In this episode, Stephan Livera discusses with Jay & Matt the evolution of Lygos Finance, a company formed from the acquisition of Atomic Finance, focusing on decentralized lending using Discreet Log Contracts (DLCs). The conversation explores the growth of the Bitcoin collateralized lending market, the unique position of Lygos in offering non-custodial loans, and the role of Oracles in determining loan outcomes. The hosts delve into the flexible loan terms and competitive interest rates offered by Lygos, as well as the platform’s global reach and future developments in user experience and funding mechanisms.
Takeaways:
🔸The merger of Atomic Finance and Lygos and its significance for DLC lending
🔸How DLCs work in a lending context: simplified signatures and outcomes
🔸Advantages of DLCs over traditional custodial lending platforms
🔸The role of oracles: Magnolia as a third-party verifier
🔸Speed improvements with adapter signatures and upcoming hardware wallet support
🔸Market size and growth of Bitcoin collateralized loans (over $25 billion)
🔸Comparison of DLCs versus custodial and multi-sig lending solutions
🔸Extending and rolling over DLC loans seamlessly
🔸UX considerations: transparency, privacy, and future seamless fiat/stablecoin onboarding
🔸Potential impacts of future Bitcoin upgrades like Taproot or Covenant support
🔸How DLCs support global, trust-minimized, and scalable lending
Timestamps:
(00:00) – Intro
(00:40) – From Atomic Finance to Lygos Finance
(04:02) – What is the size of the Bitcoin lending market?
(05:33) – Unique position of Lygos in the DLC space; Requirements for DLCs by hardware wallets
(08:20) – What is a DLC?; How does DLC work in the Bitcoin loan context?
(12:49) – How is Lygos different from other Bitcoin lending platforms?
(16:44) – What is the role of an oracle in Lygos?
(20:53) – What does taking a loan with Lygos look like?; Loan terms, interest rates & collateral requirements
(32:10) – Global reach of Lygos; How are the loans funded?
(35:33) – Is UX a major factor in developing Lygos?
(40:02) – Possibility of future Bitcoin upgrades and their impact on DLCs
(44:48) – Closing thoughts
Links:
Stephan Livera links:
- Follow me on X: @stephanlivera
- Subscribe to the podcast
- Subscribe to Substack
Transcript:
Stephan Livera (00:01)
Hi everyone and welcome back to Stephan Livera podcast. Today I’m joined by some of the team from LYGOS Finance, Jay and Matt. Now, long time listeners, you might know I spoke previously with Matt back in the Atomic Finance days and now ⁓ what’s happened is ⁓ LYGOS has rolled in or has acquired, I believe, Atomic Finance. So we’re to be chatting a little bit about DLC lending. ⁓ first off, welcome to the show guys and yeah, maybe you want to start with just a little bit of how
you know, the merger happened or the acquisition happened between Atomic Finance and LYGOS.
Matt (00:36)
Yeah, absolutely. Great to be back, Stephan. ⁓ You know, as your listeners might know, we were previously working on Atomic Finance, which allowed for folks to be able to enter into strategies or ⁓ options contracts using DLCs. I think one of the things that we learned through that whole process is that, ⁓ you know, derivatives on Bitcoin are great, but it’s really, really challenging to build on Bitcoin, ⁓ on native Bitcoin in a way that’s easy for folks, especially traders,
to be able to get the tools that they need in an on-chain environment. ⁓ And we were able to grow it to a really good size, but we felt that we weren’t really able to expand it and scale it to the level that we really wanted to. Around this time, ⁓ a good friend of mine, ⁓ Francis, was actually reaching out to me, because he was really interested in the tech that we had built. So DLCs allow for smart contracts
directly on native Bitcoin itself. ⁓ And he thought that the idea of using this type of technology for loans is something that could be really, really exciting. you know, ⁓ you know, him and Jay, who’s joining us today as well, they built up a great lending book at Anchorage back in the day. And we’re trying to figure out, how can we actually make this non custodial? And so it was kind of the perfect marriage of, you know, we brought the tech, they brought the industry,
experience ⁓ and that’s you know was the the genesis of lygos.
Stephan Livera (02:13)
Excellent and Jay do you want to tell us a little bit from your side?
Jay Patel (02:13)
Yeah.
Yeah, think, you know, Matt definitely covered the backstory perfectly, which was just that, you know, the tech that they had built building at Atomic was perfectly suited for lending purposes. ⁓ You know, terms of mine and Francis’s backstory, we had built up the lending business at Anchorage, ⁓ you know, ⁓ prior to the FTX collapse. And, you know, we had built it in a kind of like ⁓ conservative and, you know, risk-minded approach, but obviously…
post FTX, there was kind of like a freeze in lending markets overall. And what we saw on the back end of that was basically like the institutional lending market. Folks got a lot smarter about just handing over their Bitcoin to their lender or their counterparty. They started using things like triparty agreements or holding it kind of an escrow at a custodian. But for everyday Bitcoiners, the options were still a little bit limited in terms of if you wanted to get liquidity against your Bitcoin, you still kind of had to give up a decent amount of control or just hand over the keys entirely to the lender.
⁓ and so it seemed like a perfect opportunity to basically combine the kind of battle-tested tech that Atomic had built with the lending business model, which is much more suited to this kind of like bilateral relationship than something like options trading. You know, I think lending folks do like to have a relationship with their lender, know who’s on the other side. You know, if you want to do collaborative movements of the assets or modify the loan terms, you can still do that in DLC, but you still benefit from all the trust minimized architecture that.
⁓ know, Atomic was building from the beginning.
Stephan Livera (03:46)
Got it. And so just so people can kind of place this, there’s, I mean, there’s a lot of different lending products and services out there. And I guess maybe just talk a little bit about the size of this market, because as I understand, like every person I talk to in the Bitcoin lending world is telling me that like, you know, it’s, the numbers have gone up a lot over the last call it 12 to 18 months ish, something like that. Do you guys want to talk to that, like the size of the Bitcoin collateralized lending market?
Jay Patel (04:13)
Yeah, I’d say, ⁓ you know, overall the market’s grown significantly. You know, if you think you purely Bitcoin collateralized, I think it’d be excess of 25 to 30 billion at this point. Obviously it’s a little murky because there’s platforms that don’t fully disclose their balances or their loan book or, you know, even what kind of collateral requirements they require. But there’s a significant amount of collateral, you know, well in excess of 20 or 30 billion dollars that’s being used to borrow.
primarily US dollars against today. ⁓ And that’s only growing over time. I think that’s independent of all the stuff you see on exchanges. Like this is purely focused on people who are looking for the liquidity against their Bitcoin. And it’s generally not people seeking excessive leverage. ⁓ I wouldn’t try to conflate the Bitcoin lending market with folks doing like margin loans on exchanges and things of that sort. I think that’s very different.
Stephan Livera (05:06)
Yeah, gotcha. And so the other maybe interesting point, as I understand is you guys are, to my knowledge, you guys tell me the last standing DLC lending product or one of the last standing because of just kind of how things evolved in this space. Do you want to comment a little bit on that?
Matt (05:25)
Yeah, I I think I remember when I started in DLCs back in the day, there was lots of folks, there was us, was ShirdBit, there was Crypto Garage, all building DLCs. Lava joined the scene shortly after. ⁓ And I think what we learned through the process was that…
You there’s all these really interesting applications that you can build with DLCs, but if we really want those other applications like derivatives, we’re going to need covenants. DLCs today on Bitcoin are perfectly suited for loans. ⁓ And obviously there is, you know, think Lava is an example of a company that did do DLCs very recently and made the decision to go custodial, which I think was, ⁓ you know, I guess I was very sad to see that that’s kind of the direction that they went. We still believe very
much that non-custodial is the way to go that that’s the best way to protect like our users and to protect our lenders ⁓
But and I think the other thing too is I think we’ve we figured out a very unique way to be able to allow for DLCs to scale even more and for it to be used by even more people. One of the challenges with DLCs in the past was it was only available on a you know on a on an application on an app like an iOS app or is only available on desktop. ⁓ But for the first time very soon it’s coming to hardware wallets it’s coming to custodians and so the ability for us
to scale I think is maybe something that was missing for other DLC companies.
Stephan Livera (06:57)
interesting. So what change that allows it now on hardware wallets?
Matt (07:00)
So it was always, ⁓ you know what’s funny? If you look at the actual libraries or code associated with many of the hardware wallets, know, Ledger, Trezor, et cetera. ⁓
The requirement for DLCs comes down to a very specific type of signature. It’s called an adapter signature and it’s what’s used to allow for the, you know, encryption and decryption of, you know, the unique signatures that are used that allows for the Oracle to unlock one of the pathways for the DLC. And the, ⁓ the code for that is actually inside of the libraries. We just have to build applications around it that makes it possible to use. So for example, Ledger has been working
on Vanadium recently, Vanadium OS, which is a Rust based operating system where you can actually build apps on top of it. So really like the code was there all along, someone just needed to come and build the framework around it to be able to access it.
Stephan Livera (08:03)
Got it. Okay, and just, guess, high level of view just for people who are new with like DLCs, like I guess I’ll give kind of an ADIQ understanding and you can kind of elaborate a bit. So the idea is if I’m entering into a DLC contract with, let’s say I’m entering into a contract with you, it’s like we are kind of pre-agreeing certain conditions and we’re going in like signing all these different states, but only one of them will be valid at the end of that contract based on what the Oracle puts out. And the Oracle can be like a different…
you know, party who is attesting to, let’s say, the price of Bitcoin as of one year from now or whatever. And then that makes the new, like, kind of, almost like closing the lightning channel. That’s like the way that you end out the contract to either send the stats back to me or the stats back to you. And that’s sort of loosely how I understand that. Can you just elaborate a bit on how the DLC part works in a loan context?
Matt (08:58)
Yeah, exactly. ⁓ So for a loan context, you’re presigning what the possible outcomes are going to be. So for example, you know, for a loan, what’s going to happen to the collateral? Well, either you’re going to repay the loan and you’re going to get that collateral back.
⁓ You’re going to be liquidated by the price going down too far such that the lender needs to repossess that collateral. Or ⁓ upon expiry of the loan or upon maturity of the loan, the lender is also going to repossess that collateral in order to cover the cost of the debt, to cover the loan that was given out. And so if you think about it, these are three very simple outcomes. ⁓
what folks had done and we did this at Atomic as well is, you know, there’s kind of two versions of DLCs. There’s numeric ones where you try to create like this very complicated payout curve and you create all of these signatures in the background to represent every possible outcome that you could possibly imagine. And so we took it a step back and we said, why don’t we simplify this?
⁓ Why don’t we make it easier for the user to verify, number one, and why don’t we make it easier for ourselves ⁓ by instead of having this complicated payout curve, we’ll just have those three outcomes, repaid, liquidated by maturity, liquidated by price, ⁓ and then allow for, you know, in the DLC, both parties, the lender and the borrower, they basically presign for those possible outcomes ahead of time.
And we just keep it super simple, which makes it really easy for us to do integrations and also makes it really easy for the user to be able to verify those outcomes as well.
Stephan Livera (10:42)
I see. instead of like the the atomic finance model where you were signing like I don’t know thousands of states or maybe more now It’s more like you’re pre signing three things. Is that roughly right?
Matt (10:54)
Exactly. And then it’s really easy for the user to just double check. Okay. In this specific scenario, where is my Bitcoin going? And that makes it even easier to, be able to verify, okay, well what is the Oracle doing? Is the Oracle acting correctly or are they not? And, and so when it, and the, the other thing with the DLC as well is typically what you do is you, you know, deposit your, collateral that you’re to use for the contract, you presign the outcomes, and then you also create a refund transaction as well.
So in the case that we completely disappear or the oracle completely disappear There’s still a pathway for you to be able to get your Bitcoin
Stephan Livera (11:29)
Right. This is the break glass
kind of emergency sort of thing. So I guess in practice it’s four signatures. And I guess that’s also a bit faster on the signing side of things because I guess historically it took a long time for the setup phase of this stuff because it was like literally signing all these transactions.
Matt (11:47)
Yeah, well you probably remember, Stephan, like trying the Atomic app out and it would take like one or two minutes to be able to enter the contract. Now it takes like five seconds. And it’s gonna be the same thing on hardware devices and with custodians as well. know, adapter signatures are very computationally intensive, but when you only have three of them, it’s not a big deal.
Stephan Livera (12:10)
interesting. Yeah, okay. That’s cool. So that’s an interesting innovation then because as I read you then it’s like it’s a lot faster than it used to be and now there’s a possibility of having hardware wallets, which makes it more interesting because I think especially for larger loan amounts, you might not be comfortable like walk around with the private keys to that much Bitcoin just like on your phone on like an internet connected device. Now you can kind of have it out on a hardware device or hardware wallet, etc. So that definitely
is more appealing. ⁓ And then I guess, how are you guys different from, like say some of the, like, if I think about what’s already out there, right, like some of the well-known people like, Ledin is out there, Strike is out there, ⁓ there’s a few others out there like Arch and so on, and they’re kind of in the custodial lending sort of side of things, and then in the, you know, you either hold part of a key or, you know, some form of that, obviously there’s you guys, there is, ⁓
Matt (12:48)
you
Stephan Livera (13:09)
Debefy, there’s Firefish, there’s, ⁓ you know, there’s probably some others out there and maybe like, or maybe the Lendersat guys as well. So how does LYGOS compare with some of these other ⁓ types of lending?
Jay Patel (13:25)
I think that that categorization is pretty good in terms of like there’s the existing custodial lenders, which we all know. And then there is like this crop of non-custodial lenders. I think the differentiation between us and the custodial lenders is obviously that ⁓ as Matt alluded to, the Bitcoin can only move under predefined circumstances. So you don’t have to worry about any sort of financial failure of the lender or operational failure of the lender. You’re not relying on them securing your assets or you know.
their cybersecurity practices or maybe rogue ops person at the lender or anything like that, or something worse and ⁓ potentially non-malicious, which is one thing that we’ve thought about, which is that if you are a borrower at a custodial platform, if there’s some regulatory action against the lender, your assets might get frozen completely independent of you or the lender doing anything wrong. We’ve seen many examples of the SEC.
And other regulatory agencies and other countries basically saying, Hey, we don’t know what’s going on here. We don’t like it. You’re not allowed to do anything. And then all of a sudden you, as the borrower, you know, might end up in a tough situation. I’ll say on the non-custodial side, when you think of DLCs versus a two of three multi-sig, and Matt can definitely get more into the, you know, the 150 IQ, very technical part of it. But for, for the lay person, I think the, the, the, the interesting nuance here is basically in a two of three multi-sig, that third key.
Is the most important thing, right? That, that arbiter in the middle, who’s basically deciding, Hey, does the collateral go back to the borrower? Does it go to the lender? What are we signing for? And I think in general, there’s two kinds of compromises there. One is that the lender generally has a better relationship with that third party than you as the borrower. you know, they’re, you know, in a contract with the lending platform, right? So obviously, you know, there’s a tendency for them to side with the platform and you know, the
Oracle per se who’s actually saying, you know when things occur is generally going to side with the lender ⁓ In those kind of setups the other piece is obviously on on the privacy side So the benefit for DLCs is they’re basically like a two of two multi-sig So the Oracle is not like a third-party arbiter. They don’t know that Jay and Matt are the parties to this loan They don’t know that these UTXOs were the ones that were locked for this DLC. They don’t even know the contract address
or the payout addresses or any of that stuff. They just know that they need to leak a certain secret if certain events happen. So the risk of collusion with the Oracle is a lot lower. Your stance with the Oracle is kind of on equal footing as the lender versus like a two of three setup. And I think the last thing that’s really big is like the privacy piece where, you know, you can be comfortable knowing it’s non-custodial, but there’s not some third party that knows all the knowledge of who the borrowers and lenders are, all their addresses, you know, you know, all of that.
Stephan Livera (16:15)
Yeah, that’s an interesting point as well around the privacy aspect of it. So I guess it’s sort of now it’s not a totally free thing. Like obviously, you still have to understand what is the trade off or the trust in the Oracle. ⁓ So maybe you guys can tell us a little bit about the Oracle in your model. Who is the Oracle? What’s their role there? You know, tell us a little bit about the Oracle.
Matt (16:37)
Yeah, absolutely. So we use a completely third party Oracle. ⁓ They’re called Magnolia. ⁓ They’re registered. They basically are a complete third party to us where they make decisions ⁓ on their own about what the outcomes of the loan should be. ⁓ And that’s the current setup for us. So when the user is going and locking their funds into a DLC, ⁓ they’re taking that announcement.
you know, the Oracle is basically saying, I’m going to attest to whether it’s going to be a repaid, whether it’s going to be liquidated by price or whether it’s going to be liquidated by maturity. I’m going to go and independently verify that that information is correct and true. ⁓ And then when that event occurs, they obviously go and attest and they create that Schnorr signature that then allows for the user to go and, you know, unlock their collateral. ⁓ Now, one of the unique things in the DLC model is that it’s possible to ⁓
very easily expand. right now this is a very nice model because with the Oracle setup, ⁓ that Oracle doesn’t need to know Johnny over here is the borrower and… ⁓
XYZ person is the lender. Like they don’t need to know any of that information. All they need to know is, hey, this is the information I need to attest to and I’m going to do so at the time up. ⁓ And it’s very easy to go and verify that they actually attested properly to those predefined outcomes. ⁓
If you compare this to an arbiter system, you know, at the time of the event occurring, now that arbiter software needs to go and say, Hey, what actually occurred here? And Hey, let me make a, let me make a decision on the fly of what we’re going to do in regards to this, this loan, which, know, kind of, kind of makes the process of being a judge and jury, you know, much more, ⁓ much more flaky at the time of, of when that event actually needs to occur. But the other thing as well, that’s a big advantage for DLC.
is it’s much easier to expand to a multi oracle system than it is to a multi arbiter system. Because if you can think about it, like with the arbiter system, if you take a two of three, you’re using a PSBT, right? So, you you’ve got a two of three, maybe you want to move to a three of five, where you have the borrower, the lender, and you have three arbiters. Well, now those three arbiters need to go and actually communicate with each other in order to pass around that PSBT, that partially signed Bitcoin transaction, in order to actually get that
that transaction broadcasted. Whereas in this case, the three oracles can act completely independently. They can go and broadcast that signature ⁓ when the DLC needs to be executed. And then the borrower or lender can go and take that signature, and then they can go and unlock the funds from the DLC. So this makes it very easy to scale to two or three oracle system or three or five in the future, ⁓ which is what we plan, what we’re planning to do very soon.
Stephan Livera (19:39)
Interesting so I guess
it kind of depends on like if this old DLC model picks up and we start to see more oracles come online and it becomes like a thing then maybe this multi oracle model kind of ⁓ Becomes more feasible more popular more desired as an option ⁓ Interesting to see so yeah, so I guess yeah so I guess the answer then is it differs from the cust obviously the custodial lenders is kind of a bit more of a clear trade-off of like hey you actually get to still hold the keys, but
let’s say the outcome of your coins is locked to certain pathways depending on repayment or maturity or ⁓ liquidation on the price if Bitcoin crashes dramatically. But of course, the idea is you should be over-collateralized to prevent that. ⁓ And then the difference with some of let’s say mostly the multi-sig ⁓ lending ⁓ platforms is…
Matt (20:20)
you
Stephan Livera (20:35)
this difference of an arbiter versus having an oracle with programmatic outcomes is kind of how I’m understanding that. ⁓ Okay, and then talk to me about, because I’m thinking like a common use case people might have is like they might need to roll over or extend the loan. How does that work in a DLC context? let’s make an example. Let’s say I take a loan out from you guys. I’m your customer. I take out a one year loan and let’s say
Matt (20:45)
you
Stephan Livera (21:05)
It’s now the 11th month of that loan and I’m like, hey, Jay and Matt, can I extend for another year? How do we do this?
Matt (21:12)
Yeah, great question. So DLCs are very versatile because at the end of the day, it is just a two of two multi-sig with obviously these off-chain transactions that are signed in relation to the Oracle. So at any time, if you are sitting there and you’re like, Hey, I’m almost at the end of my loan. I want to go roll it over. ⁓ We have a whole process for that. We’re basically you co-sign ⁓ with the lender key to basically be able to go and move your ⁓ funds from the
old DLC to a new DLC and then you just re-sign for those same outcomes but you know you know one year later or six months later or whatever and we also use this the same process for also allowing for you to be able to you know add collateral or remove collateral it’s simply a process of taking the old DLC moving it from the old DLC to a new DLC with new outcomes and now you’ve got a new contract where you’ve got those new parameters in place
Stephan Livera (22:13)
Got it, okay, so yeah, so basically it’s similar to even just, as you said, adding collateral or even rolling over or even early ending, I guess, if I want, it was like, all right, I want to end my loan now and repay now or whatever. Okay. ⁓ Yep.
Matt (22:24)
you
Yeah, and this is very like
those that know lightning. This is very similar to a ⁓ splice in or a splice out transaction. Like on chain, it looks almost the same, which is really cool because you get that privacy benefit of this really looking like a lightning channel ⁓ on chain. So people don’t even know necessarily that this is a loan. Maybe they just think it’s a really big lightning channel, which is kind of fun.
Stephan Livera (22:49)
Gotcha. I should also ask just obviously the obvious kind of questions like the terms of the loans that you offer, like what are the interest rates? What is the length of the term or the tenure of the loan ⁓ and collateral as well? Like what are the collateral and like what’s the collateral requirements? So maybe just talk us through the basics, like what are the of the DLC loans that you offer? What are the rates, the loan term and the collateral?
Jay Patel (23:12)
Yeah, so ⁓ with regards to term, we can go anywhere from three months out to 18 or even 24 months, depending on the loan size. So fairly flexible in terms of the loan term for the ⁓ collateral requirements. ⁓ I mean, suffice to say Bitcoin only collateral, but in terms of the LTV, ⁓ you know, we limit borrowers to an initial LTV of 70. ⁓
and a liquidation threshold of 90 LTV. we give borrowers a decent amount of room for topping up collateral. And we also have a process to notify borrowers that, your loan is approaching liquidation. ⁓ Obviously, it’s up to them whether or not they want to add collateral or play it by the market and see how it goes. ⁓ On the interest rate side, I think we have kind of industry leading rates right now. We start as low as 10 % depending on loan size. ⁓
And you know, that that’s applicable regardless of the LTV, as long as it fits within those parameters. So, you know, I think the big thing for us is like, it’s not just a trait, you know, you’re not making like a, I’m going to pay more for this non-custodial trust, minimize architecture. think the longer term thesis for us is actually, you can probably pay less and still get all these benefits because unlike a big custodial lender, we don’t have a massive operations team. don’t have a massive, you know,
financial audit that we have to do and proof of reserves and pay custodians and collateral managers and you know, there’s a lot of costs that are removed because instead of trusting in kind of ⁓ Institutions and audits you’re just trusting and Bitcoin and cryptography and therefore we can kind of pass on a lot of those cast cost savings to borrowers Which is beneficial for both borrowers and lenders, you know, there’s less operational cost in the middle ⁓ and you know the security
you know, most of our bars are comfortable relying on the security of Bitcoin already. So it should, you know, it’s a given that, ⁓ you know, it’s advantageous compared to trusting in someone else’s audit.
Stephan Livera (25:15)
Interesting. the rates, like you’re saying they’re about, was it about 10 % or so? Or what’s the rough range?
Jay Patel (25:20)
Yeah,
yeah, roughly between 10 to 11%, depending on the loan size, the loan term, and the, ⁓ you know, basically the duration you’re trying to lock in the rate for.
Stephan Livera (25:32)
Okay, so
as in if you are doing a bigger loan and a longer loan ⁓ you get a cheaper rate or a more expensive rate or how does that work?
Jay Patel (25:40)
Generally,
bigger loans are slightly lower rates. ⁓ The length, I’d say, actually, you know, like the, and this probably in more credit semantics, but the term structure of Bitcoin, like the rates don’t actually go that much higher depending on the term. And I think that’s basically ⁓ because the industry, think, you know, most people accept that over time rates will continue to come down. You know, as we get more lenders in the space, as people get more comfortable with the role of Bitcoin as a collateral asset, you see more and more
capital providers who basically have a lower and lower cost of capital entering. And so we do think that, you know, it’ll be, you know, that 10 to 11 % range for now, but in a year’s time, it’ll be lower. And in a year beyond that, it will be even lower. So, you know, we’re going to continue to work to bring on lenders to pass those savings onto borrowers.
Stephan Livera (26:27)
Oh interesting, yeah, because I guess I might have thought at the start, oh maybe this 10 % is like a entry offer kind of thing to entice people, but it sounds more like you guys are saying, no actually it’s gonna be like 10 or 11 % and coming down over time.
Jay Patel (26:45)
Yeah, yeah, for sure. I think, you know, if you think about Bitcoin as collateral, a 50 LTV Bitcoin loan is maybe the lowest risk thing you can do with your dollars as a lender. If you compare it to all of the other kind of bonds or other yielding instruments out there where people take massive credit risk or, you know, no collateral ⁓ in exchange for much lower rates. So I think
Matt (26:47)
you
Jay Patel (27:08)
As people get more comfortable with the idea of Bitcoin as collateral, that rate is just going to come down. We’re at 10 % right now, but we have ⁓ plans in the works to bring on more more lenders who can lend at lower rates. ⁓ And the other benefit for DLCs versus something like a custodial platform or even a multisig is we can bring in a larger pool of lenders because in a custodial platform,
The end lender who’s providing the capital, which is generally not the platform. You know, you think of your custodial lenders, they don’t have billions of dollars to lend. have lenders on the other side providing that they have to be comfortable with. Okay. Where’s the collateral held? Is it held at a custodian in a certain country? You know, what is the process around this for us? The lender can always choose to repossess the collateral to their chosen custody, you know, whatever they like, whatever they’re comfortable with. ⁓ and as Matt alluded to, as we do more and more integrations,
You know, we basically allow a lender anywhere in the world to connect with a borrower anywhere in the world and have no trust involved or, you know, minimize trust compared to something like, you know, the world that Francis and I came from, which was like Tri-Party and institutional custody where, know, if you’re a lender in Dubai and a borrower in the U S like I might not want my collateral at a bank in the U S if I’m a borrower in Dubai, you know, like even if that’s reducing the trust required. So we solve for a lot of those things as well.
Stephan Livera (28:33)
Got it. so the, guess on the, okay, so yeah, like we, while we’re here on the topic of the lending side of this house, I guess one thing is they have to get technologically comfortable. That might be like kind of one hurdle for them to be like, well, DLC, like, this is like a new thing. Have you experienced much of that? Or are you seeing that actually you’re on the lending side of the house, people are sort of comfortable with the concept of DLC and placing
know some trust into that technology.
Jay Patel (29:06)
Yeah, I mean, I’ll definitely defer to Matt on how easy it is to underwrite DLCs, but our experience has been that it’s actually not nearly as difficult to get lenders comfortable because of how simple DLCs are. You know, we can actually let them verify every single set of messages that signed and see, know, these are the different, spending paths and these are the Oracle attestations that will be made. It’s not like a smart contract on Ethereum where you have to do some massive audit.
And it’s not like a custodial platform where you have to, know, underwrite our ops team and our, you know, finances and, know, all of these other complicated factors. It’s actually quite simple to audit the DLCs.
Stephan Livera (29:46)
And I presume you’re looking, because you are in a way, I guess, I’m understanding you as a technology also provider here. And so you’re looking for people on both sides of the house, right? You’re looking for borrowers and also you’re looking for lenders to kind of put up some fiat to, ⁓ you know, to, because they want to earn fiat yield, right? Is that, is that correct? Like you’re looking for people on both sides of the house, right?
Jay Patel (30:07)
Yeah, yeah, exactly. Whether you’re a lender or a borrower, we’re interested in having you on the platform. Obviously, borrowers, the size of loan can come much lower. For each lender, there might be 10, 20, 50 borrowers. But generally, for lenders, we look for larger lenders who have stickier capital. I think the biggest thing for us is making sure that their lenders will be around for the long term.
Like we want to make sure that the borrowers on the platform have a good experience. You know, if you’re taking out a loan, our experience in general is that most folks take out a loan and they’re looking to roll it over forever because they think, you know, I think rightly so, the Bitcoin kegger will be more than the 10 % going to 5 % interest that they’re going to pay. And so we don’t want lenders who are basically like, you know, looking to lend for six months, but then going to leave the platform. That’s probably the one limitation.
Stephan Livera (30:59)
got it yeah because you
want some continuity there obviously so that that borrower can roll over when he wants to if he wants to ⁓ and then can you give us an idea on the loan size like what’s your minimum loan what’s your maximum loan
Jay Patel (31:13)
Yeah, so I’ll say the loan sizes are generally, ⁓ you know, ⁓ 100,000 plus, we can go low as 25,000, but I’d say our average loan on the platform is somewhere between, you know, 150,000 and $250,000 right now. ⁓ I think as we open up more of these integrations with kind of institutional custody solutions, you know, we’ll be able to support much larger loans as well. ⁓ But, you know, we can, you know,
Matt (31:22)
you
Jay Patel (31:41)
The range of borrowers is pretty vast right now. It’s anywhere from individuals looking to purchase real estate, buy a house, or Bitcoiners looking to fund their business. It really varies a lot. ⁓ And as long as you have Bitcoin collateral that you can post and you’re comfortable with the LTV, I think we can definitely service you.
Stephan Livera (32:02)
In terms of regions, where are you taking customers? that’s both customer, the borrower side of the house and the lending side of the house. Like is this US only or is it just global or what are the countries that you are operating in there?
Jay Patel (32:16)
Yeah. So it is a fully global platform. ⁓ You know, we support borrowers and lenders and most any country that’s not, that’s not sanctioned by the U S ⁓ but in terms of the lender side, you know, I think that varies based on the lender size. You know, right now the lenders are primarily based in, you know, the U S and Europe ⁓ but we are looking to make like a more, more global push.
You know, on the lender side, it’s just someone looking for fiat or stable coin yields. So it’s pretty simple. But yeah, on the borrower side, you know, we’re, we’re, we’re not a custodial platform, so we’re not taking possession of your funds. you know, we kind of cut out a lot of the ⁓ difficulty that other platforms have in terms of expanding to other regions.
Stephan Livera (33:01)
see. ⁓ In terms of the loan ⁓ funding the loan so from a quick look on your side it looks like you have stablecoin support there are you doing ⁓ and are you also going to have fiat wire support so just talk to us a little bit about how the loans are funded and you know what the mechanisms are there.
Jay Patel (33:21)
Yeah, so on the funding side and maybe I’ll just take the fiat piece and Matt can take the stable coin side, but we are looking to add fiat support in the near future. ⁓ But currently the loans are only funded in the form of stable coins and that’s for pretty good reason that I feel like Matt will be able to do better justice to than I will.
Matt (33:22)
Yeah, what?
Yeah, and I think one of the really nice setups here is that because we’re using DLCs, ⁓ it becomes very, very easy to be able to add support for many new stable coins. We have USDC today, but if we want to add USDT support in the future, ⁓ that’s very easy. One of the considerations is that stable coins themselves are very easy for an oracle to attest to. It’s very easy to look on whether you’re looking on a blockchain like…
Stephan Livera (34:08)
on like random shit chains and be like, this Tron
tether payment or this whatever Ethereum USDC, like you can see that the payment was made or not, right?
Matt (34:16)
Yeah, like even though you know, all these shit coin chains are at risk of ⁓ reorgs or hard forks, you know, it is easy to verify that a know, a transaction was set. Okay, you guys did that well. ⁓
Stephan Livera (34:25)
Ha
Matt (34:31)
On the other hand, like the fiat world is still very old. ⁓ it’s difficult to, you know, as an Oracle, you know, most of these banks don’t even have API APIs. So how am I going and actually verifying that a, you know, a particular transaction has been sent? basically have to rely on the party that is actually, you know, sending that wire at the end of the day and them sending proof to me. I can’t actually go to the bank and double check that. So I think like that’s something that we would like to be able to add support for like in the future, but it’s going to take some more time.
to develop those relationships and figure out the itty-bitty details of how can we make it easy for an oracle to be able to easily attest to bank wire transfers.
Stephan Livera (35:15)
Yeah, yeah, it makes sense. think around the, you know, stable coins and you know, the various chains that they travel on. Now, around the UX and convenience side of things, that’s probably another big, maybe sticking point for people because if I’m thinking about like, tip what a typical person might want, whether they’re like a Bitcoin maxi, ideological maxi, or just kind of a convenience maxi, like
For a lot of people, they’ll be thinking, well, I want a lending product that also gives me like a card that I can just like spend on that card. that way it’s convenient for them to be able to borrow and then use that fee that they just borrowed to spend, you know, for their bills or whatever. So do you have any thoughts on that and where that’s going?
Matt (35:51)
you
Jay Patel (36:03)
I think that’s definitely an interesting space. I’ll say that ⁓ in general, our thesis is that you can build a really good non-custodial product without sacrificing on the UX. It is more work, but I think, as Matt alluded to on the integrations front, we try to do our best to basically always let users know when they’re signing a transaction, what they’re signing. We avoid the kind of blind signing.
update your loan and then it just signs a bunch of stuff in the background. We want to make it very transparent what users are signing to. But that doesn’t mean that we can’t add more complexity later on. I do think that you could get to a world where we have a ⁓ stable coin wallet, maybe it’s Spark or ARK, or maybe it’s some coin chain, wherever someone wants to receive those funds where they could spend it via a card or they could basically ⁓ pre-agree to certain signatures. I’m sure there’s some more complexity we can add there.
Matt (36:54)
you
Jay Patel (36:56)
But I think the biggest thing for us is like, we’re always going to first and foremost stick to the fact that like, we want to keep things trust, minimized and non-custodial. Like we are, you know, never going to compromise on you knowing what you’re signing and what you’re doing with your keys at, know, to slightly improve the experience. I think we can do the more hard technical work to get there without compromising on
Stephan Livera (37:17)
⁓ And I guess it just yeah, it might just mean that users have to kind of do an extra step ⁓ Because as an example, maybe it means they just have to estimate Okay ⁓ Like move that money basically it might be an extra step of moving it moving So let’s say the loan gets funded in USD see they have to then find a way to off-ramp in their local country So that might mean finding an exchange somewhere and then transferring that out into their fiat bank account so they can have
a card that they can go around and tap and pay for things. This is if they’re in the case of using a loan for living expenses, right? Obviously, if they just want to use the loan to buy more Bitcoin, well, that’s not so hard. ⁓ And ⁓ maybe the business use is also kind of different also. But I guess just thinking about what other typical like UX things that users on the borrowing side of the house will be thinking about.
Jay Patel (38:14)
Yeah, I definitely think it splits into like planned purchases versus like maybe like the lifestyle kind of stuff that you were talking about. You know, if it’s like, I’m going to go buy Superbowl tickets or something all of a sudden, you know, that might work a little differently. But yeah, generally like right now the UX is really good. If you know, Hey, I have a business and I need to cover this amount of expense or I’m, you know, building a home or, you know, buying a plane or whatever Bitcoiners might be doing out there.
Matt (38:25)
you
Jay Patel (38:41)
But I do think we can probably get closer to a world where anyone, regardless of what they’re trying to do, can find a seamless way to interact. ⁓ And the other thing I’d say is ⁓ we do have plans for the fiat side of things to make it seamless, ⁓ whether it’s via integrated stablecoin off-ramps or funding via wires. We will be able to find some sort of ⁓ approach that works for people globally. ⁓ Obviously, different countries that…
changes, you know, what the easiest way to off ramp is, whether it’s USDT or USDC, or even in some places, it’s easier to offer at Bitcoin than it is stable coin. So maybe there’s like a funding the loan and a dollar amount of Bitcoin. ⁓ But yeah, there’s a, there’s a lot of cool things you can do because the funding side is actually kind of disconnected from the DLC. Like you could fund the loan theoretically on any transparent ledger. Yeah.
Stephan Livera (39:35)
Gotcha, yeah. Yeah, so I mean, I think it’s just interesting to explore that trade-off space, least so listeners understand, okay, what are the options here? What are the different trade-offs with that? ⁓ So, yeah, I think those are kind of the key, probably the key questions that I’m imagining. ⁓ Is there anything else you guys think we should cover in terms of like, you know, what should listeners know?
Matt (39:47)
you
I think we’ve covered most of it. think the only other interesting thing is like I mean, I’m always interested in the technical side like what’s what’s possible in the future, you know if we If we ever get any upgrades to Bitcoin one of these days, you know
Stephan Livera (40:14)
Well, didn’t TxHash recently
get assigned a BIP number and some other, you know, some other, you know, there’s been a little bit of movement, but you know, it’s, yeah, we’ll see what happens on, yeah.
Matt (40:24)
I thought
folks had some issues with TX hash. So now it’s like a slight derivation of TX hash now that is the.
Stephan Livera (40:29)
Yeah, so I think there’s
now people talking about what’s called template hash, which is kind of like a variation on CTV. I don’t know enough to be able to explain the difference, but there’s like some little technical nuances around that. but I presume like the same things from our prior conversation would apply that if we got CTV or TX hash or template hash and like these other covenant things, it would make DLCs even more simple to do, right?
Matt (40:57)
Exactly, yeah. So I mean, first off, CTV would allow, I mean, this isn’t an issue for us if we built all around this, but you know, if we did get CTV one day,
You know, those payout curves could come back. It’d be instead of it taking two minutes to enter a contract, it would take, you know, five to 10 seconds. Um, if we get like TX hash, for example, then you can do transferability of DLCs without actually requiring, um, the other party to be online to do that. So, you know, just some of these like little UX things that just improve DLCs overall. Um, all those improvements aren’t really necessary for, for our product. Like we’ve built it in a way that.
kind of perfectly as Bitcoin is right now but you know just for me personally like having built with DLCs for a while like I’m excited if we do ever get those improvements because we’ll just be able to build like even more cool products on top of Bitcoin ⁓ maybe one day we’ll see
Stephan Livera (41:55)
Yeah,
we’ll see. guess the other… Obviously, it’s kind of hot right now. There’s a lot of quantum fud, right? And so the quantum fud now is that, it’s coming really soon. Now, I don’t think it’s coming soon. I think it’s being a bit overblown. I think it’s more like, you know, at least from what I’ve seen on experts say, it might be like 20 years out. But nevertheless, quantum theoretically might end up wrecking adapter signatures, right? So would that like impact anything on the DLC side? Like if this quantum computer were to come?
But I guess, I mean, given what we’re talking about, we would need to have some kind of quantum resistant thing anyway. But I guess it might change the model of how, know, hypothetically, you 15 to 20 years from now, maybe the model would have to shift a little to kind of be a quantum version of this quantum secure version.
Matt (42:36)
you
Yeah, that’s,
that’s true. ⁓ so I guess for context, like we, first of all, like I’m not, ⁓ like quantum itself, like I don’t necessarily, I agree with you. don’t think it’s anywhere like close at the moment. ⁓ the other thing too is the, the DLC contracts that we have actually use like ECDSA adapter signature, not the Schnorr adapter signature, which means that, ⁓ the
funds are actually stored in SegWit addresses. So, you know, if you are a person that’s super concerned about quantum, you know, SegWit addresses are hashed. So that’s a little bit less of a concern. ⁓ And, but, like, you know, on that, on that point, like I also think that, you know, if we’re just game theorying this out, you know, the likelihood that someone’s going to go after, you know, figuring out the Schnorr signature for, you know, to be able to create that adapter signature for funds they don’t even own on SegWit to be able to create an invalid, you know,
adapter signature for an Oracle versus like being able to go after Satoshi’s funds or being able to go after those UTXOs of 50 Bitcoin each. I think we’re going to be safe for a very long.
Stephan Livera (43:46)
So you’re not the lowest
hanging fruit.
Matt (43:49)
Yeah,
we’re very far from the lowest hanging fruit. ⁓ One day if we do decide to upgrade to like Taproot DLCs, then like, okay, we’re still very like super low hanging fruit. We’re slightly above that, but still even then I think it’s not a big concern. ⁓ But I think it is important to be thinking about it and at least have pathways to be able to upgrade Bitcoin, which I feel like we’ve kind of been at a standstill. ⁓
Stephan Livera (43:52)
Ha
Matt (44:18)
would love for to, you know, I would love for Bitcoiners and the community to come together to be able to activate something. We’ve talked about CTV for so long. ⁓ Just be able to activate something on Bitcoin, I think would be good at this point.
Stephan Livera (44:32)
Yeah, who knows? mean, maybe it’s a BIP 360, maybe it’s a consensus cleanup, maybe it’s BIP 360 and one of these covenant things. I mean, who knows? But anyway, I think we’ll leave it there. Where can people find you guys online if they’re interested? I think it is interesting that there’s people building self-custodial forms of lending. I think it’s interesting. So tell people where they can find you online.
Matt (44:57)
Yeah, just follow us on Twitter, at lygos Finance, lygos.finance as well as our website. yeah, it’s been a pleasure, Really, really enjoyed it. Great to be back.
Jay Patel (45:09)
Yeah, no, love being on.
Stephan Livera (45:10)
All right, well thanks Jay
and Matt.