{ "episodeId": "SLP762", "speakers": { "stephan": { "name": "Stephan Livera", "role": "host", "tag": "STEPHAN" }, "roy_sheinfeld": { "name": "Roy Sheinfeld", "role": "guest", "tag": "ROY" } }, "segments": [ { "speaker": "stephan", "time": "00:00", "start": 0.0, "text": "Hi everyone, welcome back to the Stephan Livera podcast. Rejoining me on the show today is my friend Roy Sheinfeld from Breez. He's the CEO and founder, or co-founder, Breez. just recently put out this new app, I believe it's called, Glow, which I, which I've got here. so Roy's gonna be joining us to chat about that. Roy, welcome back to the show." }, { "speaker": "roy_sheinfeld", "time": "00:19", "start": 19.0, "text": "Hey, Stephan. well, I'm happy to be back. I think it's my fifth or sixth time on the show. So-" }, { "speaker": "stephan", "time": "00:25", "start": 25.0, "text": "Yeah, it's about that. It's been a while, and it's been, it's been a fair few episodes by now. so, yeah, so I guess you've got this new reference app, Glo, which is an Android, iPhone, and I believe you also have a PWA, like a Progressive Web App form of it also. and I b- as I understand, this is sort of like a successor to Mist Give us, give us the, overview from your, from your perspective." }, { "speaker": "roy_sheinfeld", "time": "00:53", "start": 53.0, "text": "Yeah, as you said, we always like to eat our own dog food, and, while we're developing the SDK, every new feature that we add to the SDK, we like to have a way to test it, to test it, but not like, theoretically test it, but actually testing it in the field. so for every SDK that we release, we actually have a companion app that We use for testing to see how the feature works in real life, for the Liquid SDK, back in the day we had Misty Breez, but, recently we were heavily focused on the Spark, SDK and the reference app for the Spark SDK, we call it Glow. I-- we always had a pro-progressive web app, that you, that you mentioned, so you can go to glow-app.co and kind of interact with the app without downloading an app, but in the, in last week we kind of released native apps based on, from a technical standpoint, it's a WebAssembly that we wrap in, native code, using Capacitor for the developers out there. So we wrapped, our WebAssembly in, native functionality and we managed to release it to the store, it required a lot of iterations with both app stores in order to get this app, released. And Specifically, we decided to release it last week because of everything that, went on in Bitcoin. I felt, it's time to give, users, different al-alternatives." }, { "speaker": "stephan", "time": "02:30", "start": 150.0, "text": "Gotcha. Now, I guess let's just quickly talk through the basic features so people get an idea, because, you know, educators are listening, so you can see the app there as an example, but it's got kind of just a simple send, receive flow. It is using Spark in the background. So what, what are we actually holding here? You're holding a leaf? In, inside of Spark, and basically you can send and receive very simply, yeah. Or multiple leaves, right? and then I noticed you've got a very simple like Lightning addressing as well. So I've got Livera at breez.tips as an example, but, it, it's very kind of send, receive, QR code, and I noticed like in the settings, it's not like, it's not like you have like a million settings in there, right? Like, not like in the older days. It's just a very simple kind of send or receive." }, { "speaker": "roy_sheinfeld", "time": "03:14", "start": 194.0, "text": "Yeah, so it's more about the user experience than the user interface, to be honest, for us. we wanted to, to demonstrate the power of the SDK. I think there are-- I'm, I'm very happy that the UI is minimalistic, it, because, all the heavy-- it means that all the heavy, heavy lifting is done on the backend without exposing users to the technical implementation on how things exactly work. So I think if, if there, if the audience out there went through- Through the journey with us, starting from the, the OG Breez where we had node on a mobile device, then, went to do, Liquid with Misty Breez, now Spark with Glow, you can see how the user experience keeps, improving and, and, and, and, and simplifying. I think it's the convergence of multiple features that kind of, provide the superior user experience, and I, and I know I'm not objective But I definitely think it's like f- that from a user's, experience standpoint, it's the best, app, in the market because let's start with onboarding. When you onboard, you onboard through a passkey, and the passkey, onboarding experience is, is super simple. You don't need to write, any seed. The backup is done, automatically through the, passkey vendor, backup system. So you really get a, a very smooth onboarding experience with just clicking on your passkey. and we have a specific protocol where, again, I'm, I'll, I don't know, like, help me with a balance between kind of, the feature from a user experience standpoint and like the technical details of, of what we actually do, like how we leverage the passkey in order to derive a deterministic key that is needed in order to Use, a Bitcoin wallet. So we actually take the passkey, the vendor can be Google, the vendor can be, Apple, the vendor can be third, party vendor, and we, use a function called PRF. to derive a deterministic key from, the, the, the passkey. We also use Nostr in the background for developers who wants to have multiple wallets, so you can have one key with multiple tags. And we also derive seamlessly a Nostr account, and everything is transferred to, to end users. Users don't even know that we're using Nostr in the background to save labels, so you can have multiple labels and have multiple wallets under, one deterministic, key. Gotcha. And then the" }, { "speaker": "stephan", "time": "06:03", "start": 363.0, "text": "backup lives with that user's, let's say, Google Cloud or Apple Cloud. And then the other thing is, depending on your password manager, you might not be able to use-- so as an example, you can't use Bitwarden with it, you have to use like Google or Apple, to do the PRF thing, right?" }, { "speaker": "roy_sheinfeld", "time": "06:21", "start": 381.0, "text": "There are third-party pass, passkey provider that do support, PRF, and PRF again, it's the ability to derive a deterministic key from, from a specific passcode. Bitwarden will support it, as well, but the spec is, the, the implementation of the spec, PRF is part of the spec. The implementation of the spec is something that, the smaller vendors need to catch up, yes." }, { "speaker": "stephan", "time": "06:51", "start": 411.0, "text": "I see. Yeah. And so" }, { "speaker": "roy_sheinfeld", "time": "06:52", "start": 412.0, "text": "in any case, for those who can't use PRF, there is an ability to use the classic, passphrase, so you can always kind of fall back to, to pass, to, to a passkey, to a pass, to a pass, key recovery, a passphrase, and save the mnemonics the way you typically do." }, { "speaker": "stephan", "time": "07:15", "start": 435.0, "text": "So Roy, you were talking about the onboarding process there, and then you were also mentioning how there's auto swaps. So the idea is that you can pay in and receive things even if you don't have that thing, because you're gonna have like a swap provider there. Talk us through a little bit about, you know, how this Swaps are happening." }, { "speaker": "roy_sheinfeld", "time": "07:35", "start": 455.0, "text": "yeah, so again, like what, we're trying to push from a UX standpoint is this concept of a single balance. So you can have a balance either in BTC, or in, USD. And from that balance, you can pay, Lightning payments, Bitcoin payments, and, and also stablecoin payments, USDT or USDC. and the way that we do-- I, I guess you're asking about the, the USDT/USDC payments." }, { "speaker": "stephan", "time": "08:05", "start": 485.0, "text": "Yeah." }, { "speaker": "roy_sheinfeld", "time": "08:05", "start": 485.0, "text": "we're-- Everyone is talking about like bringing Stables to Lightning, and, and, and once Stables are on Lightning, we'll have a different user experience, and everyone will be happy. I think, I think it's, a bit of, overhyped because no one, maybe the stablecoin issuer needs stables on Lightning, but other people don't really need stables on Lightning because they have stables. Stables are readily available in, in, in, in dozens of networks out there. So the approach that we kind of decided to take is to, swap, every-- if you are holding a Bitcoin balance and you want to, make a stablecoin payment Payment, we automatically swap, on the client and then send the asset to the destination network. So for example, you want to send, USDT or USDC on, Solana, let's say, we take, the Bitcoin balance, swap it to the appropriate amount of USDC, and send the USDC out. And from a technical standpoint, we have an architecture of multiple providers, so we use decentralized- Exchange, on the background in order to do these, swaps, everything is, non-custodial. we have, Flashnet as, a main provider. We used to have Boltz until, up until, last week, but we'll replace, we'll add more, more providers. it's very easy to plug into our, to our, multi-vendor architecture, and everything that we do is multi-vendor. So we weren't affected by the, by Boltz shutting down because we had a multi-vendor, architecture already baked into, Glow. and that's, and that's basically how we implement the, the swap. I, I, I need to say that even if, stable coins will be on Lightning, there's always a swap. You need to swap the, the stable coins to BTC using an LSP or, or a swapper node on Lightning. I don't really see an advantage, doing that in a Lightning node versus doing that on the client side. I think, the world of, crypto, already kind of proven the power of swaps, and I wanted to bring that natively to, Bitcoin. Yeah." }, { "speaker": "stephan", "time": "10:38", "start": 638.0, "text": "Yeah, I see. And so I guess your point really is just that we already have swap servers and providers, that's already a thing, and therefore just use those. That's kind of, that's the point you're trying to make, right? Yeah," }, { "speaker": "roy_sheinfeld", "time": "10:48", "start": 648.0, "text": "yeah. Trust. So, so I, I think the, the, I think the-- That's the atomic swapping," }, { "speaker": "stephan", "time": "10:51", "start": 651.0, "text": "yeah, to be clear," }, { "speaker": "roy_sheinfeld", "time": "10:52", "start": 652.0, "text": "yeah. Yes, yes. And, and, and Ever ago, but, but they emerged, and when, once they emerged, I think it's, it's a very nice user experience to be able to, send, in stable coins. Polymarket just added Lightning, support, by the way, but, previously they only accepted, crypto payments, and I use Glow in order kind of to, to place my bets on" }, { "speaker": "stephan", "time": "11:23", "start": 683.0, "text": "Polymarket. Right. Yeah. I see. And so I, the other thing around, swap services. I mean, obvious question with Boltz dot exchange, they said they were undergoing these AI-assisted attacks. Do you think other swap services and providers will also face similar attacks? Like, what's, what does that mean? Is there a paradigm shift here or are they gonna survive but they will just have to spend a lot more time and energy and money on security? Wow, okay," }, { "speaker": "roy_sheinfeld", "time": "11:48", "start": 708.0, "text": "a bit of a bombshell because it's a very complicated topic and an interesting topic. So let's talk about AI first and kind of what's, what's the, what we're undergoing with kind of the AI attack, with AI attacks in the past two weeks, and then kind of we'll, we'll drill down into the swaps in that regard, because I think swaps are also, unique. So there's an inherent asymmetry, between, the ability of, attackers to use AI, and the ability of, the, of, of, of, of providers, vendors to defend themselves using AI. it's much easier and cheaper to attack. Then to defend, because we're all-- we, we all have access to the same AI models, but, think about it from a vendor standpoint, a vendor in order to defend, themselves from attacks, they need to have twenty-four-seven, defense and, and the attack surface, they need to defend on all the attack surface, not necessarily the attack surface that the attacker will attack. So they need to protect the entire surface Stack, they need, and they need to, to protect their, entire stack all the time, and with every change that they're making, and that's very costly, to do. attackers have an advantage because, They kind of know where to attack. They have more context, which is very important for AI. The more context the, you have, the more, productive you are and the more successful you are with attacks. So they can apply the same attack, gain more context, gain more knowledge, and kind of train their AI to be more proficient in, attacking. And they know they don't need to attack all the time. They, they, they need to attack only when they choose to attack. so- So it's getting more and more complicated to defend, in the era of AI. And another thing that caused the asymmetry is of course the fact that, and I'm, I'm not surprised at all that Bitcoin was kind of the low-hanging fruit, because most of the projects are open source, so, so that provides a very, a very easy mechanism for informa-for feeding information to the model, the model can find, the, the attack using scanning. It's easier to do that than kind of, attacking an opaque, vendor. I think it's good for Bitcoin, and we'll talk about swaps, specifically why, we all need to up our game. we all need to, to, to operate in a higher level of, professionalism. And maybe, maybe, we're seeing kind of the end of small, vendors and we'll start to converge, to, to larger operations and larger- Major companies, which, to be honest, I think that's what Bitcoin needs. again, I don't want to bash any team, but like, it's not surprising that, Coldcard was attacked, a small team compared to, it's not just the Bitcoin only thing, it's like the fact that, it's a smaller team, Vault, maybe the most competent team in Bitcoin got attacked, but a small team. So size is a factor. i-is a factor of the resources that, that you have, is a factor of the eyeballs that you have, the factors, it's a factor of how much tokens, you can spend. so maybe we'll see companies start converging and kind of forming a coalition in order to defend themselves against the attack, there are teams that are popping up, are actually one way to form a coalition." }, { "speaker": "stephan", "time": "15:53", "start": 953.0, "text": "so, so let's change now to the swaps aspect of it. Do you think the swap model is doomed? Or it's okay, but just there'll be a different way it works. No, no." }, { "speaker": "roy_sheinfeld", "time": "16:02", "start": 962.0, "text": "I think we need, again, to up our game when it comes to, swaps. Swaps are easy to attack in Bitcoin specifically, by the way, bridges and swappers across the board are being attacked, not only in Bitcoin. Garden was, down, now it's back up, and other swap services were affected. A lot of the attacks are going unpublished, they're keeping that, Under the radar, because, many teams don't want to admit that they were attacked, but attacks are ongoing across the board. I think with Bitcoin, it's very, hard, to protect an atomic swap provider now these days, because they are out there, they're exposed. How are they exposed? They expo-exposed through an open-s-- because they open-source their stack, one. Secondly, and more importantly, they expose themselves Through an API. So in order to do a vault swap, vaults need to, to, to provide me an API, and the API is published and known so I can reach to I can reach vaults and I can start, attacking them only because they have a public, API. I think the, the model, we, we will see a paradigm shift, to more, near intent, kind of crypto model where You will have, the ability to, declare, request swaps, and there will be, Swap providers that are-- will pick up the, the, the, the swap request and process them without public, without the public facing interface. I see." }, { "speaker": "stephan", "time": "17:53", "start": 1073.0, "text": "So you're saying it's like they're gonna go a little more private in terms of just, yes, saying, \"Here's my API, you can hit me with whatever,\" and, and it's gonna be more like somewhat private or somewhat permissioned or somewhat of a model to, to stop the attacks or at least to mitigate some of the attacks?" }, { "speaker": "roy_sheinfeld", "time": "18:08", "start": 1088.0, "text": "You will have marketplaces, decentralized marketplaces. Flashnet, by the way, is one of these marketplaces where it follows the crypto AMM model, where you have Flashnet as a decentralized exch-exchange, and you can provide provision liquidity to Flashnet in order to, in order to process, swaps, in order to kind of process the conversion of one asset to another. I think we'll- We'll see this type of model, proliferate across the Bitcoin ecosystem, meaning the next version of Boltz will be maybe, an order book, marketplace where you just publish your request for a swap, someone is watching the request, grabs the request and process them using, private or public solvers." }, { "speaker": "stephan", "time": "19:06", "start": 1146.0, "text": "also wanted to ask you, I think this is something you posted, you were saying actually a lot of the volume isn't end user wallets, which may be counterintuitive. Can you explain what you were saying there?" }, { "speaker": "roy_sheinfeld", "time": "19:17", "start": 1157.0, "text": "Breez SDK is, is, very interesting because on the one hand, yes, it's a, it's a, it's an infrastructure for building wallets, and we have many wallets and, and application building on the Breez SDK, Cake, Clever, Blink, Prime, Club Orange, Radar, the latest app from, from the Cake team. I can't keep track with the, with the how many apps are building on the Breez SDK. But actually most of the volume is driven by backend services, because the Breez SDK also provides you a Lightning infrastructure without running a Lightning infrastructure, so you don't need today a Lightning node in order to send and receive Lightning payments, so a service like Exolix, which is, is another swap provider that, wants to support, Lightning swaps, are, is using the Breez SDK in order to provide Lightning swaps. And they don't need to run a node, they don't, they don't need to maintain, a Lightning infrastructure, they don't need to secure this infrastructure, they don't, they don't need to spend any money on, IT resources or human resources in order to maintain Maintain, in order to maintain the infrastructure, you don't even need a cloud node like Voltage provide, like, you just need, the, the, the, the, the send and receive, API that we, provide you, and it takes, Very little effort to integrate, and we see a lot of backend services integrating the SDK just because they want instant Bitcoin functionality. so I think that's a revolution, a, a, an unspoken revolution, with, flying under the radar, but, I'm, I'm really excited to see what's going on." }, { "speaker": "stephan", "time": "21:09", "start": 1269.0, "text": "I think it's a good point because ultimately it's maybe a bit counterintuitive compared to, let's say, the early days of Lightning when it was like, you know, Joe Blow with his, you know, umbrell sitting back home and like he's hooked up his Lightning wallet to that and we were kind of thinking of it as like retail adoption, right? But actually it's more like exchanges, custodians, businesses, B2B, the B2B side of it is much bigger than the B2C side." }, { "speaker": "roy_sheinfeld", "time": "21:32", "start": 1292.0, "text": "And yes, of course. In, in the, it's, the, the eighty twenty party applies here as well. Like, I think, the volume is, driven by, by, by server side, adoption, it's, the exchanges, neo banks, on-ramp, off-ramp, and classic swaps and bridges." }, { "speaker": "stephan", "time": "21:56", "start": 1316.0, "text": "Yeah, interesting to see where it's going. but at the same time, I mean, getting, with all this tooling, it's getting easier and easier, right? Like nowadays you have an easy Lightning address, you just, you know, and even things like offline payments, right? Like whereas that used to be such a pain, right? And then nowadays, I think that, that offline receive experience has, has, has dramatically improved, right? It," }, { "speaker": "roy_sheinfeld", "time": "22:18", "start": 1338.0, "text": "it's there. Like you don't need to do anything. Like in the, let's compare Misty Breez to Glow. We still had to process offline payment using mobile notifications. Glow doesn't require mobile notifications in order to offline receive because that's being handled by the Spark, architecture. Gotcha. you can, you can offline receive in Spark, and that's why we using Glow. That's one of the advantages of, of, using Spark, and that provide a superior user experience. So I think all the convergence of all the features, together, test keys, stable coins, sending stable coins, having a stable balance, we haven't really kind of drilled down into having a stable balance, but now with a click of a button, I don't know if you remember kind of the stable Sats days, where you can, you were able to, to, to stabilize your Sats in a dollar value using only, custodial solutions, now you can just click a button and there's a automatic conversion of every payments that Being sent or received with BTC to a stable balance, and you also earn yield on the stable balance. so it's I actually, I, I use my Glow in a, in a USD mode where I can see my, my dollar balance, I earn yield, and when I send the Lightning payments, I can do that like the, the, in the normal way I typically do, just kind of send to a Lightning address or send to a Bolt or a Lightning invoice and every- I think works, single-sig." }, { "speaker": "stephan", "time": "23:57", "start": 1437.0, "text": "Interesting. And so in terms of the, like, just help me understand, make sure I've got the mind understanding correct here. So in Spark, as we said, you, what you are holding is Spark Leaves. So there's these Spark operators, so there are these entities, so Lightspark, Flashnet, and I believe you, Breez, as a company, is also one. And then what's really happening is instead of Bitcoins moving, it's like you're signing the ownership of that coin to, you know, another owner And then you are, part of the trust model is that you're trusting that the server, you know, forgets the old state and you just correctly signed it over to the new. and then inside of, you know, that spark concept, there are the branches, and then the leaves are what we actually hold. And then those leaves are what we, let's say, as the end user, can, if we need to, in the extreme case, we can unilateral exit. Have I got it roughly right? You need to," }, { "speaker": "roy_sheinfeld", "time": "24:50", "start": 1490.0, "text": "yes, yes, yes, yes, definitely, but you need In the unilateral exit, it's not just about exiting, the leaves, it's about also about exiting the branches." }, { "speaker": "stephan", "time": "25:02", "start": 1502.0, "text": "Gotcha. And then that's why it's so well spoken about, like you need a minimum sort of threshold above, to be above that threshold for it to be kind of economically worthwhile exiting. Is that right? that," }, { "speaker": "roy_sheinfeld", "time": "25:13", "start": 1513.0, "text": "kind of, I don't know, depends on the person that said that, because there's a lot of, optimization you can do to, to a tree. the, and the reason that we're Taking our time in, in releasing the unilateral exit, there is unilateral exit, in the, in the base, in the basic Spark SDK, but, what we're working on is a way to unilateral exit also in an economically efficient way. so we want to make sure that unilateral, unilateral, the ability to unilateral exit is, is, is there, always present and, economically efficient. And in order to do that, there's some optimization that needs to happen. on the background in order to kind of converge multiple, so to be clear, the, the app," }, { "speaker": "stephan", "time": "26:00", "start": 1560.0, "text": "yeah, so to be clear, Glow app doesn't have unilateral exit right now, not yet, but in theory, you have the passkey and you could use some kind of method to do a CLI thing if you were like an advanced user, but for now, you can't or what, what is it like? Yeah." }, { "speaker": "roy_sheinfeld", "time": "26:14", "start": 1574.0, "text": "So now we don't, we don't allow you to export the tree state out of Glow. you can use a separate command line, interface Listener, either to, export the, the live state or to unilaterally exit. The features that will, in-- will be integrated into Glow is the kind of one-click unilateral exit when all If all the operators are, offline, but we don't want just to do it for the sake of kind of checking a check box to say, \"Uh, see guys, we have, you're entirely exited, it's on custodial, let's all be happy.\" We want to do that, in a serious way where if, people will have to use it, it will also be very economical to use. there are open PRs, if you go to the, Breez SDK, repository on GitHub, you can see the open PRs. Some of them are very close to being merged. once, once merged, we'll expose that in global." }, { "speaker": "stephan", "time": "27:22", "start": 1642.0, "text": "I see, yeah. and then I guess the other big question for some users, like, I think The thing is, there's no one size fits all, right? So there may be users who are very privacy focused, and then for them, there have been people raising questions about Spark and the privacy aspects of it, whereas others see it as more just like, \"Look, this is just like spent-small spending money amounts,\" but how do you kind of, address the, the privacy side of it at least nowadays with Glow and Spark?" }, { "speaker": "roy_sheinfeld", "time": "27:50", "start": 1670.0, "text": "compared to what? we need to compare it to, to, to other- I guess other L2s," }, { "speaker": "stephan", "time": "27:57", "start": 1677.0, "text": "right? Yeah. Spark, Ark, Liquid, Lightning itself, eCash. I guess those are the main ones that people are gonna compare to, right?" }, { "speaker": "roy_sheinfeld", "time": "28:05", "start": 1685.0, "text": "Yeah, ag-again, like, Lightning in what-- Let's, let's, let's take, the LSP model on Lightning, as an example. Like, when you use Phoenix, what's, private about Phoenix? Like, they know all your transaction and they know your destination. Well," }, { "speaker": "stephan", "time": "28:21", "start": 1701.0, "text": "Asynk knows your bal- no, sorry, Asynk knows your payment, right? and I guess you, if you aren't using the default settings, you're just calling to a public Electrum server, so whatever Well, Async is doing the routing for you, but they, I think they mentioned something about, was it trampoline routing or something they were gonna do, but I don't know if that's in. So anyway, that's not one model. You need, you need, you" }, { "speaker": "roy_sheinfeld", "time": "28:46", "start": 1726.0, "text": "need the, you need the other trampoline nodes to exist in order to kind of to do trampoline. That's right. That's right. Phoenix know that your destination, the Greenlight solution back in the day knew your destination because you were kind of, routing through, Green actually around the payment, they know a lot about you, so I think the Spark, ARC model is, on It's kind of comparable to, to, to an LSP that does the payments, that actually execute the payments on your behalf. the, the thing about Spark, I think, people kind of criticize Spark because once, when it was first released, all the transactions were, were, were available in, in a Spark Explorer, meaning you could have seen the user activity, outside of the Spark operators. And to this date, by the way, shout out to Wallet of Satoshi that have, they have to fix that because they're still running in a public, in a public mode of, Spark. There are vendors that are using Spark in public mode. Now, is that for-- Is there a" }, { "speaker": "stephan", "time": "30:07", "start": 1807.0, "text": "reason for that, or is it just like they could change the setting or what is the reason for that? I" }, { "speaker": "roy_sheinfeld", "time": "30:11", "start": 1811.0, "text": "don't know. I don't know. It's very easy. I, I don't know And on Twitter, once, they didn't, respond. it's very easy, like from a technical standpoint, it's, it's a, it's a, it's a mere configuration in the, in, in, in Spark, where you say to the Spark entity, \"I want to be private,\" and that's what we do out of the box, in the, in our SDK." }, { "speaker": "stephan", "time": "30:34", "start": 1834.0, "text": "In Breez SDK, yeah." }, { "speaker": "roy_sheinfeld", "time": "30:36", "start": 1836.0, "text": "In the, you have a settings, by the way, you can change that as a vendor if you want to. There and explore experience, but they understand they, they don't think about that and they're" }, { "speaker": "stephan", "time": "30:52", "start": 1852.0, "text": "doing that choice intentionally, gotcha. Exactly. let's, let's focus on," }, { "speaker": "roy_sheinfeld", "time": "30:56", "start": 1856.0, "text": "yeah, yeah. But it's very bad, I think, out of the box without warning users to be, to, to run in a spark public mode, because then everyone can see your balance and everyone can see your transactions, like, period. that's very bad." }, { "speaker": "stephan", "time": "31:13", "start": 1873.0, "text": "Well, let's focus on, builders now, because, look, it's time to build. There's AI, there's vibe coding now, people can, o-people can build all kinds of things. I mean, even just like a little while ago, I was playing around with, with various SDKs just to like test it and see how it works. I'm not actually a developer, but, you know, I was able to just get things working basic, in a basic way. Talk to us a bit about the builder experience. What should people see as the opportunities" }, { "speaker": "roy_sheinfeld", "time": "31:38", "start": 1898.0, "text": "Well, the long tail of builders, is, is, it's just incredible. like in the past few months, we see, we saw a huge surge in the number of, of, people building on, on, on top of the SDK. Yes, of course, it's the kind of the, the, the, the, the, the power functionality, the fact that we're adding more and more features, and the fact that we're making it very simple to use. But, there's no doubt the number one driver Is AI, like a, a sim- a simple prompt, and I kind of, urge the, the listeners to, to, to do it, and we even provide the front, through our website, the default prompt to our website, you can have a one shot prompt app, using the Breez SDK, with, with the ability to transfer, value using Bitcoin. It's, it's, mind blowing. And yeah, we see, we see a lot of, apps Being built on top of, the SDK or existing, project integrating the Breez SDK in order to add, Bitcoin functionality. To be honest, there's a lot of slop. it's not like that, that these projects are changing the world. and let's, you still need a professional team, proper designers, proper UX expert in order to do, something good. but the barrier of entry in terms of adding Bitcoin Confidentiality is nothing, like really nothing. if, our marketing guy, you met Danny several times, like he's very, a non-technical user, like every, every other day he sends me a different app, that he built with, Breez SDK, it means that, yeah, sorry Danny, but it means that every, everyone can do it." }, { "speaker": "stephan", "time": "33:32", "start": 2012.0, "text": "Well, I mean, that's the, that's the cool thing, there's opportunities, obviously, that's it, I mean, of course. People will raise questions around, you know, as we said earlier, there are AI attacks too, so you kind of have to, it's, you know, op-op-opportunities, but also, risks, risks. But nevertheless, people can build, people can get out there and try to build things in, and obviously, if we're kind of, you know, we're trying to grow Bitcoin adoption, we want more people to build in Bitcoin payments, Lightning payments, so that's all there. We'll leave it there, just to keep the episode, digest Thanks for joining. Thank you, Stephan. Thank you for having me." } ] }