Momentum Ep.0

❮❙❙❙❯

This episode marked the beginning of the Momentum Podcast: an open, technical discussion for builders on MultiversX. We covered protocol upgrades, the economic and technical implications of the USH stablecoin, growing concerns around memecoins and regulation, and how AI agents and on-chain Warps could shape the future of blockchain interaction.

Hosts: Robert Sasu, Lukas Seel, Mihai Schiopu

Special guests: Beniamin Mincu (CEO @MultiversX), Ahmed Serghini (CEO @HatomProtocol)

Transcript

SPEAKER_00: 0:03

GM, hello, welcome. This is episode zero of Momentum, uh Building on Multiverse X, a new show that we're just doing because we were tired of not doing it. Um, there were a lot of different iterations of this one. Um, we've come to the conclusion that better than waiting um and getting it perfect and getting it right, we're running this as an experiment. Um, so I'm very, very happy uh to be joined here by Robert. Hello, Robert.

SPEAKER_03: 0:34

Hello everyone. It's not GMGM, but uh always it is GM, let's say in the code.

SPEAKER_00: 0:41

It's always GM somewhere. Um and hello, Mihai. What's up?

SPEAKER_01: 0:46

Hey, hi. Yeah, so it's an experiment that we are running running live. That's very important. So it's not pre-recorded or anything, it's it's just live. I'm just curious to see how it will uh it will go.

SPEAKER_00: 0:59

Exactly. And I think like a couple of things before we start, just to set the expectations. This is a podcast uh built for builders. Um, we're gonna go over a bunch of things, uh, different things from like the larger ecosystem, um, Web3 ecosystem, but also specifically on Multiverse X, specifically tech updates, uh, things that have been happening uh over the past month. The plan is for us to do this uh the last Thursday of every month. Um and yeah, we're we're talking about technical updates, we're talking about protocol advancements, we're talking about ecosystem developments, we're talking about ecosystem and community achievements. Um, and this, you know, this is live for a reason. This is going to be interactive. If you have any comments, questions, or something, we might um keep an eye on the chat and and answer some of them. And yeah, that's kind of the idea. The name momentum. I don't know, Robert Mihai. Like I wasn't involved in the creation of that. Who came up with it?

SPEAKER_03: 1:59

I just approved.

SPEAKER_01: 2:01

So you can say more about the name. I I was the one with the idea. I I just thought about something that cannot be stopped. And then I said something that cannot be stopped and has a great value of um cinetic energy, you know. Then I said, okay, let's make it something like so momentum for um the ones who are probably to a la Latin word, and um it actually means that uh it has a continuation. So every third last Thursday on every month, we are trying to keep this and uh bring news and um very interesting stuff for for everyone who is going to join us.

SPEAKER_00: 2:47

And maybe before before we start, like also we have the show kind of again, the the layout is pretty pretty clear. Again, this is an experiment, so it will evolve over time. If you guys watching have any cool ideas for it, we might be able to bring them on uh to the show as early as next month. Um, I'll do a quick introduction of who I am and and pass it to the other guys as well. Um, my name is Lucas. Um I am in like the the chief intern at X Alliance, which is the Multiverse X Community Accelerator, something we started last May, um, something that is supposed to be a little bit of a um in-between sort of nexus between what the foundation is doing and what the community members are doing. We're hosting um a bunch of events all all over the globe. Um we have a bunch of members that help uh other builders um progress in their path. We have um a lot of quests and and earning opportunities. We have our own grand program um to kickstart some of some smaller initiatives, um, some of which we'll mention uh later on in the show as well. Um so that's that's sort of my background, and I'll I'll pass it to Robert next to give a quick hello.

SPEAKER_03: 3:54

Oh, I am the second one. All right. I am Robert. Uh um your face is getting red and more red than red. I don't think if that is a no redder.

SPEAKER_00: 4:05

This is the the new camera will fix this.

SPEAKER_03: 4:07

Yeah, yeah, the new camera will fix that. I know. Uh, but um uh I'm here since 2018 from the Elrond world, and um before that I was mostly working on AI stuff uh for uh uh cars which would drive themselves. And before that, I think I've met Bitcoin back in 2011-2012, where I was keeping a few courses about why Bitcoin will take over the world and will destroy all the banks. That not did not happen yet, but um there is uh still a hope to do it, but not with Bitcoin, but with other much more powerful payment systems. So, yes, and um we will speak not only about tech but only also about what's happening, especially in the tech world. So feel free to ask anything and then we can move to Mihai.

SPEAKER_01: 5:14

Yeah, thanks. I I I just observed that I'm the only one that put uh the first name and the second and the last name. Uh yeah, so I'm Mihai, I'm the engineering manager at Multiverse X. I've been with Multiverse X for a while now. Uh, I joined back when it was called Elrond, which feels, to be honest, like ages ago. Uh the fun fact is that I go way back uh with Robert as well. I believe we are talking almost about 10 years or 11, 12 years. Uh so I I worked uh at the same company with with him, a big corporation. Um, yeah, we had some AI components there. Basically, we were focused on uh uh autonomous driving systems. I was in the software department and and Robert was in uh what we used to call uh it algo department or short for from algorithms. To be honest, I did I didn't even know uh Robert existed uh until one day when uh one of my team members came uh in the in the office and said, Yo, I'm not doing that stupid test. I was like, wait, wait, wait, what, what, calm down? What test? He goes, some guy from algo wants us, wants the entire software department to take this uh evaluation thing. And and I'm I was thinking, okay, what guy? Well, it turns out it was our very own uh Robert Seuss. And that's the very first time when uh he popped up on my on my radar. He was actually working on radar, so he he popped up on my on my radar. And um uh honestly, it's kind of funny looking back now, just because uh he was pushing for those tests because he thought some of the engineers weren't efficient enough, um, and he just wanted to shake things up. So that's probably he he's still shaking uh things up. Now it's it's it's very wild to think about the fact that uh back then uh it was all about squeezing more engineering power um into uh for solving the the problems that we had. And nowadays uh with stuff like uh AI and uh LLMs and uh I don't know, whatever you you want to mention from the AI field, it's a completely different game. Like anyway, 15 or 20 years ago, even though we had those models, um, it would it would probably have been useless uh without the the massive data that uh we've got today to train them. So um, same vibe with engineers. If companies with thousands of engineers, of software engineers, figure out how to leverage the tech right, they'll probably explode in the next few years. And and here I'm I'm saying in the good sense, uh exploding. If not, they'll probably end up cutting um lots of jobs just to stay afloat. Yeah, but probably we'll we'll dig uh deeper uh on that a little bit later. So coming back to Robert and me, um that uh test drama struck me. Um then during the COVID days, I reconnected with an uh an old high school buddy uh that I really really admire. Uh he was at Elrond and kept talking about how awesome Elrond was. And uh he's like, Hey, you gotta talk to Lucian. Um yeah, I did. I talked to Lucian and here I am at Multiverse X now.

SPEAKER_00: 8:36

Yeah, thanks guys. So so we're it's not just gonna be us um going forward, and hopefully today we might have a guest or even two. Um, but uh yeah, it's going to be some guests from the ecosystem. Uh important advancements hopefully can be discussed here uh on both the technical and project end, um, and also some unique perspectives, of course, that come with it. Um so the plan for today is it is February. It is February. I don't know, can you guys pronounce uh that? I I I I haven't quite cracked it.

SPEAKER_01: 9:09

Um it's very hard for me to pronounce it February, but and I know that the idea was to continue it with May, May, May, Mayor merch for the next month. How do you call that?

SPEAKER_00: 9:23

Yeah, I don't know. Genu AI. I have an easier time saying that February. I don't know. Um it's it's difficult. Um, but but so this month um on Multiverse X has all been about um AI and and what uh we're building on that. And there we're we'll have a whole section sort of uh dedicated to it. Um and the rest of the agenda is sort of like we'll start with a bit of a uh broader overview. Robert will leave that section um of like what's going on in the ecosystem at large. Um, then we'll dive more specifically into multiverse X uh tech developments, then we'll have an update on um projects and and ecosystem developments, um, hopefully with a couple of guests, like I said. Uh, and then maybe if we have time, we'll do some foreshadowing of what's uh cooking in the in the coming uh months or in the in the next month. Um but yeah, with that, I think we'll we'll just dive right in and and Robert, I I'll give it to you to maybe uh give us an overview of what's going on in the world out there.

SPEAKER_03: 10:25

Um not necessarily on the world because the word is about Trump tariffs, uh misinformation, deinformation, new age media, legacy media, all kinds of stuff. And uh I actually get on got somebody that I am speaking too much about Colin Georgescu. No, come on. If we have a few IQs and we are part of a democratic society, then we should use a little bit our voice here and there, um, in a nice way. If it's possible, ask a few questions and Twitter, it's a place where you can ask questions, whether the politicians will answer or the journalists will answer or not. That's another type, but at least we try that. So uh, so just for that in that comment. Um, in the crypto world, I think 2025 it did not start as people wanted. In general, it started with an explosion of the meme coins over Trump, Melania, and Libra. I think Libra was the worst of those. Um, Trump and Melania, they went down, but they were not as much sniped or as much extracted as everything else. Uh and um right now, as of today, there is like a new uh meme coin act proposed by the US government, and I think that even Europe will do something similar or even worse. I also said in January that it will bite back pretty damn hard. And it will not bite back those people who are launching necessarily meme coins, it will bite back those startups who want to do something crazily good, crazy good in the ecosystem. And it will be about okay, I'm that that new startup will have to hire a lawyer or probably and will have to fill in 100 other pages in which it demonstrates that it's not a uh uh rock pull or a meme point or stuff like that, as we have with Mika regulations or other stuff in Europe, which actually like it adds to the costs of the startups from the start. So it's not so easy to like. Um, the proposed meme coin act is it's it's actually about that public figures like politicians cannot launch a meme coin, but it was launched by the Democrats, so it's I don't know whether it will go through or not. I would agree that meme coins are not about celebrities. So if you want to make a celebrity coin, then make it, but do the benefits for your followers. Like it happened, I think, with uh SubcarPaths in our case, and then holding those NFTs, you could go into the VIP zones. Um, a lot of fun things can be happening around where where people are following a set of artists, and those artists are dropping stuff. I think Drip on the Drip house is doing a lot of cool stuff with artists and the followers. That's a fun thing. That's that's that's that's something which which has value for the long term. I the what's happening with Trump and Libra and others, I don't even call them meme coins, I would call them Rugpull coins. And not necessarily with Trump and Libra is the biggest problem, the biggest problem is with the overall sentiment and how how did it get here? And it was like a social construction in such a way where theft is somehow allowed, because actually that's theft. If you are sniping and then waiting for your friends to buy later, so to you to dump your tokens on your friends, uh, your own community, your own people whom you are chatting, uh, that's actually theft. And in sense, when you have inside information that something is going to be launched and you are sniping it and then then uh dumping it, that's again theft. But it was like you know, a period of time where this kind of thefts were smaller and they were somewhat accepted, but right now nobody cares. So when the Hawk Two girl like last year launched and then rugged their community, there was a big, big outroar, and that person, uh, the girl uh did not appear for like one month, two months, or stuff like that, because there was such a high out outrage around it. Right now, with Trump, me Melania, or the president of Argentina, nothing, it just whatever, or against those people who were saying that they were like sniping and extracting everything, no outroar, and this will I this will bite back. And and the thing is that blockchain has to do better, blockchain has to and has the ways to make some of these things much, um, let's say eliminate a big portion of these. It's also at the level of the of how the smart contracts are created, and at the level of the front end, and then the level of the socials, because you could have put only real meme coins into this kind of space by ensuring that the tokens are locked, by ensuring in the contracts different kinds of things, by ensuring that there is like no sniping, maybe have a pre-validation time, random buying, uh uh a lot of types of auctioning systems. So you can figure things out. And one of the differences which we did, for example, in multiverse six, at the first launch pad, which was raised, which was right, we have seen how people were aping in, and then how some of a few builders actually sniped and and bought the first batches, then sold after, making millions of dollars in a few minutes. At that moment, we have reacted as a foundation and given back to the users, and after that, we were thinking about how to make this kind of things better. So we created launch pads with uh price discovery, uh, with locking tokens. We introduced the MEV protection. So you can do a bunch of things, you will not resolve all the problems, but you can resolve some of it. And and the one one thing which is let's say the worst is to to put your head in the in the in the sand and then say, no, it's not our fault, it's nobody's fault, uh, it's a free market, everybody can do it. But everybody knows in the background that you are sustaining this kind of criminal activity, so and it doesn't go against uh uh one or two projects, but it's the whole ecosystem in this sense because and and all those calls that you are just one click away from making uh generational wealth or this kind of BSing. Come on, this is not a casino. Like you cannot call meme coins casino because the house wins, so it's worse than casino.

SPEAKER_00: 18:15

Actually, the ride lounge was one of the interesting, more interesting things to me that happened on MBX. I wasn't obviously part of the team in any case, but like it was so interesting to see the foundation actually take direct action um to re-imp like take it upon themselves to to then really like make people whole again. Obviously, it was an imperfect system always and and will be, and and the price discovery mechanism and everything introduced since is is super interesting. Yeah. So yeah, very interesting moments also in in terms of like how the space is progressing, I'd say.

SPEAKER_03: 18:49

Yeah, the space in 2024, it was not necessarily good in this sense, but you know, we have worse than necessarily meme coins, we have more extra extracted by uh hex than meme coins, if you if you if you want to call it. So we have like over over nine years ago, ERC20s were developed and they are still draining the wallets every single day. It's like millions of dollars every single day. And also with the Bybit hack, which is again, and a whole bunch of Ethereum guys, not the core developers, they didn't say anything because they know that it is actually an EVM fault, but all other calls and everybody's saying no, it's not EVM faults, it's it's only safe's fault. It's it's Bybit's fault because they didn't verify the contract, the information. Man, they sign blindly because you cannot really verify it. You have to take out and install a bunch of other tools to verify for yourself and a lot of complications around that. And if Bybit cannot do it with his whole OPSEC and cybersecurity team and everything around it, how do you think you or me or my mom will do it? Nobody will do it, or with the valid rates and everything. So, again, it's like the maddening thing about it is that no system is perfect, and you have to acknowledge that that no system is perfect, and that every single time there is something you can improve. And if there are something which can be improved by at the base layer, at the execution layer, and it doesn't take like re-architecting everything, you can do it. There is no reason, so like at the current moment, 90 something percent of the of the contracts on Ethereum are upgradable with proxy and delegate call, which is ugly, which is a security. If if you read a security audit report, that's the first thing, and it's big red. But everybody must go on because they must go on. Why can't you implement a function of upgrade and then show it uh concretely out there? It doesn't take so much, maybe it takes like let's say two weeks of development, and then a bunch of tests and some audits and like a month if on top at most with all the seven clients they have, and then you can go on and say about this by at every single layer. And right now, after the safe uh safe hack, everybody is saying that uh let's say um that it should be visible to see uh what you are signing. That should be the normal thing that the wallet and the directly the wallet, independent on any other system, gets the information, the transaction data from the front end, the microservice, or whatever, but the wallet interprets it, and the wallet shows the correct information out to the user. So you sign what you see, because it was demonstrated with the safe hack, and we can demonstrate it with thousands of other things that in EVM you are blind signing almost everything.

SPEAKER_01: 22:30

Here I think it's very important, but because this is nuts. I mean, the surprise probably for everyone who is not very well aware of what happened, is that the initial compromise was via social engineering. So uh that's what uh Robert is um uh referring to as um the safe uh safe attack. So it was specifically a phishing attack that targeted the cold wallet, uh, the cold wallet signers. Um I think I I've read an analysis that says that um uh safe wallet developer, not a Bybit employee, uh, was compromised, and that led to the insertion of this malicious uh uh JavaScript piece of code into the safe user interface that um um Robert was mentioning about that was not showing the right uh um string to be to be signed. And this is this code was hosted, actually is hosted basically in an AVS, an S3 bucket. So it's even crazier that this entire hack uh was targeting the Bybit wallet by going through the the entire uh other flow. That's why it's very important for everyone to think twice and to pay attention to what they are clicking and uh what they are accessing. And um, even though we have guardians and uh they save you from most of these kinds of mistakes, there are cases when not even guardians can save you.

SPEAKER_03: 24:01

Yeah, but that's a big difference because uh in how the system is created and how the transaction and the transaction manifest is created in multiverse 6, for 99% of the cases, you will see what function you are calling and with what assets, and that information that does not come from a microservice, that information is deciphered and it comes from the wallet as well, which is running on your machine in a secure enclave. So that's a huge difference there, and that's like something right now. The Ethereum community says that they want to solve it. You had like almost 10 years to solve it right now, yeah. Anyway, so uh the thing is that at the base level we can solve or it can be solved a lot of things, and then protocol any protocol who wants to be the next base layer of the internet, it has to be ultra secure and it has to be so easy to use without these kinds of hacks, and that's where we're aiming for.

SPEAKER_00: 25:13

Yeah, I think like one thing, and and it's this is also kind of sagging us into the next uh segment or or maybe uh uh part, like a lot of the things, and you mentioned Ethereum like not delivering, um, that's not a first, right? Um, but but like something where like at the base level, there's so many things that can just be improved on the protocol level to make things like this much, much less likely. Um, and yeah, so that that's very interesting. I want to um talk a little bit about overall like multiverse X um strategy and and um developments there. And I I hear that we might have a guest uh online here who could join us and maybe elaborate a little bit on that. And um with that, I'm bringing up um Benjamin Minkel. Hi Man.

SPEAKER_04: 26:03

Hello, hello, gents. How's it going?

SPEAKER_00: 26:05

Hello.

SPEAKER_04: 26:06

How are you? Great on top of the mountain, just uh making sure I build some momentum. How are things going?

SPEAKER_00: 26:15

Pretty good. Um, yeah, like uh super super nice to see you, hoping that that you might be able to um give us a bit of an overview of like what February was like um for for you and your perspective, and of course, any news from your recent trips um would be would be very interested in that.

SPEAKER_04: 26:32

Look, um I'm uh really excited to see this happening, uh to see a conversation for builders here um happening, giving them some more translations, some more insights, some more ideas, and a platform, a different platform. Um we have so many platforms now uh where builders, validators, developers, and so on can uh chime in, add feedback, add thoughts. But it's great to see another one that is specifically geared at the technical exploration of things, um where they can come in, they can come with ideas, and that just educates even more people. Because I think this is the most underrated thing. It's through the education and to the elevation of these conversations, simple things that are related to technology, to developments, to um progress, that we're actually building momentum. And I was actually um this morning thinking a bit if we should not um do a swap and actually name this podcast, um, the the full shard podcast and uh the other one momentum. But it's really cool. So um, yeah, really, really good to see this. Um, if I look at February um 1st, really great um experience and momentum being built on the AI side, uh, builders being forced and invited in nudged to look at all the AI tooling that is out there. Uh, why? Because AI will be embedded in everything. Um, and when I say everything, it will literally be everything. From you wake up, you get ideas on what you should eat, uh, you get um better tools with which you can build, you get um ideas on how you can optimize your day. And agents will probably be this interface that you have embedded and you interact whether it's just market insights um or trading or investing, ideally more investing um than trading, and then um of course building. Uh so that's that's really cool. If I come even closer, of course, I've been in the US now for uh a few weeks. I'm already saying that I moved to the US, so that's um that's cool. That's some uh some good news. Uh I'll probably spend uh much more time in the US uh than anywhere else just because this is where things are happening. Um we love to be part of the conversations, to contribute to the meaningful conversations, and to make sure that the technology we've built helps key builders all around the world. So if I'm coming even closer, um Denver is absolutely buzzing with energy builders, um, investors, people that have been with Elrond or MultiverseX before and are now looking for the next steps. Um a lot of people that have been active in the past, interestingly enough, are coming back and sending us messages that they've been seeing this very good and positive buzz around the builders, around the technology, around the upgrades that are happening. And so um maybe one thing is of course, you've probably seen. Yesterday we also met with um Charles. Um we've had some conversations before, but it was good to sit and um build that relationship. Um very cool to actually bring together many of these ecosystems in the US. So we'll gradually meet with almost each of the big meaningful ecosystems to see how we can build some brig bridges, how we can build some uh cool things together. And maybe last thing, um, I was mentioning yesterday as a as a kind of joke that people should maybe um just normalize making it a habit of looking at GitHub also whenever they look at the prices, just because GitHub will be a bit of green. You always see a bit of green prices, you never know what you get. Some days it's good, some days it's uh pretty shitty. Uh, but at least with GitHub, you have some good green in your life and uh you can have some fun. And then um building on that, I I was saying that people should just pivot to becoming builders. But builders do not does not mean that they should not focus on building value, they should not focus on uh thinking how they can improve, let's say, value capture prices and all of that. It's just they should think how they can influence it. Because the next um, let's say, follow-up that was partly joke, partly um a good thought experiment is thinking how we could have in the ecosystem an EGLD strategy, um company or movement that thinks how that could be um taken to institutionals, that thinks how that can be taken to companies on the balance sheets, um, that thinks how that strategy can be built and shared with the investors, the contributors, and so on. But that's that's a different way of looking and approaching things just because you want to be able to influence it. The prices just tell you something either of how the world feels right now, it's very short-sighted. But then if you can find a way to influence that, that's always through building. Um, whether you build some cool tools, some cool projects, some cool products, whether you just drive adoption or build a company through which you take the um token of the technology to wider adoption. Those are all really great um ways of contributing and and I mean building. So yeah, just that was just sharing some some thoughts that that I had in between a few meetings that are happening um right now.

SPEAKER_00: 32:49

Can you tell us how much longer you're you're staying or like um any any kind of strategic outcomes you're looking for there?

SPEAKER_04: 32:55

Um yeah, so we I was joking with Lucian that uh we'll uh we'll of course not go home until something um important is closed. Um just we we need to make sure that things are closed and we're we've made some very good progress um even this week. Um but uh yeah, want to this this is the checkpoint. Like want to have something very meaningful closed because then I know that the other things are just momentum building. So I we want to have that momentum already started. Uh this is um in motion and we have some good checkpoints, but uh I'm very pleasantly surprised how much the builders never went away. Um, the right kind of people have always looked and waited to see the good moves, the moves coming to the US, the moves strategically building um with other ecosystems, and just making sure that we're present. Sometimes um we were actually joking with with someone that was just quoting, I think, play Plato or Socrates, that um half of the battle is won just by being present. So you're there, you're already having some momentum because you're there. So most of the time, no overthinking, just go out. It's um 2025. You can just do things. Don't take it, don't take no for an answer, don't find excuses, just go out and jump in the water.

SPEAKER_00: 34:27

Yeah, that's literally the way we're doing this. This is episode zero for a reason, and like yeah, very, very glad to to have you on. Um, I think we'll we'll probably just let you go and and jump back into the meetings and getting some milestones done. But if you have any uh last thoughts on on next steps or anything you want to share um before you leave us.

SPEAKER_04: 34:46

No, just um definitely um do it. Just build it. No second thoughts, no excuses. Go in because um once you go in the water, you'll discover a different world. It's not that cold, it's not that bad. Even if it's bad, you're already in the water, um, and you'll push through. Uh there is no simple way of going through things. And at the end of the day, builders might um well get accustomed to um the challenges, the difficulty, the eating glass sometimes and and uh staring a bit into the abyss and also drinking some coffee and having some fun while uh while doing it. So um yeah, it's there's a lot of good and positive stuff. Uh if you can reframe um all the negative stuff into some type of lessons, that's that's really good. But uh I I've seen on all fronts builders being becoming more vocal in the ecosystem. So kudos to that. By all means, builders, please write more your thoughts, your ideas, your lessons. Those are very valuable to other people. This is how we build that type of momentum. Um, share it and then any type of contribution, any type of step. Don't try to do 100 steps today. 1%. Do that first step and then share some thoughts. And I'd be very curious to to read that. Um, and um yeah, just do it. You're already doing it, so um kudos for that. Keep uh keep the momentum going, guys. Have a good one.

SPEAKER_00: 36:28

Thank you, Benjamin.

SPEAKER_01: 36:29

Thank you.

SPEAKER_00: 36:30

Chat very soon. Um yeah, I mean, very, very interesting, right? Like maybe we'll we'll take this moment to um to discuss some of the um sort of roadmap points that Benjamin mentioned in his in his paper. I don't know if if Mihay or Robert, you you guys want to take that. Like, we have a bunch of very strategic goals for for this season, let's say, right? Um, probably the most important one being um subsecond finality. Um, but but yeah, what not probably that's the most catch.

SPEAKER_03: 37:00

So it's it's subsecond finality. Uh make it or die, or not necessarily die, but you know, that's how we talk inside that that is such an important part of it. That's everything is, let's say, everything is about sub-second finality.

SPEAKER_01: 37:17

And everything that goes to sub-second finality. So I I've seen that um we are using this uh graphic for representing the progress to subsecond finality, just because we said that the sub-second finality is the north star for let's say protocol or for the network itself. And uh there has been only 2% progress in the last 20 days, and I've seen a comment about that. The thing is that indeed, 2% in 20 days, just because now we are with two upcoming uh updates. So the Andromeda uh release, which is uh in testing, probably we will um open the governance vote in maximum two weeks on that, and also we are going to publish all the details, the technical details about it on agora, and then uh Barnard, which is uh feature complete. Uh, we just have to, you know, we had Barnard first and then Andromeda uh planned initially, and then we switched it just because we wanted to gain some momentum or to build some momentum momentum on that. Um, and then we will have to um merge Andromeda into Barnard, and we will have to retake the entire test test sets and so on. So it will take some time, but uh after those will finish, it's going to be all sub-second. That will mean um full progress and the speed will definitely increase.

SPEAKER_03: 38:48

Anything to add there, Robert, on protocol developments there, uh other the thing is that uh while the Android what what we achieved really well so far is that while the Andromeda was developed, we have developed everything around Barnard. So right now, Andromeda got is in super high quality testing, penetration testing. There are um in developers who are trying to break it, so we are pretty heavily investing in that stuff, and the whole testing team is around that. Uh, but we managed to make a small team which focused on developing the features for Barnard as well. So after Andromeda, Barnard will pretty pretty uh come uh come uh super fast. I have a question around Andromeda and Barnard. The Andromeda Barnard is is getting one shot finality, which means the block when it is finished, uh it is completely final. Uh this is not only uh Andromeda, it is not only like for the fine, not only important for finalities uh standpoint, which is um in some sense it will resolve, it will get like a UX three times better for those who are uh making cross-chain cross-shot calls, which is pretty awesome. You will not have to wait one minute for some confirmation, but around 18 seconds. Uh, but it will actually increase our security to super super high levels. This will mean that Andromeda by design will have every shard and ever and the metachain will have much bigger security guarantees than right now MetaChain has. So it any kind of stalling, any kind of um invalid block will require a full takeover of a shard, so two thirds plus one in order to make it happen. Otherwise, they just can't make anything, um, which is awesome in that sense. And you can look that for uh full shard is 400 nodes plus 400 um 400 active nodes and 400 uh validator uh waiting nodes, while the competition, let's say Sui, Solana and others have between 100 and 200 nodes. So one shard is in that sense more decentralized, more secure, more number of nodes than basically most of the chains. We are the number of chains, we are the second around after Ethereum. And after Andromeda, Andromeda is needed for the supernova release, which will reduce from six seconds to 0.6 seconds, 600 milliseconds if it if it if it sounds cooler. Uh the uh the block times, it will change the consensus model into one, which is consensus will select only the order of the transactions and transactions which will get into processing, and the processing will happen totally in an asynchronous ways. Uh, but this will come with Barnard. Barnard is a super focused on developers, it will bring a lot of new endpoints to the virtual machine through which developers will be able to develop faster contracts, much more complex contracts, improved back transfer functionality, improved multi-transfer ESDT and NFT functionality, improved execute on destination uh uh opcodes, uh new pre-built binaries to get epoch information, round information, and timestamp information. In general, this will be about Barnard. It will be a pretty heavily uh VM and developer-focused update.

SPEAKER_00: 43:30

I just like one more thing, maybe. Um, just drawing the comparison to other ecosystems, you mentioned already Solana, Sui, etc. Is there anything else that people like should understand like how significant this development is? Um, from Bernard, uh, supernova, andromeda uh Andromeda? Um, like what are sort of the main distinctions that we should emphasize here?

SPEAKER_03: 43:54

I think that one of the important things is that we get in the subsecond but real finality. A lot of the people which what they don't say about Sue, Solana, Aptos, and the others is that their finality is not as final necessarily as we call finality. We say finality, those transactions which cannot be rolled back, and that will get into this kind of subsecond uh uh subsecond timeline. Um, and on the other hand, all of this will come up on top of sharding. You cannot design sharding later, you have to start with sharding and add the things on top of it. When we started, six seconds block was let's say a really, really good number. Um, and uh one block finality was a good number compared to what was in 2020, compared to anything was there. But since then we have figured out how to change consensus, how to make improvements, you are synchronizing your drinking.

SPEAKER_00: 45:09

Uh the synchronized finality as well, like yeah, yeah, yeah.

SPEAKER_03: 45:14

And then it was always so we were always looking for a way in which machines to spend most of their time processing transactions. Right now, uh, with the current consensus mechanism, they spend at most one-third of their time in processing, and two-thirds of the time is in passing around messages in the consensus. But as math evolves, as cryptography evolves, as as proofs evolve, we can design systems where you can ensure that the system is secure through proofs. We will name them equivalent proofs in case of Andromeda. Equivalent proofs means that you have executed the block, you have reached to this conclusion, which is equivalent with everybody else, and then you accumulate the accumulate the signatures on top of that. Everybody can accumulate it, and but this in general it took progress. We were the first one, I think, who introduced a very a new version of the BLS signatures when it when we launched in the mainnet, not new version, in a new usage of the BLS signatures. We were the first one to introduce uh verifiable random seed in the blockchain, and uh we will be probably our new consensus model has a lot of new and new things, um, especially this kind of fully asynchronous execution, with the need to only agree on the list and the on the list of the transactions as the order is baked into the protocol with the math protection. So it will it will bring us all the so and the the the thing is that with all the innovations we made around our front ends and everything with the transaction manifest, I was speaking about that with Wasm and everything. Um this will add on top of it, and we will have subsequent finality.

SPEAKER_00: 47:30

Okay, I think one one last thing too, which is uh the sort of low spec machines that um MVX is running on right now. I think it's it's a significant um difference, especially when we're thinking of Solana, where like starter kits uh start at uh 10k or something like that, and maintenance is extremely um high, costly, and and like is that something that will be affected in and in which ways um by these updates yeah uh this this is not going to be affected.

SPEAKER_01: 48:01

So um neither for Andromeda, neither for Barnard, nor for uh supernova. Uh but uh as we I think we mentioned yesterday or the day before yesterday, uh we will probably um need to make improvements after we will finish with supernova for the parallel processing uh upgrade, which will bring um more uh bandwidth, which that will probably bring us uh uh in in top three the most uh uh capable blockchains out there. And um, why is that? I mean, even though when we are saying about improving the low spec requirements, we are not saying like it will become um so expensive that you cannot um pay for them, like it will probably double or triple in cost from point of view and also from performance. And even though uh if we take into consideration uh the evolvement of the hardware, we will probably meet with um mid from the from the cost point of view with mid um costly um hardware for it.

SPEAKER_00: 49:15

What's like a what's a number for that per node or something?

SPEAKER_01: 49:18

So if if if we are looking right now, it depends on the setup that you have, it will probably cost you around$200 for um a low spec machine. And if you want to have something uh higher spec, it will go up to 400 or 500 uh dollars. Uh back then we will expect costs, and here I'm talking about monthly costs. Um not keeping uh on bare metal or keeping the setup uh at home where uh obviously the cost will go even down, you even lower. If obviously you you have a Raspberry Pi for it, it will go even even lower. Or if you we did not test running on a fridge, but uh we've seen in the community some um builders that tested running a node by using a uh Samsung A50 or something like that, which was quite impressive. Uh, and uh we really encourage these kind of approaches just because it demonstrates the beauty of the technology. I mean, the the main principle of blockchain is that anyone should be able to run a node anytime anywhere.

SPEAKER_03: 50:30

I would double down on this and I will say that I am pushing for lower and lower requirements. So let's make everything possible from let's say from the code perspective. Right now, the requirements are around uh quad core machines and eight gigabytes of RAM, but actually you could run with dual core and eight gigabytes of RAM or two four gigabytes of RAM in some cases. I think the A52 machine had has four gigabytes of RAM, if I am uh not mistaken, and it it can run. And with all these new improvements, actually, the the not only the time spent on processing is getting bigger, you have more time to spend on processing, but actually the processing gets much optimized from IO perspective, from processing perspective. And um uh to add that I think that we don't want to go over quad core machines and 60 gigabytes of RAM. So that's like the top in our maximum in with what we think, and that is just like your all not not average laptop, but let's call your average laptop can do that, even a Raspberry Pi Pi 5 can do that. And I would personally recommend everybody to run bare machines because that's like really the beauty of the grip the blockchain space where you run it in your own your own garage. And it's only not only one just one more thing, it's not only about low specification machines, but it's also about uh the bandwidth requirements we need. And right now we don't want to change the bandwidth requirements at all, uh, so it will remain the same. We are we do not want to go in the direction Solana or the others are taking with gigabits per second or 10 gigabits or 100 gigabits per second bandwidth requirement because that's costly, and that actually kills all the garages because you don't have kind of uh uh system in your garage, you are dependent on cloud providers, so or making your own cloud, but that's costly.

SPEAKER_00: 52:54

I want to change the topic a little bit and demonstrate how cool we are. Um, about 10 minutes ago, um, Victor asked this question about um the Haddam stable coin. So we got on the phone um and made sure we talked to the person behind the stable coin. Um, and somehow he answered the call and is on stage with us right now. Um, welcome, welcome, Ahmed. How are you? Hey.

SPEAKER_02: 53:21

I'm doing good. How about you? How are you doing?

SPEAKER_00: 53:24

Yeah, all good. Um very happy to have you. Um thank you for for taking this call on such short notice. Um and just to answer Victor's question here. Um, so I'll I'll just pass on the question uh to you directly. Um, how is how what are how are things going with with this stablecoin we hear so much about?

SPEAKER_02: 53:45

Well, uh I would say that uh I'm super excited for what's coming next. Like we've been working on uh on this specific product for about like a year and a half now, and uh it is aimed to be, you see, like uh one of the biggest launches that we will see like on multiverse X. So we can't wait for uh for Monday to come, actually.

SPEAKER_00: 54:09

What would you say? And like this is maybe just like taking it back a step, right? Like this is the first native stable coin on Multiverse X, it's overcollateralized. Let's get a little bit into let's let's say the details of it. Um, Robert and Mihai, also if you guys want to jump in, but but first, of course, Ahmed, like what's what's the new thing here? Why why do we need this? What does it do? How can people benefit from it?

SPEAKER_02: 54:32

So let's say that uh during our activity like on mainnet, uh through Hattam and the lending protocol, so something that we really enjoyed was the high yields and the high demand on the stablecoin money markets. So whenever some new liquidity was supplied, there has always been multiple borrowers who directly like took it and they were ready to be exposed to some high yields just to be able to get access to that liquidity, basically. So what we decided to do, and like before that, so even if the yields like the supply APYs were like super high, there wasn't a lot of liquidity that has been breached to multiverse x to be supplied there. So we understand that there was like a real need for stablecoin liquidity. And this is how like the idea of USH has come. So what USH will allow you to do is to be able to access this stablecoin liquidity without having to be exposed to this variable rate. So you will be able to mint it either at a 0% interest rate or at a fixed one through the lending protocol and actually allow people to access this liquidity. But I would say that this is not the only benefits or purpose of USH actually. So besides tackling the stablecoin liquidity issue, it is also helping to reduce the selling pressure on e-gold by allowing liquidity pairs to be formed with USH instead of e-gold. And this is the exact same transition that Ethereum went through back in the days. And it will also provide access to stablecoin liquidity by using e-gold as collateral, eliminating like the need to sell it for the different users. So uh yeah, I would say like this is the reason, like the main one behind creating like this new product.

SPEAKER_00: 56:30

I'll I'll take I was wondering if if Robert and and and Mihai had a question.

SPEAKER_03: 56:34

I think like no, I don't have a question. I just uh I wanted to congratulate in the two million EGLD locked in uh I think in Hatam, but uh that brings uh another type of discussion. Like we not necessarily, we don't necessarily want the leado kind of situation with 30 33 percent. So sorry, Ahmed, but uh there will be a need for others to join the system. But you know, a little bit of competition will make uh all the products shine and then them, and after that, you need market uh not market makers, but uh those who are making the markets efficient. So congrats on that. But uh but you know you have to prepare yourself for a little bit of competition because otherwise uh we get into the other aspects of the decentralized world, which which we don't want that of course.

SPEAKER_02: 57:31

So uh no, thank you very much. Uh first, and for us, you see, like competition is always good, and what is like most important for us is for multiversic to thrive. So anything that pushes forward is is a pretty good thing. So uh yeah.

SPEAKER_00: 57:47

I I want to know like one of the things that I was wondering, like in what's in your best case scenarios, right? Like, how do you want users like just regular retail users to interact with the protocol? And then also on like after that, how do you see project um projects benefiting from it? And then like how do you want builders to build on top of it? I guess three very different questions, but maybe like in these in these steps would be interesting to hear.

SPEAKER_02: 58:11

Well, uh, like uh what I would like users to do is to mint ush and uh directly like long equaled with it, okay. Like uh jokes aside.

SPEAKER_03: 58:23

Uh there's the chat that's the strategy. There's the strategy. It's okay, but you know, you need providers as well.

SPEAKER_02: 58:29

So um I would say that the use case of USH like would depend on the specific case of each user. You see, like through Hatham, uh we have like multiple products ranging from Lending protocol, liquid staking, and other cultural stablecoin. So the use cases are different. Uh, there could be like some simple strategies and some complex ones. So it really depends on the user itself. But what we want to achieve actually with this new product is to have like a steady, healthy, and constant growth similar to what they had easy, like since it started. So uh yeah, I would say like this is our objective with the release of uh of this new product.

SPEAKER_01: 59:13

I think um uh again, um, myself, I want to congratulate you, uh Ahmed. Anyway, we we had several discussions, but I think this is uh clear um evidence of the fact that the ecosystem gets more mature and it will get more and more mature uh as protocols like Hatam uh will evolve and will grow. And also, as Robert was mentioning, uh we encourage new builders to come and uh we even support Hatham and we support new protocols just because it will make everything uh more um it will bring opportunities for everyone in all kinds of ways.

SPEAKER_02: 59:54

Totally agree, totally agree, and uh thank you very much for the kind words.

SPEAKER_00: 59:58

I have a last question before. we let you go. Um the launch is planned uh for March 3rd, correct? Exactly. Yes. Take us take us to the war room. What's what's going on? What are the final um preparations? Like how are you uh I don't know. Uh what is it called? Uh preparing for impact, you know, like how how do you see things going down?

SPEAKER_02: 1:00:21

Yeah. So like uh let's say that everyone is uh I would say like a bit stressed but in the good way uh like those matches uh always create like this sentiment a sentiment of excitement too on our side I would say like to give you some some more uh information about like what's going on so we are like working on different fronts like from a tech point of view we already like took care of the deployment and upgrade of the required like smart contracts to make sure that we have like enough time before like Monday. So trying to reduce like the the charge of work. From a functionality like point of view uh what we did is we also like included uh multiple members of the community during like the heavy I would say like last testing of the last version like of the protocols that would be like deployed on mainnet. So we have a group of more than 20 people like that they have been like heavily like testing I would say you see like the latest state for the past week already and like it's still ongoing. And from an economical and financial like point of view the release of USH must come like with stability. And to be able to achieve that like from the beginning what we need is to have like a deep liquidity for USH from day one. So we already have deals I would say and synced with different partners to be able to create like different LPs for USH from the beginning. So those LPs will be live at the moment that USH could be minted. So we already like shared I would say like some more detailed info on our protocol update but this includes like LPs with eGold uh with Max UTK Ash Foxy2 and HTM and uh yeah and also of course like uh USDC and USDT LPs that we paired like with uh with USH so those are like the things on which like we are focusing but we are making like some pretty good progress and everything is set so we like like on uh on the 3rd of March at 4 p.m. I think that you will not be the only one in war rooms I think that we will look with curious eyes for the next big test on the multiverse X ecosystem because um this is something uh which did not happen yet and um with all these new pools and all these new markets I think there will be a lot of uh a lot of uh circular feedbacks and stuff around how the money flows how the smart contracts are interacting with each other how much gas will be used here and there so it's it is pretty okay that we have fixed a few issues uh we had in September I think last year where uh we had discussions with Hattom and their their transactions were not landing in the safe price in their record price mechanism and stuff like that so those are fixed but uh it will not be only Hattom in the war rooms for those uh those the first days let's say after that it will be settled if we don't discover something but um you know uh it will be a bit of fun fun fun days around so I would say like same here same here like everyone will be like in a in a warm world uh you see like uh even on our side just like we are uh a pretty big team actually on Hatton but people or let's say the community like they are seeing me alone and maybe like sometimes we we choose like schedule a call or you see like an AMA where we bring you see like the whole team so so they can see how big uh I would say like the team is and uh besides this I would add that even like on our side we had like a lot of preparation in terms of how the impact of USH like will happen what could be like the interaction that people arbitragers or any like participant could make and what would be like the repercussion on the different like products on the ecosystem.

SPEAKER_01: 1:04:42

So uh yeah like uh dozens of analyses have been made and we can't wait to see this live and even like after the lunch what we aim to do so we already had an initial round of audits that were like performed for USH and we aim to have like uh a second round afterward to really like make sure that uh everything is uh is as secure as possible amazing and this this had a more room idea i'm gonna i'm gonna hold you to that like uh organize like a big call everybody can can like just see how many people you are and like how things are are working maybe um for like once you once you're done with the real world room we can simulate it um in a in a more uh chill environment and maybe all celebrate together for this uh launch again a year and a half in the making uh congratulations for everything um thank you for for building all of the stuff you guys are building um and yeah best of luck i guess for for march 3rd yeah thank you very much guys for having me and uh yeah wishing you a good end of the day so thank you thank you by yeah that's that's super interesting like of course like there's the the had them end but then of you guys must be stressed about all of this stuff um as well to to make sure all the systems are ready yeah uh i don't know like um stress is good of the live it's like your phone is uh we have a few contacts in in i am in do not disturb in general my phone at night but there are a few special people like lucian who get who are whitelisted let's say for a call um and back and forth in that sense uh we had a few war rooms i i think we might uh make a story around that in the next in in the next podcast just to just to show you around what happened in the war room uh it is exciting it is stressful you fear a lot of things but um the things what you can let's say reduce test uh stress is by more and more testing which you do beforehand so and always to be prepared that anything can happen and then to be prepared to be like fully there and resolve resolve the things just one note that I think in the last one of the last war rooms uh it was like 12 o'clock at night or not not 12 o'clock it was 2 a.m and after 10 minutes there were over 20 people in the call and we were in that call for like the whole day long so um I think uh war periods make strong software developers we can say that right because experience during war rooms and war periods and late night calls uh just bring the the need of urgency and that's when real uh teamwork is happening and when uh most innovative and fastest and most optimized solutions come up to to life just to solve the the issues as fast as possible and let's not forget that a very important part in in being on call and being fast reacting are the validators so we have so many validators involved but even though just because they care about what we're building and what they are supporting and what they are validating for um we had I think the the the fastest response time was somewhere around six hours to have more than uh 66% of the network upgraded so that's very important for us to have the validators close we have a uh red uh red line to say so with uh with most of the validators and when something happens we uh directly call them and um yeah we are happy to have uh such great uh validators with us yeah and i it like actually this is also a war room in the sense that the three of us had an agenda we're like we're not even halfway through i think we're an hour 15 in um so we can decide on the spot this is this is episode zero so we can decide like um I think a little bit let's let's go of course through the MBX tech developments um we've already touched on on some of the um ecosystem developments with Haddam but want to spotlight a few more things um but yeah want to uh pass it over to you Mihai to maybe talk a little bit about February I um on multiverse X I did that well this time yeah that that sounded cool yeah well February and also how uh Benyamin uh positioned the entire direction like global autonomous markets uh by talking about an autonomy a level of autonomy um and I think uh he already mentioned once or twice during the his presence here we are talking about and thinking about agents and um we are having a hackathon ongoing right now it's an AI megawave hackathon and one of the um tracks of the hackathon is called AI agents but for everyone who participates and watches uh probably it's very important to to understand that AI agents at their very core are basically some programs and they use the artificial intelligence and here generally speaking uh in one way or another to do some tasks for us they can I don't know answer some questions or they can um handle some support requests if we are talking about customer service um if we are talking about developers they they can assist with coding they can assist with debugging and um if we are looking more and more on the on on Twitter because Twitter became the let's say to go platform for news and for uh hypes we we see that people have mixed feelings um on one way we can see the the the versatility of AI agents so they automate all kinds of routine tasks and they can support on kind all kinds of processes but the public perception of AI agents is mixed some of the people uh around 60 are quite um they they consider ai being helpful and and that they they they feel supported but more than um 39% of them feel that they there there's something more than that they express more concern than excitement about uh about ai and now um to understand why probably and and to understand how those uh ai agents um come into fruition to say so i will not go much into detail probably uh we will ask for feedback after the the episode if people want to go into real details but uh from a very high level point of view we have a few components which are very important for AI in general so is the perception which means that um the AI systems or the agents have to um be able to gather the data we had sensors where when we uh worked before the company we worked before uh now we have APIs and we have text so um actually information and and and uh data input has to come from somewhere uh obviously we have NLPs for or um natural language processors for uh uh taking the the text as input and then analyzing it we have computer image uh processing uh algorithms or computer vision for images uh we have speak recognition for for audio and all kinds of uh of other uh solutions then is the processing part um and here things become interesting because um in us as as developers we we think about uh if else if else also as a program that can be called as an AI agent why just just because it reacts to an input uh and gives you a predetermined output this is actually what an AI agent is so uh it he or it it or however it's able to uh take an action after processing the input that you you you provide it now thinking about uh real intelligence here um we have to uh to say that uh those systems have to be able to learn now uh learning can be of several ways like supervised learning and so on but i i i've read or i i've seen a video uh quite a couple of days ago and and i i think ilya uh soutzkever said that um the first step for a gi is combining unsupervised learning with lots of reinforced reinforcement learning at the at the heart of it and what does it mean it means that for for um thinking about unsupervised learning it means that the system should figure out things uh by its own um without anyone telling it um how to use so it used some clustering algorithms and then a reinforcement learning it means that the agent will learn by trying and uh error so he's chasing rewards like good results now my question for a long time is that wasn't this how um humankind or in general uh nature evolved because this is actually what uh AGI is is doing right now anyway cutting short the story we have uh heard about Eliza OS Eliza OS is um the track uh is the sponsor the AI agent's track sponsor um we have heard about cursor uh probably most of the builders have heard about cursor um then we we are using the mcps which is uh a term uh introduced by anthropic the model context protocol and we are going to hear them more and more just because Eliza offers this opportunity to create uh deploy and manage AI agents uh cursor provides all the developers the ability to make use of those uh agents and AI in general and then we have um mcp the the the this protocol that allows now uh only anthropic or cloud to access databases to make calls to uh api calls so basically to bring data to them now um i intentionally chose those three examples just because i i want to go through something what what uh is happening now in the market and and um what um we will probably observe more and more to happen um in the in in the in the future um i've had a discussion with um one of our colleagues uh just the other uh day and um i trust him a lot he's an exceptional developer he's one of the of the exceptional developers so um somehow he's not fully convinced or he's not fully bought about ai and um just because i trust him a lot but i i i also trust myself um i said okay i have to to search the the twitter the internet i have to deep search everything just because uh i'm more than sure that this is not happening only in in multiverse x but is happening all over the world where developers or senior developers very experienced developers feel like ai will never take our place or will never have uh uh an important uh or at least not in our lifetime an important uh saying and um actually I found this uh curve which uh thank you for sharing Lucas uh I I found this curve which is very very important and I I'll go very fast uh over it just because um it's a lesson that has to be learned by junior engineers who are probably at the university right now and they are just starting uh an LLM and they are using it just because it's cool. You know they are putting all the pieces together they they hit an error they chase just copy the error paste it to the LLM and that's it the LLM solves the the the problem gives you an explanation you need to make an update to the library that's fine um the LLM can help you. Well this is a trap because if you're just copying and pasting code from it without understanding how and why it works that way you are not growing as an engineer. So you are not growing you are not understanding the systems behind of it. And back then when I started um programming and also probably Robert we had to figure things out the hard way I mean taking the book opening reading blogs trying you could actually do stack overflow and copy paste but after that it it might have worked or not. Yeah but you you would have had to understand the problem. So um it's important for uh everyone to understand that LLMs can help but they should not be a replacement for uh you searching learning and understanding what you are actually doing there. So they can speed you up but you don't um you don't you don't have to let them think for you that that's that's a problem. And then we go to the mid um mid experienced engineers to say so um those engineers know their their way around they can make coding faster with LLMs they use copilot um it autocompletes a lot and and all stuff all kind of uh advantages but the problem is when um uh during midnight Lucian calls you and says um hey we have a problem or um you have to trace uh a race condition in the debugger or uh you have to to fix um I don't know a problem reported by one of the developers in uh or builders in in telegram those are still on you as mid uh engineer you you still have to to figure out what was the problem if you were the junior engineer who did not learn you cannot support anyone at this point in time um then we go to senior engineers and here I believe it's the the the point where I wanted to to touch senior engineers actually have the full picture so senior engineers probably wrote half if not more than half of the code base themselves they feel like um coding uh faster with an LLM but um they don't feel like the the LLM is actually uh doing the the same level of of quality when uh developing these kind of uh engineers are defining roadmaps more uh they are um hunting down extreme complex bugs and and they are working a lot on on fighting and fixing though those bugs and they um draft all kinds of design documents and here it's a here is a uh um heads up for Robert because I've seen how happy he is from time to time when he has some time to write some some code just because rest of the time it's just um yeah doing all this kind of of stuff now at multiverse x our systems are quite complex uh and um llms don't have that deep insight yet probably they will have I don't know in two or three years just because on average uh LLMs at this time um at this point in time have a capacity of around 200 000 tokens um which is not that much it means something like small project small to mid projects i i would say as full uh context input but uh probably this will change but even though um for for seniors I would say that LLMs are cool for experimenting asking some questions but when the real challenges hit um we we have to count on on on the senior developers and very fast on on staff engineers where I I consider myself that's why I'm optimistic and I feel like uh the LLMs and also uh AI agents and um AI in general will bring a lot of uh benefits we we see it from a different uh point of view we we see a different story here so our jobs is more to find the way our our job is to find new ideas to experiment very fast and LLMs are great for this uh we can do proof of concepts very fast um proof of concept that took probably weeks before now it takes only two or three days and I I'll just give you one example just the other days I had a request from our countic team that they they they need a better visibility and and and way of exporting data um about I don't know efforts in engineering teams and so on. They have all kinds of strange conditions and questions around how the data should be combined and um how how should they how should they uh configure and automize it and make it fast. Well this would have been impossible for them to do using Excel or whatever whatever tools are there. So then I thought about an app and I just did it in two days and now it's it's working and it it helps a lot. Otherwise it would have taken quite a few of days because we have a back end we have authentication we have um um all kinds of services there is a database we have a front end for it right and this is what um I would say it's very important for all the hackathon participants right now because we said for the hackathon we need MVPs we need proof of concepts for which we will offer support afterwards with developing on the other hand just because we have the other hand as well is the growth games or the growth grants for growth grants we are not looking at proof of concepts we are looking at real projects that can be real utility and and which can support themselves afterwards um just to to to also cover that uh on short and uh to provide a little update on that we have until now more than 70 applications on um on growth games uh out of which we accepted only eight uh but we are in discussions with uh 20 of them just because we we've seen the potential uh behind the idea and the rest were just um projects that would probably would would fail just because they do not bring the utility and this is um in general for startups if you solve a problem you will definitely have bigger chances of success on the long term i want to a little bit challenge you here and then uh challenge everybody who is listening if you are coming to a hackathon and you want to prove a point with your project you can you even if we are in a crypto world you have to respond to a few questions and think about how you are going to make money when you are going to be self-sufficient uh what's the timeline to get into the breakeven zone and those kind of questions are probably the first one if you are not in web 3 in crypto because in web 3 people are saying okay we launch a token and then we sustain from there and the token goes to zero and and it's a meme coin and you can call it a meme coin but yeah so let's focus on that because um those are some of the things and AI can help you on that to to to create some of the ideas um I wanted to make a joke that you consider better than senior developers as a staff but you know the staff are in general making uh clean uh cleanups uh sorry for being blunt on that um I think that right now in the people are not necessarily realizing it but I think that the Groc 3 moment right now or the O3 Pro model right now is as big of a step as GPT 3.5 was when it launched and um I think that's such an amazing like an exponential stop step from there.

SPEAKER_03: 1:26:39

At the current moment if you go down the rabbit hole and you start to understand how to discuss actually and how to prompt these AIs wow the the it is also scary and it's also amazing in both of the times it helps a lot on research I think that no developer should fade it anything in any kind of PR pull request you make or any kind of few lines of code you write you can actually make the LLM the first reviewer and it might find optimizations or rethinking of the architecture and stuff like that.

SPEAKER_01: 1:27:24

Yeah you have to apply after that and you need the knowledge to be able to do this kind of stuff but let's not fade these these are like amazing tools I don't know what will happen in the Singularity it is definitely pretty damn right now it's not the answer to put all the power in the AI's hand because they would go they they will do their own language but you know they will probably do their own language and protocols and and so on and um and and this is a discussion for multi-agent systems which are uh trending more and more uh and by the way as far as I've seen I think they are going to announce chat gpt 4.5 4.5 today um as far as I as far as I got yeah so it's it's quite interesting um for all hackathon participants they still have a chance to use chat gpt 4.5 just because uh we will the the submissions will end on the 3rd of March uh afterwards we will have two days of uh evaluating the the applications and on the 6th of March we are planning to host a demo day where we will invite uh top nine projects that will have been evaluated by the by the judges to come and um and present in in in front of them we have incubators accelerators we have um judges from all kinds of companies with different experience so they can uh help you on um providing uh very good feedback and uh um acknowledgeable feedback um to your to your project and afterwards on the 7th of of uh march we in we are uh aiming to announce the winners so basically this is the the the schedule that we have for for the hackathon and then let's also quickly mention growth games are going on you mentioned 70 applications eight chosen so far it's a 1.5 million dollar uh growth program uh grants um in uh i think in three different categories that is uh build it um accelerate and launch it um so to different categories depending on where you're at with the project um the first build category even has like just call to for um proposals so if you

SPEAKER_00: 1:30:00

Want to uh build something that um the chain and the network needs, there's even clear instructions what to build. Um, and you can dive right in. I think that's that's a great overview. I wanted just to touch on like because Robert, you said there's this watershed moment with with Grok. Um what does that mean? Like, is it is it it has a different way of understanding code that is uh you know basically replaces a lot of human labor, or what's what's this sort of like new moment?

SPEAKER_03: 1:30:28

I mean Mihay too, I'm not sure who wants to answer that, but I think that the first models like 3.5 were good on summarizing things, but uh with uh O3 or Groke, it's not only about summarizing from one thing, but to gather information, especially Grog, to gather information from various sources and to bring it together in some sort of format. And when you start to ask him to reason about it, then it's pretty amazing. Uh after that, you can um in the first versions like 3.5, when you were making a code review on on asking him to do the code review, or like we tried, I don't know if it was 3.5 or 4. Um we just uh run on the whole on our whole architecture or on small pages or everything, and it was giving giving really like you know those general cases, what you know from any kind of book. It didn't give any kind of specifics, deep things which can be improved. These the new ones, like three, actually gets into the nuances, and some of those nuances you are not seeing, and that's amazing.

SPEAKER_00: 1:31:58

Yeah, I mean super interesting. I think so. Uh just wrapping up this sort of like multiverse X tech developments, we'll we'll dive maybe a little bit more into some other developments that have to do with more the ecosystem at at large, but uh tons of opportunities right now to get started through grants. There's an ongoing hackathon you have until March 3rd, also, yes? Uh Monday.

SPEAKER_01: 1:32:21

Same day with Hatton launch. Exactly. So USH.

SPEAKER_00: 1:32:26

Try out the new um uh AI models and and build an MVP and submit it. Um if there's anything that that we can help uh with, there's also some X Alliance initiatives. Um uh keep an eye keep an eye out for that. Um we'll probably do some uh office hours or something to get you across the finish line if you need any help. Um but yeah, follow follow that page, X Alliance um on Twitter and and uh stay up to date and also join all of the telegrams, of course, there. Um in the last section, and I'll I'll go back to our presentation mode here. Um I wanted to kind of showcase a few things that are going on or have been going on in February February. I tried, I failed. Um, and and one of them is the development of the War protocol. So the War Protocol is something that um we launched at the very end of December, if I'm not mistaken. Um basically an interesting way of bringing on-chain actions um to the web. And um, you know, I'll I'll let uh a more technical person explain sort of like what the idea is with warps and how they work and why, you know, we're calling them warps, as in like uh bringing this uh warp speed to blockchain actions. And maybe Mihai, you can give us a little bit of an overview what struck you perhaps as interesting about what warps are and then like kind of how they work on the on a technical level as well.

SPEAKER_01: 1:33:58

Yeah, I just uh I think very shortly I touched upon the um MCP before, and then coming here to the warps that works, which works works, it's it's a very nice combination. Um it it makes sense just because you were mentioning on-chain actions. This is this is actually and this is generally available. An on-chain action usually or most of the times is a transaction. So uh what actually happens when we are talking about words? Uh they encode transaction data into a URL. Why? Why URL? Probably you've heard about blings from from Solana before, it's something similar to that. Just because this data has to be parsed by the the wallet, and in our case, we have an integration with the extension wallet. So, for example, if you uh for example, you can use the extension wallet when you navigate through Twitter or through Facebook or whatever you you are using, and the extension is able to introduce some minimal uh code like CSS and HTML and and so to the page so that you are able to see a button or two, or however you design this uh functionality. And um when it finds a uh uh key code or a keyword or something in in the article or in the page, uh it's actually where um an action happens. This is a button. If for example, you you are scrolling Twitter uh threads and you see um staking provider saying something, and then a button with um uh prompt for hey, do you want to stake to this staking provider? Um or whatever you want to swap um USH for USDC just because you have seen an offer or something. Um the user clicks the button that appears on the on the web page, the wallet the wallet decodes those parameters, such as I don't know, if there's a smart contract involved in it, uh, or a function that has to be called from a built-in function or whatever, and um it presents this transaction for user to approve. So if you have the extend we usually have on the right or on the left, I don't know, um you would move the transaction or to sign the transaction. As I was saying, this process is similar to uh what uh uh was uh launched before with with Solana. The um beauty of it and what uh I think uh Michael wanted to to uh uh achieve with bringing warps into the AI agents world is that uh they should take go over um so you should not use them only for paying, voting, swapping, minting, or whatever. Um they should also make interactions much easier for agents for developers of agents to be uh introduced. Um so it's it's also uh the perfect combination for web 2, bringing um web 2 users into web 3, and it's also the perfect combination for bringing um AI agents closer to accessing databases or to uh modifying databases, accessing all kinds of uh APIs and um making them actionable because as I as I was mentioning, it's an AI agent when it can do an action, and warps uh help AI agents to do actions.

SPEAKER_00: 1:37:38

And here we have two of them uh uh actioning um each other through warps. And uh I'll mention two things on top of this. So so one, super amazing to see, and shout out Michael who who built this protocol. It's open source, it's community uh-driven initiative that really enhances the capabilities. We now, as Mihai mentioned, we have a sort of like deep ecosystem integration through something that MultiverseX um built or the foundation built, which is the DeFi wallet, meaning that literally when you see an on-chain action on Twitter or something, you can just click it and uh execute the transaction right there and then uh without ever having to go through the usual things that you have to go through when interacting with the blockchain. So that's um yeah, a huge thing, I think. Like developers that aren't the foundation obviously pushing the network um much further than it could go um alone. And yeah, shout out to everybody building such things. Um we even took it a step further, as Mihai just mentioned. Uh, these warps are also very, very interesting for um AI agents. And because we have this AI hackathon going on, there is even a track that is a warp track with, I believe,$6,000 in prizes that um, if you integrate warps in a meaningful way, um you should run in that track. Um, and yeah, this is uh sponsored by X Alliance and a very, very cool initiative from sort of like this prototype that is uh developed by a contributor to the network to integrating it uh into the uh wallet and then to even have a track for it. I think um super impressive. And yeah, congratulations there. Um next uh up, I wanted to bring up uh a very cool growth uh um project that has been doing great work, just kind of a shout out um for all of the stuff that uh CoWat has been doing in the past month. Um they have been voted as one of the top 10 emerging charging networks in France. Um, if you're not familiar with uh CoWat, it is um an EV uh charging station that uh also deeply integrates um Web3 through membership models, through token um tokenization of the energy you can put into your car. Um they have been expanding very, very aggressively, as you can see on this graphic from uh 1320 to 60 stations to now, I think, opening basically a new one every week at least. Um, and so that is accelerating greatly. Um, yeah, shout out to what they're doing. Um, they also just announced that this year is going to be the first year they they'll reach breakeven point before the end of the year, um, which is also huge news, no longer depending on fundraising, you know, becoming profitable within, I think, three years of launching in a very, very competitive market in Europe is very, very impressive. Um, have also announced um bringing more value to the token, uh, the co-aut token um by burning the supplies. So, yeah, super, super big shout out um to co-watt for what they're doing. Um, and another project um launching or having just launched on uh multiverse X is Pell. And Pell is very, very interesting. And I'll let Robert talk about this a little bit because Pell finally brings um Bitcoin very, very close to not just our hearts but our network through restaking. And um, maybe Robert, you can kind of take it. Why is this sort of um innovation something that is needed? Um, how can we interact with it? Um, what's what's going on and what can we expect from from Pell sort of as from a from a network perspective here?

SPEAKER_03: 1:41:33

Sorry, I think I had to snooze.

SPEAKER_00: 1:41:36

Um yes, not muted. This is live.

SPEAKER_03: 1:41:42

All right, I wanted to mute, but it didn't work it out. Sorry for that. Um I I thought it went away, but it didn't. So uh Pair Network, they are from Asia. Uh they have built this on the chain Bitcoin risk taking protocol, which means that they deploy multiple contracts in multiple ecosystems, and there, if you have Bitcoin, you can wrap Bitcoin of various versions. There, you can uh use that Bitcoin in a restaking module, and then projects can tap into that re-staking module to have much bigger security of their own network. Uh economic security, I am speaking about here about economic economic security, and on the other hand, they will be the ones who will launch the gravity ristaking protocol as well for eGLD. Uh, meaning eGLD restaking for them, and in case of their EGLD ristaking, they will integrate with I think right now we have three or four liquid staking modules. If I am not mistaken, it is the Hatom, the SOXNO version, and the and the Salsa version. If if I'm not mistaken, those are the three right now working. Um, all of them will be integrated. Um when they are everything is, let's say, uh open sourced, audited, and everything around it in order to make it happen. Um, in our case, it makes pretty easy to integrate any kind of liquid staking protocol because everything is native, so you have the native assets, and that's why the composability is much, much, much, much bigger. And then uh the idea with gravity ristaking and or Pell ristaking of Bitcoin and of EGRD is that projects can launch new sovereign chains or light speed chains. Uh, we have to uh we have to figure out figure it out the new name. It's temporary. Yeah, yeah, yeah. We have a few more days to figure it out. Uh we'll be able to tap into the risk taked tokens, and then they don't have to make such a high inflation and uh initially um you can support as a supporter if you have EGLD, you take a few a risk on them a little bit, but the only risk is is your your EGLD cannot be uh taken away. The only risk is if you let's say re-stake your EGLD to a wrong validator or a wrong project, which gets slashed, and this the idea of the slashing it is anything which is slashed is distributed towards those who have restate so it doesn't get into the bad hands. What Pell is making it even more uh more interesting in terms of liquidity is that they are bringing Bitcoin into the play. You know, bitcoin it might it is an asset class, but it cannot be used for so many things, whatever the Bitcoin maxims and the salary is saying. Actually, even sailor is saying that Bitcoin is not even money because you cannot lend it on chain. And everybody knows that you can there is no lending and borrowing protocol directly on the on the Bitcoin layer, and it will never be. So you have to wrap it somewhere and use it. So, but you believe in Bitcoin, it offers economic security, you can use it to secure some of the operations economically. This doesn't need to this doesn't need to have to be a completely new chain, it can be an Oracle service, it can be whatever you want in that sense. It it can be even a sequencer, and then you know the sequencer is posting optimistic rollups and if something or ZK rollups, but if somebody catches that sequencer pushing wrong data, then it can be slashed. So those kinds of things can happen. And when you have a protocol with a deep enough liquidity into which people and new projects can tap into it, it will happen. So that's that's the word they are opening up.

SPEAKER_00: 1:46:48

Is there any specific um ways that that you know builders, developers should interact with Pelos specifically, or is this something right directly to them?

SPEAKER_03: 1:46:57

They are really good developers and they will help on anything.

SPEAKER_00: 1:47:01

Amazing. Um, okay. I mean, by the way, one thing that you mentioned that isn't in our highlight reel here is um XOXNO or Zuxno, um also with their liquid staking module, um open source, uh came out this month, I believe, as well, or lending borrowing is also coming. So that's something I think to to also give a shout out to. Um even, you know, like a lot of things launching, I think a lot of things uh happening. Um and you know, we're about I think an hour behind on schedule. We thought maybe we can get this done in in 60 minutes. Um clearly we cannot. Um, so this is something to to think about for the for the next um one. And this is also some um a place where you guys uh we're super happy to get some feedback on whether this felt um too long, where you like what kind of other sections you might want to um see, any sort of feedback that you have that can uh help us improve this for the next month is super appreciated and super welcome. Um maybe, maybe we'll have some some interesting new things we were talking going into the show about having some very cool like gamified uh developer uh sort of challenges in parallel running to the show. Um hopefully we can start introducing them. Um, if not next show, then the one after. Um should we exactly one to two weeks uh maximum on that? Um what how do you guys feel? What um yeah, any any closing thoughts?

SPEAKER_01: 1:48:30

Uh I'm just looking that we have uh quite a couple of things on the agenda, so we we were quite optimistic about it. Yeah, uh, but probably we will keep them for the next uh for the next um podcast.

SPEAKER_03: 1:48:44

No, I think that the the crypto world is changing so fast that our agenda will change completely, so we don't know what we will talk about next time.

SPEAKER_01: 1:48:54

Yeah, uh I'm quite curious to receive some feedback and and to see um how uh the community wants to position this uh this show. Um we can continue the same format. We will bring um the same as we mentioned in the beginning. We will have guests, we will try to make it probably more dynamic, um, making some uh demos probably whenever we have uh some uh MVPs or proof of the this was the demo, actually. Exactly. This was a demo, correctly. Um, and probably we can bring some hackathon winners and and we will see. I don't know, we will make it very um we will try to make it very um inspiring and also helpful for for developers and builders.

SPEAKER_03: 1:49:39

Also, we need some cool idea to make uh this, let's say, not on-chain, but some kind of activity on-chain while we are doing this kind of talking. So if anyone has some ideas, just drop us some message and maybe or let's say wasn't one of you guys supposed to build uh an on-chain thing?

SPEAKER_00: 1:49:58

I totally forgot about that.

SPEAKER_01: 1:50:00

I think it was it was me. We we wanted we wanted to to vote something or to do something on-chain. I don't know. I think I know this is the different the other thing or for the hackathon. Sorry, sorry. Uh I'm confusing the the things.

SPEAKER_00: 1:50:16

Well, that this should be the last um not non-warped or not on-chain um uh action uh momentum, but this is why it's episode zero. We haven't even started, we're we're very early. Um day one. This is this is day zero. Um and and we'll we'll improve over time. Yeah, any any closing thoughts on the market new developments you're looking forward to in March. Um, anything you guys want to say, and then we'll wrap it up. Uh Robert, first.

SPEAKER_03: 1:50:46

We have a lot to develop, so I I I hope that we will give a big chunk of updates on that and what we are doing there. Um, we have a pretty big mission to to do it this year. Or or I I think for only for until end of Q2, we have a pretty, pretty big insane mission. So let's deliver on that.

SPEAKER_01: 1:51:12

Yeah, and and March, um in theory, is the month when we will when the community will first touch and feel the Andromeda on testnet. Uh, we will encourage not only validators but everyone to come and participate in testing it. Uh, then we will have new, we hope lots of new uh projects coming with with the hackathon and also with the growth games is actually um starting to happen, and um applicants will start delivering on the first milestones. So um, as I was mentioning, lots of growth is going to be visible starting with uh with this year, February, March. And um this does nothing but builds momentum.

SPEAKER_00: 1:52:05

Well, that's that's the way to close this uh first episode of Momentum. Well, the zeroth episode of Momum. Um, thank you, Robert. Thank you, Mihai, for taking the time. Thank you, everybody, for tuning in. Um, you will see us maybe in this constellation. Maybe half of us got fired in the meantime for doing such a poor job uh on this. So we'll see what the constellation is next month. But um, we are we are coming back in some formation around this time next month. Thank you so much for tuning in. Um, again, we'll see you. It's time to build all of that. Um, and yeah, be well, do some stuff and do it today.

SPEAKER_01: 1:52:44

Talk some. Thank you.

SPEAKER_00: 1:52:46

Bye. Bye guys.