Where do you see something specific of like companies coming in and using these really as if they were employees for specific departments? Like, what do you see as sort of the prime use case for something like Joy?
Speaker 6:Well, actually, I have to say, and being fully honest, I wasn't sure if I'm going to take on the swarms for the hackathon, because it was very ambitious to do this in like two weeks, as we thought it was, and then, like we had the extension to three weeks, but we're still very extensive, regretted it every day, um, but uh, actually like really happy with the result.
Speaker 6:I think also the the video shows it quite neatly um, of what is what is possible.
Speaker 6:And to go to the question, um, like all the work like people working in an office there's all like computer work, all all the work like people working in an office, there's all like computer work, all all over all different kinds of works, and there's nothing that prevents an AI to do perform this work.
Speaker 6:Ai is software, um people just moving bits around, and AI can do the same. And I think if they are connected to the blockchain, then they can actually manage assets for the company, for a person, and perform all the tasks, and the human operators can just approve, decline, give some feedback for the AI to do that work for them, and that's actually the grand goal. There's a tight integration with Pyramid, which is all the goal, to bring teams of any kind, whether that's multi-sig based or whether that's a dial based um to the blockchain and then, once you run a pyramid company, then the goal would be and I think we're quite getting quite close uh, is that you can you hire and a worker and employee basically with just a click and they can perform some tasks for you.
Speaker 1:I think one last question, like next, couple of steps for Joy just in one or two sentences, and then we have another track that you can probably speak to, so we'll keep you on for another minute.
Speaker 6:Sure thing.
Speaker 1:No, just give us a quick Joy like your next couple of steps or something what your future development is looking like.
Speaker 6:So the next step for Joy is definitely not to immediately release this one protocol. I think the biggest challenge is to get things stable and reliable. There are good results, but you want things to be. You don't want the hallucinations to be reduced so that there is nothing performed that the user doesn't want to. Because we have different modes. We have manual approval modes, we have autonomous modes, which is some know it. In Cursor, for example, it's called the YOLO mode, where it just performs steps by itself if you're high risk. But of course, these example it's called the YOLO mode, where it just performs steps by itself if you're high risk. But of course, these things, as AI improves, as joy, ai improves, as there's more guardrails, all this is going to improve and the first step will be, of course, to have individual agents that can perform work to you. And as they get stable, then we're going to move over to warps, no to swarms. And as they get stable, then we're going to move over to warps, no to swarms.
Speaker 1:Well, since you're talking about them, let's talk about the next track. Mihai give us an intro, and then maybe Micha has a word or two to say about that.
Speaker 2:I just want to say also the other winners of the previous track, say also the other winners of the previous track yes, so we have Laika, which is again. So for the second place, we have Laika, which is a project built by a community member, el Pulpo. If you know Mr Pulpo, a very interesting AI agent based on Eliza. He added new actions that could interact with HATM or X Exchange, which was very interesting. So anyone who wants to build on top of that can freely do it. And then we have Hivex, which is another version of or lower version of, swarmz, which was thought about being deployed multi chain. Then on the third track, which is warp. So we had a track only for warps, the protocol that Miha developed. If you could change the slide, mr Lucas, thank you very much. We have one click, warp, which is the winner of the track, if Micha or Lucas would like to add on top of that, just because so, x Alliance was the sponsor of the AI works track and during the demo day probably I did not fully describe what demo day was.
Speaker 2:Demo day was. The purpose of demo day was to bring in front of the judges the most voted projects for being, let's say, seen in action. The judges had to vote which project they wanted to see, so it did not have any connection to the evaluation of the project, because they would probably. We had the situations where the judges would say, okay, this project is five out of 10. But I really want to see it, and after seeing it he changed his evaluation. We had this kind of situations and the project in the demo day should have presented the way how the product, the platform, the application, the agent in some cases would act and how it would work. How it would work. We had judges from Alibaba, from Google, we had judges from Tada, we had judges from Skynet, from Animoka Brands. So it was quite a very it was an interesting discussion. I'm curious how Micha saw the entire dynamic inside the demo day, the entire dynamic inside the demo day.
Speaker 6:Yeah, it was really smooth. I liked seeing all the other products that were built. I think people had really good presentations and yeah, of course, getting in front of judges from all of these industries and all of these companies is a bit nerve wracking. You want to have a really good presentation and, yeah, I think it was a great experience and hope to see more of it.
Speaker 2:Great, and would you like to add something on the one click warp and or any of the warps projects that were? Maybe let's take a step back and make him so.
Speaker 1:Micha is behind warps um projects that were uh, maybe let's take a step back. So, misha, is um behind warps? Right, that is a protocol. That is community development, uh, developed um. It's something that x-alliance also helped launch with a, with a small grant, um, but it's. It's a great um, I think, advancement and in what um mbx can do and also what blockchains can do in general, and then also what exists that's similar on other blockchains that may not be named on the show, but maybe Micha give us a quick intro like what problem are warps solving? What are warps? And yeah, then maybe talk a little bit about the hackathon winners and how happy you were with this track that was dedicated to your protocol.
Speaker 6:Yeah, okay, so there's a lot to dive into. So first, I was really happy to see it was a great surprise to have Warps being a separate track, an individual track, because I really believe, like warps can make a huge difference for the ecosystem. Like warps is, you know, this standard way of doing of instructions, instructions to perform a transaction, and this or warp can then be used. You know, by a simple link you can share it with your friends, you can share it on social links, in private messages. But even in other integrations we have new integrations coming up. I'm not sure how much we can link there, but I think Mihaly mentioned it. So I'm going to say it's like part of the DeFi wallet, the browser extension which is going to unwrap even Warp Show to you directly on the XFeed. There's going to be more integrations, like with virtual worlds, so each of the with the AI agents within choice, so each of these integrations all feed of the single base of warps that we have. Like once you create a warp for a specific smart contract, then all of these integrations benefit from it, and this means that it's really important to have a huge repository of warps to cover as many smart contracts as we have. I know there's a lot of developers watching, so I want to give this a shout. Uh, this call to action if you have a smart contract, create a warp.
Speaker 6:We have new integrations, uh, new um tools created as part of the hackathon, that help you with that. We have useWarpto, where you can create a warp without any coding. We have even cooler like AI. Even cooler, we have AI generations coming up with Enjoy as well. All of these tools are going to help you create warps and once you create a warp asset, it's going to work across all the integrations that we have. Lucas, need you to give me a hint on what I missed, because there were a few questions in there.
Speaker 1:Yeah, I think that that was a great overview.
Speaker 1:I think, just in general, talking about warps are really the idea is to to warp the blockchain into something that is um, user friendly and and you don't need to leave um the internet anymore the web 2 internet to to perform blockchain actions.
Speaker 1:Um, I'm running the uh, the warp uh protocol website here at the bottom, um, if you to check it out, but I think that's so fundamental that one it gives a tool to AI agents to create things and really directly interact with humans, but also humans can obviously create these warps to just let users stay on the website they're on and not have to jump through a bunch of hoops, and and that's just really great. Um, again, we do have to give a shout out here to solana blinks, um, that that kind of pioneered this concept, but, um, on a technical level and maybe you want to say just a couple of things on that, michael, because you'll you'll be much better on that um, the warps are a huge improvement upon what Solana is doing with warps, uh, with links, sorry yeah, um, so, as, as with everything on motor is xb.
Speaker 6:Uh, robert talked about the permission less dash, zero, zero, zero um standard or or what's upcoming, but, as, like, it's all about putting things on chain and warps, um emphasize on this like they build on top of this, because with blinks, they're very server-based, which means you have to run servers, which comes with costs, which comes with uptime and downtime, like, of course, blockchains can go down too, but we're safe almost for a sec, so all good. Um, so there's always uptime for warps. Um, also, a cool thing is, once the warp is created, it's it lives within a transaction, in the transaction data, so it's immutable. It cannot be changed. Like you get what you see um like if you use a warp in in 10 years from now, it's going to be the same. No one has changed the code on a server, as you can do with blinks, which has some security aspects to it, blinks, which has some security aspects to it.
Speaker 6:Um, yeah, probably missed some. There's a nice graphic. Maybe we can pull that up. Um that we have. We have, uh developed a protocol a lot further since then. We have like branding for warps now upcoming, which is also um branding, information inscribed, and with inscribed that means stored within the transaction on a blockchain which is going to pull in. You can do custom branding, you can upgrade warps, you can assign aliases to works. We have the registry, which is also off chain. On Solano, for example, we have this as a smart contract, so that all warps are registered within a smart contract, which is also pretty cool to have this on chain, as we do in motoristics. Uh, I think, uh, I personally uh like how, how it has turned out.
Speaker 3:Uh, back then at the collabathon, I briefly talked with robert about solana blinks and and uh, the whole concept was very uh blurry to me back then, not sure yet how to design this, but in the end I think it worked out great robert, you wanted to jump in yeah, I wanted to to to stress a little bit on the warps and how can the warps as a standard to be part of the standard of the whole multiverse ecosystem and what would be, from your step, do you think, Miche, in how could, let's say, this kind of standard get everywhere on the whole ecosystem to use it?
Speaker 3:This is one of the things. And the secondly is how can Warp be made in such a way that it is, let's say, there are no trust issues, it is totally safe, totally permissionless, totally readable for the user what he will actually do and all kinds of steps like that, Because you know, sometimes people on the blockchain they don't really want, let's say, blind signing or stuff like that. It happened with Bybit. They lost 1.5 billion. So, and in terms of standards and getting a standard to be used, there is a lot of work of business development, not necessarily business development, but adoption work. So what would be the steps for you to think that the WORPS is, let's say, adopted by everyone and we start to use it and we start to see demand for it?
Speaker 6:oh, very good question, and it brings me to a point um, I was thinking about, um, actually, we had the during the demo day, so to make warp six uh adopted, basically, I think they need to become very accessible and very easy to create. That's the first step. I'm, with the hackathon, we have done a huge step in that direction with AI creation of warps, abi to warp converters I think that was a, the winner, actually, and also a second project that took this task on. So, abi, for those who don't know, it's like the definition file of how to into how to interact with a smart contract. Uh, it's, it's automatically generated. Uh, when you build, when you uh call the smart contract and you can translate this into a warp, and, with some ai magic on top of it, which, you know, adds a nice title and description, um, you end up with a, with a warp directly. So I think, if we automate this, if we make this as simple as possible, uh, then it will be like then we will end up with it with a huge set of of warps and, yeah, you can even go as far, I think, as of you know, scanning the blockchain for newly deployed smart contracts, um, and ai maybe even decompiles that this was the topic of the we had in the chat at the demo. They decompose the WASM bytecode that we have and generates a warp based on this, so that there's some magic there.
Speaker 6:I think in the end we will end, we will end up there, and the second part of for adoption is definitely also there, like it has to get to the point where if you don't have a warp for your smart contract that you developed, then you will miss out on so many opportunities, and with opportunities I mean people will not be able to use your smart contract with integrations like, for example, a virtual world, the metaverse world, cho AI and other AI agent won't be able to interact with your smart contract if you don't create a war. So there's some pressure, um dynamics there too, and for the security, which was the second question or it's, it's a very deep dive, I think. Um, there's like different dynamics that we can have, like trust status, some governance functionality which can assign trust status as part of the registry to a warp, so that scam alerts and all this can be presented to the user. But of course, security never ends and warp is a community protocol. It's not just done by billy by us.
Speaker 6:Um, so, if any feedback is welcome, um, as we see, new threats emerge, always, of course, trying to be pre, uh, you know, looking into the future, seeing how we can prevent, uh, any, any issues. But, of course, uh, if there, if there's anything from the MultiOS 6 team, anything from developers, anything from anyone really like who wants to participate. Github is everything is public. We have a group and, yeah, would love to see more people contribute to the protocol because, as I said, I think it's going to be very useful to the ecosystem.
Speaker 1:Yeah, I'll just add here this is fully open source. Anybody can jump in. These are very much some improvements through the hackathon that we just saw, that can be built on. Now it's very easy, I think, to create warps of different forms, even complicated ones. I do want to do one quick demonstration of it, and this is you know, this is an interactive show, of course, right? Oh, geez, where's the oh? Here it is we.
Speaker 1:This is a warp. So everybody watching you have, however long this QR code stays up here, scan that. That will warp you in for a proof of podcast, proof of pod um, nft, um. Anybody sending a tiny transaction to it, um, will receive an exclusive nft um. So please go ahead and do that now. Um, maybe we'll share the link also a couple minutes after the show, leave it open for a little bit and then close that warp. But this is something you know like. You can literally just go on your phone. If you have Xportal installed, then it will be quite quick and should work, hopefully flawlessly. Micha had about three minutes to code this, so I hope that everything worked out on that end. But, yeah, this is to Warps and yeah, thank you so much, micha, for all the work you did there and, yeah, great, great job. Can't emphasize enough how cool a development that is. So we'll see how it goes. I see Robert has some problems. Robert needs to jump into development himself.
Speaker 2:Before, before having me go, I know. Mika is a very hardcore developer. I would just have a question, mika how much are you using AI in your development activity lately?
Speaker 6:Let me quickly step in about the invalid QR code. I hope it's not there anymore. Maybe, lucas, you can bring it in. Don't use Xportal. You know that will connect the camera, but use the normal camera that you have. It will prompt you to open the website, which is usewebto.
Speaker 6:And regarding AI, I think like there's certain dangers, like if you only use AI and you as the developer have to gatekeep what the AI is able to merge into the code. So I personally mostly use AI, especially if high stakes you know to quickly code complete, tap completion just to be faster, but not really to, you know, write complete features for you. Maybe you know like do some prompts and then do some edits afterwards. But I think AI it's definitely getting to the point where you're just prompted and it will write you the code and yeah, but I don't think we're not quite there yet. So of course, be careful.
Speaker 6:There's a lot of developers posting on X. I've wipe coded my whole app and now I have to. Now I have to take it down because I'm not able to maintain it anymore, but of course things keep improving. I think it's definitely a thing, web coding, especially when, if you use it for you know side projects, super cool, you know, to create quick games and all this. And now my question to you, to me, hi robert, maybe out there, and also, if he's still here, um when? What coded supernova? Uh, maybe next month uh.
Speaker 3:Supernova. You know, when you have this kind of super critical stuff, ai cannot really help you on that. On the super deterministic, super exact code, it's not there yet. It can help you a lot on test cases and building tests and creating tests and reviewing the code and everything around it, but for those super technical stuff, not yet. I'm glad you say this, never.
Speaker 6:I don't want to see Supernova next month completely webcoded. It's going to be quite dangerous.
Speaker 1:Mikia, thank you so much for being here. Thank you for vibing with us. Even not vibe coding yet perhaps, but that's certainly coming. That's also going to be our next topic. But yeah, congratulations on the amazing work that you do every day in the ecosystem. Congratulations on the hackathon win. Congratulations on birthing Worf's protocol getting its own track even and I think there was some great progress. Looking forward to seeing what is happening with Joy and Pyrami and everything else that you do next. Hopefully we'll have you back very soon. Thank you so much for being on.
Speaker 6:Thanks too, and yeah, enjoy the evening. Thanks, miche, hopefully we'll have you back very soon. Thank you so much for being on.
Speaker 1:thanks, too, and yeah, enjoy the evening thanks, miche Mihai, back to you, back to vibe coding back to vibe coding.
Speaker 2:Yeah, it's not vibe talking, let's say like this Robert was part of the judging of the hackathon and um you guys, I cannot hear you uh, I can hear you, robert.
Speaker 2:Very well, robert, you're here. Yeah, we can hear you. Um, anyway, so Robert was part of the judging, or judges, of the AI hackathon and when he'll be back with the sound and everything. I just wanted to discuss a little about the fact that I think we had around 20% of the projects being built fully by AI tools or in any way, ai in some way or some form, and we managed to figure out which code was generated just because parts of it did not make sense generated just because parts of it did not make sense.
Speaker 2:And this is probably taking us into the subject of vibe coding. This is a term that probably, if you go and scroll the Twitter or technical papers and articles, you will see that it becomes a thing. And for those who do not know what vibe coding is, it's actually a new way of I don't know if I can call it programming, but it's a new way of building or describing what you want another program to build for you and you describe it in simple words or, let's say, in English. I think I've read somewhere that in the future, people will begin programming in the English programming language, anyway, so you send those words, simple words to the AI and the AI will generate the code for you. It is very cool, and the AI will generate the code for you. It is very cool Now thinking that you are just using English or Romanian even as programming language, and it actually changes the focus from writing the code manually to just guiding an LLM or a tool into AI generating code.
Speaker 2:And what does this mean? It means that people who never wrote a line of code in their life will begin experimenting and building products or ideas. They will not be products ready for production. They will not be products for mass uses, because end users are not safe. So end users the people like us who are staying on the other side of the screen and looking at a product that a vibe coder created is not safe. Why? Because, first of all, the quality of the code is bad. If the product itself reaches a certain level, like 5,000 lines of code or even more, and you want to add extra complexity to it, the code is not optimal anymore and is not efficient anymore, and you have some bugs that you don't know about and you don't understand what's happening there.
Speaker 2:In general, the vibe coder to say so. He does not need to understand what's there, but for the safety of the user, of the end user, he has to understand that. And this takes us to the next point where it should raise huge security concerns, especially when we are talking about blockchain related apps vibe coded just because there are funds and value involved. Ai generated code could and will introduce lots of vulnerabilities if not properly vetted, and it will definitely raise privacy and security issues, and this is something that we experienced today. We were looking through some something that we experienced today. We were looking through some product that was vibe coded and we saw some very strange behavior and we had to understand why was that happening. And let's say, we have a programmer's background, so it took us a few minutes to understand what was happening there, but I'm thinking about someone who never wrote a line of code in their life.
Speaker 3:What will he do, you know? The thing is that I had to challenge this. I get that.
Speaker 3:AI is helping us a lot and it is great. And wipe coding actually started back in the 2000s when those cyberpunk people were wipe coding different types of 2D and 3D games, or when Flash started and people were just starting and building games in a few days because it was easy and simple building games in a few days because it was easy and simple. But in all of that, you know, sometimes you have the creators, you have those people who have the ideas, but with the ideas, if they don't have the basic knowledge about programming, about that, they don't have the basic knowledge about how to prompt the AI to do the stuff it needs to do, because you don't have that kind of knowledge around you. So you must know a little bit around it. Yes, it makes it much easier to develop, to create, to be creative, and that's super, super nice. However, and wipe coding is fun when you are, let's say, senior or junior developer, but for normal people I wouldn't say that, let's say, in one year or five years or ten years, eight billion people will wipe code.
Speaker 3:No no, no, as you don't have, let's say, eight billion content creators on the Internet.
Speaker 2:No, we don't have eight billion.
Speaker 3:It will not happen with AI wipe coding neither. So it will be a set of people, but the thing is that it makes you much, much more efficient and it creates this kind of loops where you can iterate over and over again Around the wipe coding. One thing which would be nice is to have, let's say, the when somebody posts about an idea, about the wipe code and how he got to that result, to be able to replay it.
Speaker 3:you know, yeah, that's a very nice idea yeah, that, that's that's another hint for something, but for something for those who understand what for something.
Speaker 2:For those who understand what, anyway, it's very funny how you say it it's it's vibe coding with. You have to feel the vibes of it and and don't get me wrong, what I'm trying to say here is the fact that people have to be aware of the risks and challenges. Uh, that vibe coding brings them to and to take it as a new educational opportunity. And you are experimenting and you must learn what's happening there. Why did it write that line of code in that way? Why did it create this wrapper? Why do we have this callback?
Speaker 2:This is very important and it enhances, enhances productivity. What does it mean? It means that if we, for example, have a team developing a backend for some kind of product, this team does not have to wait for the frontend team to develop the product, the front end itself, but they can generate a mock of it with minimum functionalities and they can start testing their own functionality. And actually for programmers and experienced builders, it's actually a way of also learning and understanding how other products or other parts of the entire product itself if it's a bigger product are being developed and how it better contributes. So it brings a better lens over the, let's say, high level or big picture view of the product itself. I have a question on that.
Speaker 1:How good of a learning tool would you say the vibe coding is Like I've never coded anything complex. I'm starting with the AI. Is this something that kind of stunts my growth because I heavily rely on it? Should I first go through a real I don't know Python course or something to have some basics and then jump into vibe coding? Or is it a cool idea to just start with the vibe coding and then get all of that?
Speaker 3:I think that just reading a few tutorials and seeing how others are doing, then you start to get hang of it. It gets frustrated at the moment. When you get the code, you start to run it and then let's say the compiler or anything, visual Studio Code or everything, it gives you hundreds of errors and you don't know where to look at it and you will prompt the errors to the code who generated it and it will still give you erroneous code Because you don't know programming. You will not be able to fix it. But that's one of the things in this sense.
Speaker 3:But once you get the hang of it and then you start to eliminate these basic issues. It makes pretty simple to iterate on your idea whether it is feasible, it is okay, it would be used by others. It looks as cool in an app as it looked in your head. That's what it is, and then you can directly start it on mainnet. Why not? Why?
Speaker 2:Creating a special mainnet only for that.
Speaker 1:That's interesting.
Speaker 2:Anyway. So we had a colleague of ours, anca. She never as I see Alita in our chat is saying she never wrote a line of code in her life and I told her why won't you build a Flappy Pig game? And she did, and she said that it was so fun to do it and it was so fun to play the game that she thought about to do it, and it was so fun to play the game that she thought about even though probably it was a suggestion that she, she envisioned how to design and what kind of effects and what kind of colors to use, and so on.
Speaker 2:I believe that it definitely. It will bring a lot, a lot of fun and why not? Innovation, just because people from other industries or other segments, with different backgrounds and knowledge, will start thinking about products and experimenting but being dependent on ai. That's very wrong and this is what I I also said last time and I this is also as a response to your question, lucas I will suggest that first you will have to understand the basics, learn them. What's a variable? What's?
Speaker 1:a.
Speaker 2:I don't know a while loop, what's a for, what is for, it's not for, like a number that's for two and what's a function, and things like this. You you definitely have to understand them before going deeper into building your product itself.
Speaker 1:I actually know the answer to all all of these questions? I think, so I'm going to start vibe coding. I already promised, uh, you that I would vibe code something, but maybe we'll do that on a um, yeah, on a different stream. Um, I had this slide up the entire time. Can you tell us anything about this slide that I'm bringing back up?
Speaker 2:yeah, this is, this is a slide that brings good vibes and, yeah, just because summer is coming, uh, that's why probably every uh person, every, that's why probably every person, every, let's say, community member and not only can start and building good vibes, and probably we will follow up on this. It's going to be so fun and so interesting that I believe we can create I don't know a vibe out of it and I think vibing is going to be the, the activity of um, of the next uh, of the next month oh, this is interesting.
Speaker 1:It's, it's not may, but I don't know. Vape, I don't know. We'll need to figure out, uh, how to market that, but but definitely very interesting. Um, I think the, the vibe coding, I don't know, it's becoming a bit of a movement with all of these great things. I don't know if it's a moment or if it's something to stick. I don't know, robert, what's your take on fad versus something that's here to stay and maybe bring some innovation, and then we'll move on to our next and last segment, because we've been going for however long almost two hours our next and last segment, because we've been going for however long almost two hours again we are taking too much long.
Speaker 3:Um, the the funny thing about white coding is that if you can make people to recreate the same thing, and then you build a lot on this open source and then you build in such a way that it is permissionless, you do not add any kind of control of admins, and you deploy it directly on a layer one. Then it will start to bring a lot of value in that sense. So even the permissionless 0, 0, 0 can make it super efficient because you can wipe code a few front ends, you can wipe code a few back ends, make the system to compile it to Wasm and directly deploy it and then you just enjoy and have fun with it. So especially, let's say, through when you have permissionless, composable open systems, the thing is that let's say you don't like something on the front end, you don't like something how you interact with an application. You can just wipe code the front end for it or whatever you want in that sense.
Speaker 1:You can just iterate a little bit.
Speaker 3:You can just run around see okay, this is the existing one in the open search as possible, I would like to have this, that and that, and then let's see if it's feasible, why not? That's how you do it and all these kind of you know innovation cycles. It's all these kind of you know innovation cycles. It's all about trying out more things, trying out new and new things, until something sticks and white code can make. You can make the like 10x or 100x speed. Yeah, on seeing what will stick and what doesn't stick speed on seeing what will stick and what doesn't stick.
Speaker 2:Yeah, and the process is so simple. So you just have to choose an AI coding platform. Just because I see that Philip said something about Coursera, you can use Coursera Replit, bolt, lovable, windsurf, so many. Then you just have to describe the idea so many. Um, then you just have to describe the idea, and here are so there's so much content about how to describe the ideas in a better way, in a more descriptive way, in a more technical way, in a more artistic way, in a more vibe way, and so on, and then you just have to review the product that the ai generated and then you just have to refine it and say, yeah, I don't know, decrease the padding in this area, move this.
Speaker 2:Uh, you. So for me it would be probably easy to move this object somewhere, but I don't know how a person who never, uh, coded probably this rectangle more to the left, or give me this circle, I don't know what shape and color, and then, very important, and this is probably which brought the most of the popularity to replit, but nowadays also, course or gives a lot of support on that is deploying it, because that was a problem. So, linking all those tools and what kind of version I have to use here, what kind of wrapper I have to use here, how do I link my product to a database and things like this. Well, it's so much easier to do it with help an assistant like Coursor and it's even called Coursor agent or Replete agent it's so much more easier to to do it. Yeah, definitely, uh, we will uh come back on this and it is going to be something very interesting for all of us. Um, probably next time, uh, we will have uh very good.
Speaker 1:One to two weeks. I also want to talk about vibes, but IRL vibes for this last segment, just because we're so far advanced in time won't linger on too much for too long, but very important I think. There's so much going on in the world and like so many developers actually coming on, just wanted to highlight sort of the the outreach that has been going on for the past six months or so, with different hackathons in different countries, with an entire university actually attending. It were like select students engineering students at a nigerian university attending their first ever blockchain event. Um, we had some great hackathons in in london, I think. Yesterday we had another event in london. There's something I think tomorrow um coming up in brazil. We had an irl hackathon in turkey. Um also, I think at least one team um may have won a hackathon prize in in the big one miha you can correct me on that. There were a few entries at least.
Speaker 1:Um from turkey from turkey, yeah from, or that we can see on the top right there.
Speaker 2:Yeah, there were quite some.
Speaker 2:I think WarpX and DeFi Verse were coming from this stream of projects and builders. So what we have seen is actually that through our initiatives from Turkey, through our initiatives from Turkey with the help of Ryzin and Patika, and also going through them into India, Indian, let's say community communities of developers, we've seen that almost 20 or 25 percent of the participants at the hackathon were coming from that effort. So if there are discussions about what's the outcome of all those initiatives that we have in the educational area and educational initiatives, well, the numbers will probably change in the near future and the results will probably be visible in the mid term to long term. Just because vibe coding a product, it's easy and you can directly deploy it on mainnet as robert is suggesting.
Speaker 2:So so much on doing it. But building a good product, a solid one and a secure one, a good product, a solid one and a secure one on the mainnet it takes a little more time. On DevNet, probably we will see lots of projects, but for having them on the mainnet and fully recommending them it will probably take a little more time.
Speaker 1:Yeah, and it's just really to show like these developments are active, ongoing. There is a new partnership with NearX in Brazil that's also ongoing, quite similar organization to Ryzen, but just wanted to highlight you know, obviously Robert is active in one of these pictures himself. This was in Spain. So there's a lot of things going on at the moment, and just wanted to also show some some real faces doing some real work. Bottom right, we would find at least one hackathon winner in there. And yeah, just just good things going on and and this keeping this momentum and bringing this in. Also want to talk about a brand new initiative that we launched last week, which is the MVX Ambassador Program. Just quickly on the idea, and then maybe you can add a few things there, mihaly.
Speaker 1:This obviously was something that was planned for quite a long time. We went through different iterations of how to do it. Should it be an application process open to everyone? Should it be just a select group of people that already are doing like admin work, if you want to? And like what kind of people are we looking for? And so right now, what we're starting with is six people in a pilot program, just seeing really what works.
Speaker 1:These are people that have been in the ecosystem, around the ecosystem for a long time have really proven themselves, and there is an emphasis on building communities really on the ground.
Speaker 1:So they are geographically strategically placed in Australia, in the US, in Europe, some outposts there, and that is going to expand from here on out.
Speaker 1:But we first want to really try to see what works, because, you know, before we put 100, 1000 ambassadors somewhere and have random application processes that are hard to oversee, we need to build best practices and see what really works, and this is exactly what this is designed to do now.
Speaker 1:The ambassadors will be equipped by the foundation with basically a set of of tools that enables them to go to events, run events, throw events, uh, attend meetups and do all of these things and really start spreading the word locally and also through blog posts and educational posts online. So there, it is really this two-tiered idea of having in real life events attending, hosting and then having educational courses, both live, for example, you know, coming on the show or writing a blog post or hosting a webinar can really be anything and then, yeah, just starting to spread the word, and I think that's a, it's a really, it's really important that this is finally happening. It took perhaps too long, but it's being taken very seriously, and want to iterate on that very fast and maybe, mihai, you want to add one or two things that I might have forgotten here.
Speaker 2:Yeah, so you started it as a brand new initiative. Well, it's quite brand old initiative, so, yeah, but this time it's a pilot for three months. We are running it just because we want to learn how we can make it as efficient as possible with best results as possible. So we, for the next ambassadors, we will open the list for anyone, basically, and from day one they will receive a pocketbook describing hey, this is what we have seen, it is going to happen if you do this and that, this is how we are going to support you, and we have some very well and very specific ways of supporting, either from the foundation or either from X Alliance, either from the foundation or either from X Alliance. What actually we did? This time? We are trying to bring attention back to the program itself and, as you very well said, put some boots on the ground. This is very important. Put some boots on the ground and we funny enough, we were joking that we are targeting to have at least 36 countries covered by the end of the year.
Speaker 1:All right, 36 out of. How many are there on the globe that are officially recognized and big enough? Too many many are there in the in the on the globe that are officially recognized and big enough, too many? Well, and you know, like this is what the the visual that we were seeing there is is really like putting the flag down as well. I mean, we we see it right now with the ongoing efforts in the us um that are really accelerating things there. Um, we really, really, really proudly, I think, had our first event in Nigeria recently.
Speaker 1:Brazil, again. These are huge countries with huge potential, great talents there that are looking for opportunities to jump into the space. That you know, and like there's some other foundations also doing work on the ground, and I think it's also important to recognize this is something like the more people we get into web 3 in general, the better for the entire space. This is not really just a rush to to get as many people on boarded to MVX. It's. It's really something to to introduce the technology to people who can use the technology and build the next great application Vibe Code, the killer app and obviously, as the Multiverse X Foundation, as X Alliance, this is something that's very, very important to be part of that and to really open up opportunities for local talent, and I think that's really something that the ambassador program will accelerate significantly and really looking forward to the results there. And yeah, so that's sort of the program.
Speaker 1:I think we already talked a little bit about the future development as well. Robert had some spoilers there on some new developments I think in there. I hope they went unnoticed, or maybe it was the async v3. We may have something in terms of eas login functionalities. Maybe one of you guys wants to just give a quick teaser what we can expect there, if there's any new developments that we can tease here, because, again, we've already taken about two hours here of I think that we have teased enough, we tested too much.
Speaker 3:We have to, let's say, do it in one and a half hours and then keep it much more shorter on everything, and that's it. We have one beautiful month ahead of us. In the next month we will have Andromeda live on the mainnet until the momentum to start, and we will see after that how much of the teased things are live and ready to be enjoyed by everybody next time is is and on andromeda, on mainnet, yeah why?
Speaker 1:yeah, definitely great yeah, any closing thoughts here?
Speaker 2:yeah, um, it's very important to to notice that we are doing our best here into doing what we know best to have, and up until now, we believe that we have one of the best technologies out there Life, as we are saying, zero seconds downtime. All what I'm saying is that by the next momentum, we will fight to keep that and also we are doing our best to find the next killer app, if not finding, but building. Definitely, the entire industry is moving from narratives around infrastructure narratives around and Robert did not have enough time to talk about L2s on Ethereum, but I know he would very much love to do that but definitely it's very important to find the real app killer app and the real utility, because that's the moment when everything will just skyrocket, and here I'm just talking from utility, usability and block space usage.
Speaker 1:JOHN MUELLER block space usage. Block space usage. This is like increase bandwidth, reduce latency and also block space usage. Gentlemen, thank you so much for taking the time to give teasers and real updates and everything in between Political thoughts and opinions. Next time, perhaps. Again Thank you so much for tuning in. This was momentum episode one. If you have feedback, very much appreciated. There is a quest running right now where you can make real EGLD by producing content, so encourage you to do that, encourage you to tune in this time next month that is the last Thursday of April, I suppose of AI Pro and, yeah, great to have you all here. Thank you so much for spending the time with us. I think there's a lot of interesting, exciting updates coming up and looking forward to dive into the next updates yeah, next month. Thank you, mihaly, thank you Robert, and thank you everybody for tuning in. Don't forget to do the warp thing. I hope it works and we'll be back for some live testing on Mainnet next month. Cheers everybody.
.jpg)
.jpg)