Mike Sall & Blake West: Goldfinch, uncollateralized loans in emerging markets.

Sina [00:00:18]: Hey everyone. Welcome to another episode of Into the Bytecode. Today, I sat down with Mike Sall and Blake West, the two co-founders of Goldfinch. Goldfinch is a protocol facilitating uncollateralized lending, and it's being used by a growing number of borrowers in emerging markets. In this conversation, we talked about the users actually benefiting from Goldfinch on the ground, which I think is really, really cool because the project is actually doing good in the world.

We talked about the complexities the team has faced ranging from hedging foreign exchange rates to working within the regulations of different countries. And we moved on to diving deep into the protocol and the design. Hearing about their experience with shipping a production smart contract system that has millions and millions of dollars flowing through it. And we also talked about their recent experience with launching their token and decentralizing to open governance.

Nothing on this show should be construed as financial advice. My guests and I may hold positions in the assets we discuss on the show. And that goes for this episode and any other episodes. With that, I hope you enjoy this conversation.

So probably a sensible way to start is just by describing what Goldfinch is. So if either of you wants to take that, that would be great.

Blake [00:01:43]: Sure. Mike, do you want to take that one?

Mike [00:01:44]: Yeah, I think a good way to describe Goldfinch is also just to share some quick background. Which is that I mean, one of the reasons I originally got into crypto was being excited about the potential for crypto to really expand access to capital and financial inclusion around the world. And I know those are kind of like buzz words a little bit in the crypto space, but I really believe it.

And while Blake and I were previously at Coinbase, we were just seeing a lot of the infrastructural pieces coming into place that could actually make it possible to have that kind of impact in the world. And so we really started Goldfinch with the idea of being this bridge between all of the capital that's sitting in crypto and borrowers out in the world who can put to good productive uses who have traditionally been underserved.

So the high level of like what Goldfinch is is this bridge between all the capital that's in crypto and these borrowers out in the real world who can put it to work. And then, sort of in crypto terms, Goldfinch is an under collateralized lending protocol, which is like compared to, say, Compound and Aave, where you need to put up crypto collateral in order to borrow other crypto assets. With Goldfinch, you don't have to put up that collateral. You need to instead convince the community, essentially to determine your creditworthy and give you the capital that way. So that's a high level of what Goldfinch is.

Sina [00:03:07]: That makes sense. So when you're talking about being at Coinbase and realizing that the infrastructural pieces already, were there specific things that happened that triggered this realization in your mind? Or how did you actually choose to make that transition?

Blake [00:03:22]: Yeah, I did; there was like, there were a few things. So kind of like late 2018, early 2020, Mike and I were first seeing, I mean, not first, but like you were noticing DeFi was really starting to take off, right. There was just kind of these glimmers. It was getting to like the first billion in TVL. And I remember at the time, Mike and I being like, oh my God, there's a billion dollars in TVL and DeFi. This is so nuts. Like, wow, this is going to be like a real thing.

And so, that alone, right, this idea that now there was like these, there were these users who had capital and they wanted to lend it out. They were starting to like seek yield and wanted to like do things with their crypto. Because kind of before that and before DeFi really got going, crypto was really just huddling, like that was pretty much what you did.

Sina [00:04:03]: Huddling and very like long-term idealistic visions of it's going to change nation-states and whatnot, but not so much that was applicable right then.

Blake [00:04:13]: Exactly. And this was kind of the whole magic of Ethereum, right. That was like, now you could really do stuff and build applications on top of the blockchain. And so that was one thing. Just seeing that there were people starting to have this expectation of earning yield and wanting to do stuff. There was like this customer base, if you will, on that one side of the market. This kind of lending side of the market.

Mike [00:04:32]: I would also, by the way, add to that about the demand. We were talking about this way in 2019 or early 2020.

Blake [00:04:38]: Kind of like late 2019 or early 2020, we started getting this. Yeah.

Mike [00:04:40]: Yeah. So this was before Compound launched COMP and started; I feel like the whole DeFi summer with yield farming. And like technically, yield farming already existed with synthetics, but it wasn't like a whole lot going on. But at that time, I think it was in 2019 that Coinbase launched Staking. Was it Tezos staking? I forget which one it was.

Blake [00:04:59]: Tezos was the first one. I was actually working on that project while I was there.

Mike [00:05:03]: Exactly. And I was running the data side, like the product analytics team there. Analyzing the numbers for that. And like a lot of people, like a lot [00:05:09 inaudible & cross-talking].

Sina [00:05:10]: You had kind of a front-row seat to see how the adoption was coming through.

Mike [00:05:14]: Yeah. And it was like notable, like a lot of people were for doing that. And it was like, oh wow. When you offer like yield to folks who have traditionally just been thinking about huddling like, for me, that was a bit of a light bulb moment. Like this is like a huge thing in crypto. And then, of course, once all the DeFi yield farming started during the summer of 2020, then it became super clear. This was like a huge component of crypto. But we were starting to see there were early signs of that. I feel like in 2019, early 2020, that like crypto holders looking for yield was like massive sort of area that was still untapped.

Sina [00:05:49]: Right.

Blake [00:05:49]: Yeah. And we were seeing that as like, oh Compound and Aave were starting to do this with the over collateralized thing. And we were like, that's awesome. Right. But that's like a slice of the market only kind of, that's really for like margin traders. People who want to go like basically take leverage and go extra long on their paddling. Or for people who've had a lot of gains and they want to get some liquidity but still like maintain upside exposure, that kind of borrowing against appreciated assets.

That's pretty much the use case for compound borrowing. And we were like, well, if we can do decentralized lending, that's under collateralized, oh man, that's going to be like a hundred X, thousand X bigger market. So that was part of what was getting us really excited. And to get back to, Mike was mentioning some of these infrastructural pieces that were coming into play. One of the other big ones at the time was just Stable coins and USDC tether. But there were a bunch of Stable coins coming out at the time, even though those two are really the major ones now, and that was starting to pass.

I don't know exactly what the benchmark was there at the time. Maybe like 10 billion or something like that of volume, but we were like, oh, these Stable coins are real. There's like a pretty sizeable amount of liquidity all over the world. And that really unlocks the ability to make loans to people in the real world. Because, but before that, you're pretty much what are you going to do? Lend out someone ether and like,

Sina [00:07:04]: Right.

Blake [00:07:05]: take all that volatility risk as a borrower? Like, you'd be kind of crazy to do that, especially if you're like a real business in the world, who's like, they think about 1%, 2% on an interest rate that's like can be meaningful to them. And so, you can't have your borrower [00:07:17 inaudible],

Sina [00:07:18]: Right. The yield you're getting is totally outweighed by the price movements.

Blake [00:07:22]: Yes, exactly. So without some sort of stable asset there to lend against, it really didn't make any sense. There was like no way to do real-world lending. So that was like another big piece. And then, a third big piece was seeing how exchanges were starting to proliferate around the world. And you could turn crypto really easily into basically any currency in the world. You had like finance and Coinbase really starting to launch in countries everywhere.

Latin America, Africa, and Southeast Asia. All of them were starting to have solid exchanges with real liquidity. And that's also important because the borrowers of Goldfinch, they take the crypto from the smart contracts like USDC. And then they go to whatever their kind local exchange is and turn it into their local currency. And they use that in the real world. And so we really needed all three pieces. Otherwise, we would've been just operationally and infrastructure; like much more difficult to build this business. And I don't think we really could have done it before mid-2020, which is when we launched.

Sina [00:08:17]: Right. Interesting. So you're partially relying on the local exchange infrastructure in these different markets where the borrowers exist. Right? Like how, yeah, maybe let's move to talking a bit about who the actual borrowers are. Right. So the borrowers are mostly credit funds who are focused in these different geographies and are borrowing USDC and then kind of doing their own underwriting and due diligence for like totally different verticals and types of more like low-level borrowers, right.?

Blake [00:08:53]: Really, I think it's easier to think about the borrowers; you're right. Like these credit funds are coming on to Goldfinch, and they're a huge channel for the scale of the protocol. But I think it's easier to kind of think about the borrowers as FinTech companies around the world,

Sina [00:09:05]: Right.

Blake [00:09:05]: in these different markets. They're the ones who are really using the capital like that gets borrowed. And that whole piece was actually like, just a big part of Goldfinch from the beginning. Because we did a lot of customer calls as we were thinking this whole under collateralized lending thing. We were like, yeah, okay, but like, where's the borrowing side? Like the lending side was easy. There was just so much money in crypto, and everybody wanted to earn yield. That was like obvious.

But then the borrow side. When we were talking to people initially about this idea, they were like, oh, well, like maybe it's crypto miners want to borrow. Maybe it's individuals who like are shut out of the credit card system or something like that. That was kind of the easy answers that people had. But we would talk to people and we kind of dug into that more. And like, just none of those really kind of made sense. Either people didn't really want to borrow, or the other options available to them were probably better than what crypto could offer. Because crypto still had a lot of its own disadvantages of user experience and gas costs. And like, it's slow. I mean.

Sina [00:09:59]: A lot of like nontrivial obstacles there.

Blake [00:10:01]: Totally. And so we were like, no, we need somebody who has like a huge pain point. So that they're really willing to go through all those nontrivial obstacles. And I've kind of had this like general longstanding interest in microfinance. And eventually, I was like, oh, I think there's, there's a lot of companies in emerging markets who don't have that same level of access to capital. They might be willing to jump through a lot of these hoops. And we just started reaching out to those businesses through some people on our personal networks. And we talked to dozens of these businesses before starting Goldfinch. And we just kept hearing that same story from them, which is I spend half my time trying to raise capital despite the fact that I have a profitable business and I have a lot of track record.

And what we came to realize is they were kind of falling into this gap of financing between like say 500K and maybe like 20ish million dollars. We're like if you wanted to borrow less than that in, a lot of these other countries, not in the West, their capital marketers aren't as developed. The pools of capital aren't as deep. And so, if they wanted to raise less than 500K, that could probably be done. That was like maybe pretty easy if they talked to family and friends or there are some investors there. You want to do more than like 20 million or something. Now you can start getting the attention of bigger institutions who are in Europe and America.

Sina [00:11:12]: More global.

Blake [00:11:13]: Yeah. More global. Exactly. But there was a huge gap there in the middle. It was just really hard to get capital. And it wasn't really commensurate with the risk of the business or the track record of the business. Like, I sort of should be able to get capital. And so we thought this was like a really good wedge to get into where you have all this money in crypto. Doesn't have a lot of other opportunities for earning yield. Especially stuff that's uncorrelated with the rest of crypto. So that we could offer a unique product there and that they would be willing to lend these places all over the world.

Because most of the crypto community is like so global anyway, and so that was kind of a thing for us where like we think this is a really good, unique borrowing market to get started with. And that was initial. I feel like I'm not sure what the tangent that brought us down to this. There was like an initial question.

Sina [00:11:54]: That's super interesting. It's kind of surprising in a way how disjoint the global like capital market is, right in a world where I feel like everyone's kind of chasing yield—chasing returns and going through a lot of effort and like a lot of lengths to get a return. But then, on the other hand, you see this in the venture world, too, where like startups that are based in other geographies just don't get the valuations that startups in like the valley do. And that's kind of starting to change also, but it's interesting how much geography is still a barrier.

Blake [00:12:32]: Yeah. You're totally right. And this was something we heard from just not companies. We also talked to just a bunch of we might just call experts in the field. People who've dealt a lot with financing in emerging markets over the years, and we heard that over and over again. There's just so much kind of; I don't know, geographic bias. Or just like sort of, I don't understand that market. So like, I don't want to go touch it because I'm not really sure what's going on. And I think a lot of it is starting to change slowly. Crypto is helping obviously remote work and just virtual everything. It's helping a lot there, but it's still there.

Sina [00:13:04]: Yeah. And so, who are some of these borrowers then? Because there are some like really interesting stories that you've published on your blog.

Mike [00:13:12]: Yeah. I can share some examples. So one of the borrowers is a company called Tugende. They're based in Uganda, but they were looking for capital to expand their operations in Kenya. And what they do is they provide rent-to-own loans for motorcycle taxis. So kind of like the Uber of Kenya for like taxi rides there they're on motorcycles. And they have like tens of thousands of customers who get these loans for motorcycles. And then they pay it off over a couple of years, and they're able to make a living by doing these kinds of like taxi ride shares. And yeah, so Tugende is one example. They borrowed like 5 million dollars to expand their operations in Kenya. They're already like very well established in Uganda and like expanding throughout Africa.

Another example is a company called Divvy Bank. They're like the kind of Clearbanc of Brazil, and they provide working capital loans to small businesses in Brazil. And so they've borrowed money now to make those loans. And then a third example is a company called Greenway there in India. And they have a really interesting thing. They have these sorts of like eco-friendly cookstoves that they provide to low-income households throughout India. And they make money on the carbon credits that they get from those cookstoves because they're, yeah, carbon-friendly or eco-friendly. And so yeah, they have a loan to expand their operations there in India and even beyond.

So those are like three examples, and it's like, we were across all of the FinTech and lending businesses that the capital's gone out to. There are now over 230,000 borrowers who are like receiving capital now. And it's cool -

Sina [00:14:50]: Like, would you count the motorcycle taxi owner as a borrower?

Blake [00:14:53]: Yeah.

Sina [00:14:54]: Like end-users?

Blake [00:14:55]: Yeah. Being financed by the loans.

Sina [00:14:58]: That's so cool.

Mike [00:14:59]: They probably, they don't even know probably that it's like crypto capital that is helping them grow their businesses or get their cookstoves and things like that. But yeah, it's very cool that like now, like all this crypto capital is like actually in the hands of these folks allowing them to grow their businesses.

Sina [00:15:13]: That's so cool. So you actually get to feel that you're doing good in the world, and you really are, right?

Mike [00:15:19]: Yeah. I think like going back to Blake's description of like us targeting this market, it was like a win-win. Because from an analytical standpoint, it made sense as like an initial market to go after. But also from like a mission standpoint, it was like, oh, from like day one, we can be achieving the impact and mission that we want to have at the same time that it like makes the most sense from like a go-to-market strategy too. So yeah, that's really cool.

Sina [00:15:42]: That's so cool.

Mike [00:15:43]: Yeah. That's cool.

Sina [00:15:44]: So then these borrowers like Tugende, right. So are they kind of, are they directly interfacing with the Goldfinch product and protocol and then kind of abstracting like absorbing all of the crypto complexity and then abstracting in a way from their own users. Right? So,

Blake [00:16:03]: Yes.

Sina [00:16:03]: they take in USDC. They convert it to their local currency. They have like substantial crypto operations under the hood, I imagine.

Mike [00:16:14]: Yeah. By substantial, I mean, so Tugende, yeah, tthey set up their own MetaMask, and we connected them with an exchange where they cooked up their bank account, like a typical like centralized exchange. And so then they use meta mask to draw down the USDC and send it to the exchange and then yeah, like basically withdraw that capital to their local bank account. And then they handle it from there in Fiat. And yeah, so Tugende is doing that.

Give the example of Greenway; Greenway is going through the credit fund Almavest. So, in that case, Almavest is like the actual borrower on the protocol, and they have their own exchange account, and they get the Fiat, and then they send it to Greenway. So it varies, but like divibank and Tugende, they're both, yeah, setting up with meta mask and using the protocol.

Sina [00:17:01]: That is cool. And how have their experiences been like, are you on the phone, like doing customer support for meta mask? Like for someone based in Uganda?

Blake [00:17:12]: We've done that.

Mike [00:17:13]: Not us. We have some great team; I mean, actually, Blake and I were doing it early on, but yeah got on the phone with them. Also want to make sure that like they hit the buttons correctly and don't like accidentally -

Sina [00:17:24]: Right, because there's no recourse. You can't get it back.

Blake [00:17:27]: Yeah. Yeah. You put in that address wrong; poof, it's all gone.

Mike [00:17:30]: But interestingly, like Tugende, for example, afterward, I think we had like a couple of small little hiccups in the process, and then we were chatting with them. They were like, honestly, this was way better than anything we've like ever experienced –

Sina [00:17:41]: Oh my God. Comparing it to like the, yeah, documents passing, word files passing back and forth and -

Blake [00:17:48]: Banks and yeah.

Mike [00:17:49]: Yeah. Even they were like supposed to receive a wire, and it got lost in like, I forget what country it was. There was like some bank in another country that like lost the wire or something. They had to like redo it. And it was like a whole ordeal versus crypto; they see it like, within a few minutes, basically right there.

Blake [00:18:05]: I remember them even, they told us like, I remember like they were trying to draw down and we had some bug on the front end. And it took us like a few hours to like end up fixing it and saying, oh, you got to come back and do your drawdown again. And they did it. And then they were like, yeah, we once had a bank just hold up our funds for like nine days just to like ask more questions.

Sina [00:18:24]: I mean, it happens in the Western world too. It's not even a foreign concept.

Blake [00:18:29]: Yeah, no, still they were like, oh, this is still way better.

Sina [00:18:34]: Yeah. Wow. I wonder what kind of other second-order effects this will have to have entrepreneurs on the ground and like businesses start to actually use crypto. And you're probably seeding this level of interest. That means that now they're going to go explore what else is happening. And,

Blake [00:18:54]: Yeah. Well, it's funny, you mentioned that. So we've worked with businesses in Nigeria since like the very beginning, and their crypto adoption is really taking off. It's like one of the highest sorts of like, I guess, densities of crypto users in the world. And they were mentioning to us that just as businesses, they wanted to keep more of their currency in a Stable coin. Because they were kind of concerned about inflation of the Naira, which is their local currency in Nigeria. So businesses were doing that. They were also telling us that their customers kept asking them like, Hey, can I get crypto with you? Or can I just like, hold money as Stable coin or something with you? And so they were talking to us about, like, what APIs can I use to make products that like will use crypto?

They just kind of wanted to get our opinion on that. And it's not something that we do at Goldfinch, but they, are getting super interested in this, and the users in, like, the people on the ground in Nigeria are getting super interested in that. And to us, that was like for Mike and I; we felt like, oh, this is kind of the Canary in the coal mine. Right? Like, eventually, everyone's going to be pretty interested in this. And if the users start having crypto accounts and merchants start having crypto accounts, then like eventually they're just going to be able to transact with each other natively, like through crypto.

Sina [00:20:06]: Totally like leapfrog the old kind of financial infrastructure and just move straight to crypto.

Blake [00:20:12]: Yeah. I think that whole like vision seems like it's getting closer. Right. You really need the L2s, right. Like you can't Ethereum just like so expensive right now that people in Nigeria can't be sending any money to like a merchant on that. But what we're getting there, right. All the L2s are starting to get pretty real. And so I think within the next maybe 18 months, or so we can start to actually see some of that more like retail usage.

Sina [00:20:35]: And the particularly interesting thing is that a lot of these people, you'll know that they're unique human beings, right. Because of like the KYC that like say the borrower is doing, the borrower lender is doing on them. And so that opens the door to, I don't know, like UBI of various kinds or voting like various kinds of governance on the local level. It could be really interesting.

Blake [00:21:00]: Yeah, it totally could. And, I mean, that's another thing we've been doing at Goldfinch. And we released that product a couple of months ago, called UID, which is this NFT for representing ID on-chain. And we wanted to do it for all the reasons kind of you're mentioning here. It, it opens up a lot of design space on protocols and on voting. So we're actually now doing quadratic voting now that we've released our token like that's only really possible because of this UID, which represents the fact that a user has been KYCed on-chain. But other protocols can use it. They're free to integrate into their designs as well, be totally free for other protocols.

The other stuff you're mentioning, like UBI and that kind of thing. I think to me, that's one of the things actually really exciting about crypto generally is like, if individuals and businesses start having sort of known public accounts on-chain, the types of like philanthropy and the types of the things that you mentioned, like UBI, like all that stuff becomes just so much more efficient than it is now. Right. Like you have,

Sina [00:22:01]: Totally.

Blake [00:22:01]: things like give directly. I don't know if you're familiar with them, right? Like, they just give money directly to people in villages and stuff like that around the world. And it's like, it's not even that much money. They give them like a thousand dollars. And that's like the equivalent of like,

Sina [00:22:12]: It's a huge difference.

Blake [00:22:13]: four years of salary for them. And they can, like, it can be transformative, and like, that's hard. You have to have the whole organization, and all this overhead to, like, make that work. But if people just had crypto accounts and you kind of had confidence that it was going to them, I feel like the amount of outpouring of philanthropic support from the West going to these other places would like increase significantly. I mean, they'll do, I hope. Who knows, but.

Sina [00:22:34]: Yeah. Does having legible addresses for anyone, even in the Western world again, right like a musician having a known Eth address that you can send funds to just because you're vibing with what they're doing.

Blake [00:22:46]: Yes.

Sina [00:22:46]: is like, there's no way to do that right now. You have to listen to their music or buy their merch or something like that. And so it's just a direct link that is super valuable in its own right.

Blake [00:22:57]: Yeah. Especially when you're talking about global stuff, right. Like it's one thing in America, maybe you go see someone live, and they put their Venmo on like a little sign or something. But everything, especially on the arts, is getting so much more digital and global. Like that they can't do that for their fans who are in like Europe or India. Right. It's harder. And I think you're right. Ether is like such a great way to do that kind of thing.

Sina [00:23:18]: So when you're thinking about more like global adoption of crypto in these emerging markets, like, what are the real gating factors of that right now? What are the bottlenecks, like layer twos is one because the gas fees are too high? Are there any other things that are fundamentally missing that would be needed before this all happens?

Mike [00:23:39]: I would say a couple of things. Well, one is the hedging of the foreign exchange rates. So when all of the, like, for example, Tugende, they need to get hedging instrument. And also, companies Nigerian need to get like hedging instruments to hedge against like changes in the price of the Naira, like local currency. And that's really hard to do today. You have to usually,

Sina [00:24:00]: Interesting. So when they lend it out, the price could change on them. And then, when they come to pay back their loan in USDC, they might not be able to, even if there's been no kind of default or anything on their side.

Mike [00:24:12]: Exactly. Yes. Right. Because, because they're, they're borrowing in USD terms, I mean, USDC, but it's; basically, it's USD terms. But they're lending out in different currency, and it's just like standard practice to require, I mean, and also just a good practice for all of these borrowers to, yeah, hedge the exchange rate through like forwards, I think, or futures. I forget how they do it. But it's expensive, and it's like really hard to set up, and you have to have like good relationships with the banks. And so, if you're a small company, the idea of like getting into all these hedging instruments is like a huge hassle.

And so part of what's good about working with credit funds a little bit is they sometimes have good relationships already set up, and they can like help get that in place. But that's a big hassle. And we've talked a lot about, oh man. Like how could you do that with crypto potentially if you started to have actual Stable coins in all these different currencies on crypto? Then you could start creating just like automatic hedging instruments and derivatives so that we could just solve this kind of in an automated way, which would be like amazing for these companies. But right now, that is like a real kind of hassle to get started.

And then a second big hassle is all of these areas. They have their local kind of like lending regulations where you have to register certain things. So, for example, in Brazil, we have to set up all these special kinds of accounts, and you register in certain ways. Same thing with like Kenya and Nigeria; you have to register it with the government. And so that's another reason why it's been helpful working with credit funds. You like are already familiar with these different jurisdictions because they kind of like manage that process and make it a lot easier.

But those are two big things where if, over time, we can kind of make it more automatic or like make like templates of how to do things more easily. That will be a big help, but that's been like a big kind of like friction for us to get through on these initial deals. But a nice thing is that is like, once we kind of figure it out in one country, like once we have something working in Kenya, then it -

Sina [00:26:13]: Kind of works for everyone.

Mike [00:26:14]: to do a lot more deals in Kenya. Yeah,

Blake [00:26:18]: I was going to say on the, on like the retail side of adoption, I think it would be probably pretty helpful if we saw more Stable coins in more currencies.

Sina [00:26:28]: Yeah.

Blake [00:26:28]: Because I think a lot of people's daily life, it's not in dollars. And even though a lot of people might like it to be in dollars, and that's kind of what was happening in Nigeria. I was mentioning they want their lives to be in dollars, but a lot of countries, it's not. And I think for crypto to really take off, they're going to need their own version of a Stable coin and the currencies around the world so that merchants can feel good about accepting it. They're not going to want to deal with the exchange risk that Mike's talking about by accepting some currency that they don't pay their taxes in.

Sina [00:26:53]: Yeah. And have you looked into like how that might be possible? Like are there Stable coins in some of these currencies? Like what are credible paths for getting there?

Blake [00:27:05]: We've started to see some notion of other Stable coins. Things like the Euro and the Yen, I think have their versions of Stable coins right now. We even, there was some company that was like pitching one of our borrowers that they were going to have like a Stable coin of the Real, which is the Brazilian,

Sina [00:27:25]: Brazil currency.

Blake [00:27:26]: Yeah. And I haven't seen a ton of others. I'll be honest; I haven't been keeping my ear too close to the ground on that one. I would imagine it would be the same thing as what Coinbase; the centralized path is so easy there.

Sina [00:27:39]: Right.

Blake [00:27:39]: And I also feel like it's something that governments can get like a little bit more comfortable with. Like they still have an arms-length way to talk to Coinbase in the American example or other banks around the world who may want to do their own currency. Of course, you could do a decentralized one as well. Some sort of like, you could have like a [00:27:58 inaudible] that's pegged to whatever currency you want in theory. Right. There's nothing that stops that from happening. That has all its own problems. It's less capital efficient to do that kind of thing.

Sina [00:28:08]: Right. Yeah. If you can design some sort of a programmatic algorithmic, Stable coin design, then you could anchor it to anything at all. I feel like that's holy grail.

Blake [00:28:16]: Yeah. That's still the holy grail that's missing in crypto.

Mike [00:28:21]: This is making me think of like a broader kind of macro question which is like, which do you think is more likely to happen first? Like a whole bunch of Stable coins in different currencies around the world or increasing dollarization of some of these kinds of economies potentially. I feel like there's like two paths it could potentially take from like a big macro view. As you end up just having hundreds of stable coins in different currencies, or you have dollarization with like USDC or other kinds of stablecoins,

Sina [00:28:48]: Right.

Mike [00:28:48]: in these market.

Sina [00:28:49]: What's your thesis on that? Have you thought about it much? Well, it feels like the dollar is kind of in danger in its own way.

Blake [00:28:56]: There's more and more inflation and stuff. And if you're -

Sina [00:28:59]: Increase in debt.

Blake [00:28:59]: [00:28:59 cross-talking] he would say. The dollar is not going to be; you should be sure of the dollar over the 10-year horizon.

Mike [00:29:07]: My personal view, it would be like larger economies will probably have their own Stable coins. But like a lot of the markets, we're going into their smaller economies. And so to me, it seems a little bit more plausible that they just start using USD Stable coins versus creating a whole, like, for example, like in Mali, I'm no expert in Mali, but like the idea of like a local Stable coin there sounds less likely to me than like adoption of other existing Stable coins.

Sina [00:29:37]: Yeah. It'll be very interesting and like kind of path-dependent, how all of this evolves too, and how the different governments react to it. Because if I imagine it's happening to some extent by people on the ground using USD, but then for the government, it means that you lose control of your monetary policy,

Blake [00:29:55]: Which is a big deal.

Sina [00:29:56]: and is why in like Venezuela, Colombia and stuff they limit, like you can't actually like buy too many dollars. Right.

Blake [00:30:03]: Yeah. Right. I actually think from the US perspective it would, it might be the sort of the enlightened path would be for them to actually encourage this dollarization through Stable coins around the world because that would just drive more demand for dollars and generally keep the prices up. And it's also a thing where like, that's not like a diplomatic thing they need to like push. They can just sort of let it happen, and it would just happen.

Sina [00:30:29]: Right.

Blake [00:30:29]: Yeah.

Sina [00:30:31]: Yeah. So one of the things that's been interesting about Goldfinch for me is that I think a lot of people building in crypto kind of probably rightly so gravitate towards very crypto internet native type ideas that minimize their touchpoints and interface with the real world because that's just going to involve a lot of complexity. And like we've talked about some of this complexity, right. Of like the exchange rates risk and all of this sort of stuff.

Sina [00:31:01]: So, yeah, I'm curious, like has it been harder, easier than you expected? How's that all gone?

Blake [00:31:09]: I mean, I'd say like broadly, I think the, one of the things Mike and I have talked about is that this real-world thing, it is harder because it has these frictions that we're mentioning. Has these complexities and more like legal and regulatory issues and stuff like that. But when things are harder, that's also where opportunity tends to be.

Sina [00:31:29]: Right.

Blake [00:31:30]: And so we kind of think that's an advantage. It's a place where a lot of other entrepreneurs in this space maybe don't want to even touch. Right. They're just going to have a sort of; I think Paul Graham calls this like Schlep Blindness. Like, oh, it just feels like a Schlep. I just don't really want to deal with it. And so we think that's part of an opportunity. And I also just think kind of more broadly, like crypto, needs to get into the real world. Right. Like there's just so much economic activity here. So much value.

Sina [00:31:54]: At some point, you to cross this [00:31:56 inaudible].

Blake [00:31:57]: Yes. And whether that's us pulling it into the crypto world or crypto sort of expanding and pushing out, it needs to happen in some way or another. So I think it is harder, but I think that's where the opportunity is. That's part of why we're kind of excited to be doing it.

Mike [00:32:11]: Yeah. Well, I'll speak for Blake, but I think we kind of like viewed it as a feature of doing it.

Sina [00:32:16]: As a positive.

Mike [00:32:17]: Oh, like we want to go and do something super hard because we were ready to like really work on something that's really hard. And then that's where the bigger opportunity is. And like probably like less folks all try and do the same thing because it's so hard. And getting into it, we said, how does it compare with expectations? My expectation getting into it was that it was going to be like extremely difficult. I would say, compared to my expectation, that it was going to be really, really hard. It has been easier than I expected or, and like what I would say there is,

Sina [00:32:48]: Everything is relative

Mike [00:32:48]: I feel like something we didn't quite grasp when we started is like, there's a lot of people who were like rooting for something like this to exist. And, like we were talking to the early borrowers, and like at the end of the phone calls, they were like, Hey, let me know when you have something ready because we want to start borrowing from this thing. Like there was, there's like a lot of like momentum and energy. Like people actually really wanted that. Like once we started having versions of it, it was easier than I expected to get folks who wanted to participate and like try to make it work.

And a lot of the early borrowers and partners that we're working with like they're brainstorming with us. How do we make this happen? Kind of. And like, so like it's like, yeah, it's hard, it's a pill, but it feels like we have more folks all trying to like make it work because we all want to like see it happen. And so that was something that I wasn't totally expecting, and that made it frankly easier than I was sort of prepared for essentially, because I was just thinking, this is going to be just crazy difficult to get off the ground.

Sina [00:33:46]: Yeah. It's like that thing of building hard things is sometimes easier than building easy things because you get people excited. Like people want to help you.

Blake [00:33:56]: Yeah. That was also, I would agree. That was something that was surprising was, just like just sort of, it was almost like people were like waiting for someone to like say the idea. And then they were like, yes, yes. You should go do that. And that was,

Sina [00:34:08]: Like, I don't want to go do this because it sounds really complicated and hard but I will root for you.

Mike [00:34:12]: Exactly.

Blake [00:34:14]: Yeah. We had, I think, more than one investor turn us down, being like, I'm not sure if this will work, but I really, really want it to work, and you guys should know that.

Sina [00:34:22]: Right.

Blake [00:34:26]: Yeah.

Sina [00:34:26]: That's awesome. Yeah. I mean, well, it's definitely one of the good, like positive stories. One of the positive crypto stories. So one part of this interfacing with the real world that I wanted to talk about was on the legal side. Right. Of like interfacing with all of these different governments and the US government where you are based, I think, and or the company has been based. I don't know how the decentralized protocol is structured.

And it seems that you've kind of consistently taken the path of working within the legal system. Right. Like doing things that are compliant. You're doing the KYC. You're doing the UID. So I'm curious how have you thought about this path? Like maybe what some of the specific constraints that you've had to design around have been? What its been like interfacing with other countries, if that's at all been relevant to this point?

Mike [00:35:25]: Yeah. Oh, I was going to; when I think of the biggest constraint, it is; actually, it's less from our perspective as like creators of a protocol. It's more from the perspective of the borrowers who want to use the protocol. So these are FinTech and lending businesses that have pretty stringent regulations and compliance requirements in their different jurisdictions. And the biggest one is around AML and KYC. Like they can't take money from folks who they don't know who they are.

So when we decided to get KYC implemented in the protocol, that was less of us thinking from, oh, is a protocol like what does a protocol need? And more from thinking like, what do these borrowers need to get comfortable borrowing from it? And like, they need to see the KYC there because they have much more stringent requirements than like crypto [00:36:18 inaudible] around the world. And so that has been, I think from the like regulatory, and like flip perspective, the biggest lift for us is really trying to think through the borrower's needs in all of these jurisdictions. And like, how do we make sure the protocol is going to work for them so that they're willing to grow with the protocol?

Blake [00:36:37]: Yeah. I would just kind of add to that part of the reason that we find that is really important is because we want this to be really decentralized. Like there is a different path that other protocols won't get into, like names or anything have taken where like the protocol is totally open. And there's no KYC, and it's all anonymous and stuff. But the borrowers who are companies, guess what, they still have the same concerns. So the like centralized company, that's like helping to build some of the protocol. They bare all that risk because they sign a deal directly with the companies who are borrowing from the protocol.

Sina [00:37:12]: Right.

Blake [00:37:13]: And like, that path is available. It's actually kind of the easier obvious path. You don't build this thing in crypto,

Sina [00:37:18]: So in that path, you'd be signing like an off-chain contract kind of assuring,

Blake [00:37:25]: Yes.

Sina [00:37:25]: them that like, you're going to give them some subset of the liquidity that's KYC?

Blake [00:37:30]: No, it would just be that the company. So like the borrower, like it would be as if Tugende to use the example here or just whatever FinTech company signs a deal with some company. And like that company is sort of the lender there. Right. So that company has been KYCed, and that part's fine.

The borrower doesn't have to ask questions about where; they don't have to go like multiple layers beyond the chain. You know what I mean? So like that,

Sina [00:37:54]: That's centralizing. Right. Because like, that company owns that, right?

Blake [00:37:58]: Yes. And that's all anonymous over there. So that company's really taking all that KYC risk. But to us, like that's not a decentralized protocol. That's just like somebody raising money from the crypto sphere and then lending it out in a centralized way. And we really want to build this decentralized thing. That's why, like, we are not involved in any of the deals in sort of like the legal way. It's these borrowers going directly to the community of backers. Right. And the community of backers, therefore, have to be KYCed because those borrowers have to know where they're getting their direct money from. And so, yeah, I think it can be a little a sort of non-intuitive tool, but just like the KYC is what enables the true decentralization.

Sina [00:38:35]: Yeah. So have you baked the KYC into the protocol, or how would you think about like, like there could be another level of abstraction on this where like these different stakeholders could interface with each other in a non-KYCed way? But then, like they could opt to only kind of take counterparties that have the KYC? But then, on the other hand, I imagine like a lot of the core incentives and like security considerations of the protocol around like civil resistance, like collusion, like all these things fall apart if you don't have the UID, right?

Blake [00:39:12]: That's right. So the whole trust through consensus mechanism, which we haven't really talked about much on the show, but is like this kind of core part of Goldfinch. It requires civil resistance. Right. And so we knew kind of from the beginning, we were going to have to have some form of civil resistance and ID that doesn't necessarily have to be KYCed for that part to work.

Sina [00:39:29]: Right.

Blake [00:39:29]: But it turns out that KYC is pretty good at that, and our borrowers need it for compliance. So it sort of killed two birds with one stone. And so that was kind of why that decision is there. And it is in the protocol, like it's on-chain, that the lender, the individual, or the backer has to have this KYC sort of NFT that we have, which is the UID. But, it's built in the protocol in such a way that they can sort of swap out any provider of idea that the protocol wants. Right. We sort of think about it a lot like Stable coins, right.

Like the Stable coin you see is centralized, but any protocols just used that. Right. And like that doesn't make the protocol not decentralized. Or it doesn't make the protocol centralized just because they use a centralized Stable coin. Right. And in the same way, like the KYC is provided by Warbler Labs, which is an alternative company that does the UID thing. And Goldfinch protocol happens to use that UID product, which is provided by Warbler Labs. And,

Sina [00:40:25]: But that could be like abstracted the way it could be like any number of these UID equivalents could be used.

Blake [00:40:33]: Exactly. And many are going to come up and [00:40:35 inaudible] may decide to switch to another one if it seems like it's a better option for whatever reason.

Sina [00:40:39]: Right. It makes sense. It's like threading the needle between like yeah, like interfacing with the real world and being like native, doing what's needed right now for these borrowers versus what could be the ideal in like X number of years when there are other like civil resistance. Or like proof of unique human-type methods out there.

Mike [00:41:02]: And I personally think a big part of the future of crypto is going to have this kind of like KYC on-chain in some form. Because early on, when we were working on this like, we were trying as hard as we possibly could to not require KYC on-chain. And, but like the more we talked to the borrowers, we're like, this is a straight-up blocker. The only other option is like what Blake described of like, we just set up an entity, and that entity does all the loans. And then that entity takes on all this risk. And it's not actually decentralizing. Like we can't imagine a future world where a protocol can actually serve like a substantial portion of global debt in the world where it requires like an entity like that to be sort of like in the center.

So we're just like, it is going to require KYC on-chain. For what we see is like Blake mentioned a hundred X, a thousand X with the kind of lending we already see on crypto. We're going to need that KYC on there to make that happen around the world and with the kind of impact we want to see. So like my personal view of it too, is like, it doesn't feel like it's taking, like, doesn't feel like a big trade-off. Because I think the future of like a big portion of the future of crypto that's actually serving real-world activity will end up having some version of this on-chain. I don't know if it'll be exactly like the way we've designed it upfront. Will probably change over time. But yeah, I view it as like just getting an initial piece in place that's going to help like get to the next stage of like the broader sort of crypto landscape.

Sina [00:42:34]: Totally. It's definitely a needed stepping stone in the least and like potentially relevant, longer-term too.

Blake [00:42:42]: Yeah. I actually kind of personally think that this whole like ID piece, whether it's KYC or other forms of civil resistance that are starting to pop up, I think this is going to be like a big sort of growing theme of the crypto landscape over the next year and two years. Because not only does it enable like KYC specifically, I think enables a lot of real-world use cases that are dicey now and swiftly getting enterprises on, institutions on who have just so much capital, right. And like this, KYC thing is a real blocker for them. And I think the capital's totally worth it, in my opinion, for the growth of crypto long-term in this trade-off of doing some KYC. But also, it just enables like so much more design space for the protocols. Right. You can do things like referral programs better.

You can do so much more interesting governance systems. You can do so much more interesting, like voting things like, and quadratic funding and like protocol funding of like public goods and stuff like. Just so many things require ID and some form of civil resistance. And even our stuff like trust through consensus, right. You just can't do that kind of stuff without some notion of sybil resistance. And so I just think without it, you're really actually handicapping your design space, which is only going to get more, like there's going to be more opportunity cost to that as we start having L2s. Which also just expands the design space of what like developers can do and the gas costs and stuff. Also, things like just being able to have like NFTs that are like specific to and address for things like vesting and these other sorts of design things. Like all this is very sort of address specific. And I think L2s and sybil resistance stuff are going to add a lot of this design space and protocols they're going to want to use it.

Sina [00:44:15]: Totally. And so the way the UID works, you're using a third party to do the KYC. Right? And none of the information is stored on-chain, obviously.

Blake [00:44:27]: Yeah.

Sina [00:44:27]: But then are you as a, is the company, does that have like the kind of underlying data or does that all kind of stay with this like third party right now?

Blake [00:44:38]: Yeah. So Warbler Labs is the company that handles UID, and they have like; basically, all the real data is stored with the third party, which is Persona. That's the name of the company. And Persona has like super

Sina [00:44:52]: Super clients.

Blake [00:44:52]: stringent, yeah, super good clients. They're a really large company. And they have super stringent security policies and data storage policies and stuff. So we feel really good about using them. And then Warbler Labs sort of handling the on-chain aspect.

Sina [00:45:07]: Got it. So they just kind of like ping you back, and they're like, yes, this is the real person. Yes. They're unique. They haven't signed up before. And then you kind of take that and map it to Ethereum?

Blake [00:45:18]: That's right.

Sina [00:45:20]: Yeah. That makes sense. One question that came up for me. So when you're talking about the borrowers needing this KYC, so taking the example of Tugende or another one of the borrowers, like, are these, is this an entity that's established in Uganda, or do these borrowers kind of set up US entities just to be able to borrow from this market? Like, what does it look like under the hood?

Mike [00:45:48]: So it varies. So a lot of these, like Tugende, I think they are a Uganda entity. I'm pretty sure. And for example, we worked with QuickCheck, and they are a Nigerian entity. Aspire they're in Singapore. They have, I think, a different entity, but they're not US.

Sina [00:46:07]: That's so cool.

Mike [00:46:07]: There are some companies what they do is they do set up, some of them will set up US entities so that they can get funding from US institutional investors. Because US institutional investors find it way easier to fund to other US entities than like to set up a whole operations around understanding Nigerian law. And like what has to be registered there and things like that. So the more established businesses that can afford to like set up a US entity will do that and have the time to do so. But a lot of these companies, they're in their local jurisdictions and they'll often like a lot of the legal agreements will go by like London Law, or even like US Law rather than local law, but they still have their entities set up locally. And you still have registration requirements often with the local governments.

Sina [00:46:58]: That's so cool. So you're really like kind of acting in this international playground

Blake [00:47:03]: Yeah. It's in 18 different countries right now.

Sina [00:47:05]: Yeah. So, yeah, maybe let's move to talking about the protocol because I am curious about digging into some of the design decisions. Like maybe a good starting point would be to describe like the brush strokes of how it works and also has it changed? Has it evolved over time?

Blake [00:47:24]: Yeah. So I can give some of the broad brush strokes. There are three main actors in the system. There are borrowers, what we call backers, and then there are LPs. And so we've talked a lot about the borrowers already. They tend to be these FinTech companies though there's nothing specific in a protocol that requires them to be FinTech companies in emerging markets. That's just who the initial market is.

And then the backers are, you can think of them as sort of like active investors. And so they look at specific deals. They assess the risk of a company. And the company they'll present tons of information to the backers. They have these data rooms. They have channels, where you go and answer questions. They do YouTube live AMAs. They show their track record and all sorts of stuff. So backers can use a lot of information. That's all off-chain to assess the risk, and the backers put money into a specific deal. As I was saying, they look for that, get like a fixed rate of return, whatever that is somewhere between, often between like nine and 15%.

Then there are these LPs. And so, the LPs are putting money into a senior aggregated pool. We just call the senior pool. And that money is more passive. It's kind of quote-unquote, set it and forget it. And it earns a lower return, but it's; also, it's diversified across all of the borrower pools. And it has these backers as kind of first lost capital. So an important thing to understand between the backers and the LPs is that backers are first loss. So, in traditional finance, you call this like a Junior Tranche and a Senior Tranche. So the backers are junior, and the senior pool is not surprisingly senior tranche. And so,

Sina [00:49:00]: So the backers have a lot of skin in the game.

Blake [00:49:02]: They have a lot of skin in the game, and this then is the whole reason why the protocol uses this idea of trust through consensus. To say, if we see enough backers putting their own hard-earned money into a given deal, then we can trust that that's probably a pretty good deal. And therefore, the protocol can automatically allocate capital from the senior pool into that same borrower pool that the backers have been putting money into. And so typical kind of leverage ratios as we call might be like three X on that. So like, roughly 25% of the capital would come from backers, and 75% would come from the LPs or the senior pool. So like, just as an example, right. If there was like a million-dollar deal, 250K we'd be coming from backers, 750K would be coming from the senior pool. And part of the economics that makes all of this system work is the idea that 20% of the interest from the senior pool is reallocated to the backers. So the backers end up earning,

Sina [00:49:57]: 20% of what?

Blake [00:49:58]: of the interest generated by the senior pool is reallocated to the backers. So depending on the leverage ratio, it'll like vary slightly, but backers will often end up, they can be making sort of like a 15 to 20% interest because of that reallocation of the interest. And so that gives,

Sina [00:50:16]: It's like a fun model for them where you are sourcing all of the capital. And if they perform, they take a performance fee on it, basically.

Blake [00:50:23]: That is a way to think about it. Yes.

Sina [00:50:25]: But they also get to lose. They don't only get to play with other people's money. They also get [00:50:29 inaudible].

Mike [00:50:31]: They have to lose.

Sina [00:50:32]: Right.

Blake [00:50:32]: They have to lose. Yeah. All of their money gets lost first before a dollar of the senior pool money gets lost. Right. And so, but that's their incentive, right. Do a good job, and they can make this outsize return on their money, and do a bad job, and you lose all your money. So, the economics are aligned with everyone. And this is why the senior pool can kind of get comfortable following this like wisdom of the crowd from the backers.

Sina [00:50:56]: Yeah. So these backers are they also kind of existing research and due diligence like underwriting firms? Or are there like new backers entering the system who are doing this for the first time?

Blake [00:51:13]: It's both. We've seen sort of traditional finance people come in who have a lot of experience looking at deals like this. We've seen them be backers. And then there's also a lot of what you might call Crypto Degens. Individuals coming in and taking a look at these deals because they're offering solid, Stable coin returns, and they want to get in.

Mike [00:51:34]: And I would add to like this, the community of the backers is something we're still in the early stages of kind of forming and learning who are the kind of folks that are going to find this to be most compelling and want to spend the time evaluating it. So like, I would say over the next year or two, I wouldn't be surprised if the composition changes as we potentially like attract more folks from traditional finance involve. And then help educate existing folks and like see new people from the crypto space come in. And so, but yeah, our hypothesis is that it will be a combination of sort of new folks learning how to evaluate these borrowers. Because we're learning, it's like, it involves some education, but it's not like crazy. It's definitely like a learnable thing.

Plus, getting folks who've already been doing this, like existing analysts in traditional markets, to want to participate in crypto because the economics are really compelling. And so I think that's going to be like, we see like, yeah, this early sort of stage of folks who are participating. Then I'm kind of, I'm like personally very excited to see how the community kind of evolves over the next year or two and like how the processes and norms get formed.

Sina [00:52:39]: Yeah. And so, while this is kind of in the earlier stages or like before decentralizing and opening the system, I imagine you can kind of be more hands-on with who the backers are, who the borrowers are. But I imagine that the intention is that like people can literally come and start using the protocol. Right. And not really interface with the core Goldfinch team.

So what are the incentives at play here? I imagine the borrowers, right, like if you truly are a FinTech and an emerging market, this is basically you have access to this capital pool. And so you really care about that. You don't want to kind of like mess that up. I know you have this concept of like auditors. I know there's like the junior tranche. Like how do you think about these different pieces on what the risk model is here?

Blake [00:53:33]: We can talk about the auditors and get into sort of what their place in the protocol is. So the auditors their main job really is to handle and prevent fraud on the system. Right. And collusion in the system. So, borrowers could, for example, collude with backers to sort of fake consensus and then essentially steal all the senior pool capital that comes in on top. Right.

Sina [00:54:00]: Right. Like you and your friend could do, the KYC could like go into the protocol, and one decides to like underwrite the other, and then you get a leverage ratio from the senior pool.

Blake [00:54:10]: Exactly. That would be like the kind of most obvious and like worst kind of fear of the protocol in general, that that would happen. And so this is where the trust of your consensus thing comes in like, okay, yeah, I guess you could do that with one friend, but like, can you do it with like hundreds? Right. And what is the number of people that it's going to take to actually be able to get real leverage in the system? And so those like exact numbers are kind of still being figured out, but like, it's probably going to be substantial, right. It's not going to be you and your friend who can then get like three or four X leverage ratio. And so,

Sina [00:54:45]: Oh wait, because multiple backers need to kind of confirm the deal?

Blake [00:54:48]: Yes. Multiple backers need to put into a given deal in order for the protocol to consider that like trustworthy.

Mike [00:54:55]: It is worth pointing out the general sort of model is that the leverage ratio increases as the number of backers increase. So,

Sina [00:55:05]: Right.

Mike [00:55:05]: it's like directly proportional to how many people have actually provided that first loss capital.

Blake [00:55:11]: Yeah. And so the auditors can step in there as being, they're essentially randomly selected. It's a little bit like a proof of stake mechanism in typical blockchains where like they stake the token of the protocol based on the amount of token they stake gives them a higher chance of being selected. And they have to perform this task that the protocol needs. And so, in the case of auditors, that's performing this task of just kind of looking at like really basic stuff about the borrower and the deal that's happening. Things like, okay, does this borrower look legit? Do they have like a real website? Do they have backers who seem like they've actually participated in other pools, or all the backers, like brand new for some reason, right?

There's a lot of kind of like basic information. That's this human-level check. And because they're being randomly selected, it would make it much more expensive to like bribe all of them. Right. You'd have to bribe like the entire thing. And, the KYC aspect actually makes for a different type of security model, I think, than in a proof of stake mechanism. In the typical, like proof of stake stuff. It's this pure kind of economic model where it's literally just about the money you have. And how much money could you acquire to then try and sort of attack the system.

But in the Goldfinch case, it's different than that because you'd have to also coordinate with people, humans, and like a lot of humans actually. Or you'd have to like, buy all of these addresses and buy these like IDs to do the KYC. But you still need to do it with a lot of humans. And it's not just this pure money thing. You just like go to some marketplace, like buy all of these. And so that coordination cost is, it's an interesting thing. We've talked about it. Like, we're not exactly sure how to model it, but we know it's greater than zero.

Sina [00:56:48]: Right.

Blake [00:56:50]: It doesn't really matter how much money you have. It's still tough to like, just literally coordinate with a lot of humans. And there's not like one price that every human is going to have that you could like model in any way. And so it's been tough to, like, think about that sort of quantitatively, but we're sure it's more than zero. And we think it's actually pretty substantial that like having to get a thousand people to do something all at the same time is actually like much more difficult than just like some amount of money.

Mike [00:57:15]: Well, what we've talked about is like, you'd actually, it's easier just to start up like a straightforward lending business and like borrow money reasonably well, and use it for productive means than to try to like convince a thousand people to all coordinate.

Sina [00:57:28]: Right.

Blake [00:57:28]: Yeah.

Sina [00:57:30]: And it is kind, like the auditors being selected randomly out of like the people who are staking the Goldfinch token. Like the kind of proof of stake mechanism is also interesting. Because like the security of the network is going to scale with the total value of the network, with like how much Goldfinch is worth. And so, there will just be more and more at stake securing each of these pools. And so, I imagine it's going to only become more secure over time. Right.

Blake [00:58:05]: We would hope. And also, we would hope that there's a larger and larger pool of auditors who are willing to stake their token, which means that the random selection, there's like a lower chance of being randomly selected at any given time. Which means you'd have to either bribe more people or coordinate with more people to get them to actually work as an attacker.

Sina [00:58:23]: So, like the role of these auditors, it's not a software client you're running on your computer that's performing some automated task, right? It's like, you need to do something in the real world when your turn comes up.

Blake [00:58:38]: That is correct. Yeah. You need to do something. It's a human-level check. I think as much as possible, we would love to turn that into some software client that people could just run, but it's, it's not going to be that.

Sina [00:58:49]: Yeah. And then, okay, so this idea of the leverage ratio of like the senior pool backing the backer pools, how's that determined? So it's the number of unique backers who've kind of underwritten a particular deal. How does that work in the protocol? Have you thought about that?

Mike [00:59:09]: It's yeah, and this is one of these things that we'll need to introduce. Like that will need to be, get introduced to the community to approve. It's not in the protocol today, but the -

Sina [00:59:20]: And then this is all like really interesting stuff because this is where the rubber hits the road of like doing uncollateralized lending. Right? Like this is like the meat of how this all works.

Mike [00:59:29]: Yeah, exactly.

Blake [00:59:30]: Yeah. A truly decentralized way. Yeah. An automated way.

Sina [00:59:33]: Yeah.

Mike [00:59:35]: Yeah. So the general idea is that, say, okay, like two or three backers are providing to a loan like that's not really enough. Like, is there 10, that might not even be enough to call that like consensus of a group of people. So there'll need to be some parameters like the minimum number of backers to start seeing any kind of leverage. And then as the number, like the actual number of backers who participate in that pool, goes up, then the leverage ratio will increase up to some maximum leverage rate.

So I'll just throw out random numbers. We'll have to see what the community decides for the parameters. But like, let's just say you need a hundred backers to get even a penny of leverage for every dollar or $10 that goes into the pool. And then maybe as you go from a hundred backers all the way up to 10,000 or a hundred thousand backers, I'll say something crazy. You can get a leverage ratio that goes from zero all the way up to like 8, 9, 10 X leverage potentially. Or maybe we start off being more conservative as a community. But that's like the general idea behind a leverage ratio is that it is sort of on this scale based upon how many backers are in that particular pool.

Sina [01:00:48]: That makes sense. And how do you, so the role of a backer it's kind of this underwriting role, right? There could be people or entities who could perform that role without having like capital. Right. So do,

Mike [01:01:03]: Yeah.

Sina [01:01:03]: do you have any thoughts on how to like decouple it from needing to also put capital at play?

Blake [01:01:11]: We've thought about that as like kind of this reputation interesting idea that could be introduced to the protocol. And it's the type of thing that like is like, it sounds really cool, and we want to do it, but we were just like, weren't sure how. Like it's sort of a complex thing to think about early on. And so we've mostly been sort of pushing it down the road. But I think the crypto landscape is going to make its own thing. This doesn't even have to live within Goldfinch specifically. Right. I think what you're really talking about is it's the same kind of thing that's happening in traditional finance forever if someone proves that they're somehow good at picking things. And then they're able to raise money themselves, even though they don't have the capital, and then do a thing. Right. And so I think this,

Sina [01:01:52]: Right. It could even happen offline, right? Like [01:01:55 inaudible].

Blake [01:01:55]: Could even happen totally off-chain. Exactly. I bet that this is one of these things that's also going to be really coming to life in crypto over the next year or two. Individuals just like essentially raising all their own funds on-chain. I think like there are protocols that are set up to do this kind of thing now. Like Enzyme, I think is one, and there's like Iron Bank was like sort of like this. Yearn is even really basically like this. Right.

Sina [01:02:17]: Right.

Blake [01:02:18]: Then I think this is just going to keep exploding. And so those individuals who prove themselves completely outside the scope of Goldfinch may then decide to want to participate as backers. And I think that's going to happen whether we do anything about it or not.

Sina [01:02:33]: It's like another way to get leverage on and monetize the reputation and trust that you've built, right? In this part world.

Blake [01:02:43]: Yeah. And that's also another example where that whole system of like doing funds, it works better if there's a notion of identity and that it's like hard for you to switch to just like spin up a new address. Because otherwise, you could raise much money, you could lose it all. And then up, I'm just going to set up a new address and start another fund and do all my same marketing tricks. Right.

Sina [01:03:03]: Totally.

Blake [01:03:03]: But it would be really nice for the crypto space if there was some like hard to change address, it's like if you lose the money, you're going to have to deal with that consequence.

Sina [01:03:13]: Yeah. I mean, like anonymous, sybil resistance, like proof of unique human identity is like another one of these holy grails that people have been trying to crack for like ten years, probably more. And it's a really hard one. So I do think that it's going to go in this direction, and it's sort of like the right trade-offs are being made. Right. People's information isn't being put on-chain. It's one global system.

Like you're now trying to control this like registry in any particular way. So I'm excited to see where this all goes. So one question on more like the building all of this, like this complex protocol. What's it been like shipping a production smart contract system that has millions and millions of dollars flowing through it? And has that like, has that experience been different from previous engineering experiences you'd had at Coinbase and other places?

Blake [01:04:16]: Yeah. That's a good question. For me personally, like my software engineering career has, I'm just actually putting this together like right now. It's pretty much only been at companies where like we deal with money processing. Because the first one was technically a healthcare company, that's how I normally think about it. But really, we actually just processed a bunch of healthcare payments. And so systems there we're dealing with money. Coinbase obviously deals with like a huge amount of money. And I was like on the core payments team there and worked on a lot of the reward systems and stuff for paying people out.

And then now we're at Goldfinch. So like, it is interesting. I think one of the things about crypto that feels different than the other two, like Web2 versions that still had tons of money flowing through them, is just the like lack of guardrails that are present. It's this really adversarial environment where people can all see your code. It's all like public, an open-source. And if you have bugs, like somebody can spot it. And then if they take the money like that's kind of it. They just took the money.

Sina [01:05:14]: It's so wild to me that this is the state of affairs, that any of this stuff works at all. Just knowing how difficult it is to write software and that there are bugs everywhere.

Blake [01:05:24]: Yeah. And that it's so hard to fix bugs too. Like the idea like a hotfix in crypto is like, it is actually like it is really different. Like usually, crypto code gets audited, and like, it costs like an insane amount of money to deploy stuff too. It's like one of our deploys end up costing like over 30K.

Sina [01:05:40]: Oh my God.

Blake [01:05:41]: Yeah. I know. There were like NFT drops happening that day, and gas prices were crazy. So the idea of like, oh, let's just deploy it again. And like, let's have five deploys a day. Like most about two companies are, is just like completely off the table. So you have to take this like different approach. It's not; you can't be like quite as just like iterative and like move fast and break things. It's just like, not really an option in the crypto world.

Whereas even at Coinbase, you make a mistake, and like you credit company, you credit the user with like the wrong amount. Most of the time, not always; you can just undo it. It's annoying. But you can just like write a script to like fix it, or you can call their bank and say, Hey, sorry, that was a mistake. You need to send it back. And like, they'll do that. So you still don't want to make mistakes because it still really pisses people off when the money's wrong, but the money doesn't feel like it's at risk of like truly being lost the way it does in crypto.

Mike [01:06:30]: This is reminding me, Blake, of when we deployed the very first contracts we were using with our first customer, and we got some money in that. And I think it was like within 24 hours of deploying those contracts and putting money in there. There was like some like failed transaction from some address that like called the withdrawal function on the contract. Like we had never shared with anyone that the contract had been deployed. It was just like a hundred K that was sitting in there and like some bot or something must have noticed a withdrawal function and like called it and like gladly we had checks in there that say you can't just withdraw the money just by calling the function. But like it is just kind of wild that as soon as you deploy the contract immediately, there's like a hack attempt just automated on top of it.

Blake [01:07:16]: Yeah. And actually also remember that one of those very first versions, it was actually you, Sina, who put us in touch with his first name is like Bigelow from the Ethereum Foundation.

Sina [01:07:26]: Yeah. Scott Bigelow.

Blake [01:07:27]: Scott Bigelow.

Sina [01:07:28]: Amazing person.

Blake [01:07:29]: And he did this like short audit with us and within like,

Sina [01:07:34]: A spot check.

Blake [01:07:34]: Yeah. Like spot check

Sina [01:07:35]: like quick security review.

Blake [01:07:39]: Yes. Have to thank you for that. Because within about like five minutes, Scott had been like, oh, bug there, you could lose all your money. And I was like, what?

Mike [01:07:48]: I remember that. Because I was on the couch, and you were like over on the table talking to him. And then I just saw by being like, holy crap. That was -

Sina [01:07:56]: Wow.

Blake [01:07:57]: Yeah. So Scott was great, and thanks for saving us from deploying something that would've lost a bunch of money.

Sina [01:08:03]: Oh, I didn't even realize that, but I'll take credit for it indirectly.

Blake [01:08:07]: We did have like a real audit scheduled after that. So hopefully, they would've caught that too, but Scott was great and caught it before that audit happened. But I mean that stuff's there, and if it was going to happen, then like, woo, it's all over.

Sina [01:08:20]: So one of the things I've admired about Goldfinch from the beginning is how consistently and quickly you guys ship. Right. But in this environment that you're describing the same methodologies of, do we like deploy code frequently. Like move fast, like that sort of stuff is risky. So I'm curious, like have you kind of walked the line between iterating and finding the product-market fit that you have while operating in this kind of risky environment? Like what is your product development philosophy, almost?

Blake [01:09:04]: Yeah. I'll give some quick thoughts. So I mean, definitely just general product development philosophy still is iterate and like shift as little as you have to, to like learn the next thing. But I also think that as the protocol has gotten more mature and there's more money at stake like, those trade-offs need to shift. Right. It was one thing when we were first starting, and we actually like the very first thing we shipped; we didn't tell anybody about it. It didn't really even work for like a full life cycle of a loan. But it was something that allowed a customer to draw down money and pay it back. And I think the accounting was so all wrong.

Sina [01:09:38]: So unaudited, like probably sources isn't even verified on Etherscan.

Blake [01:09:43]: Yeah.

Sina [01:09:43]: You're just like putting out a prototype to like get the money flowing.

Mike [01:09:45]: Yeah. We had the, our first customer, which was this company, QuickCheck, in Nigeria. And we got on a call with Louis there who like runs the company, and we're like, okay, go drawdown from this thing and then pay it back just to like, make sure he could actually do it.

Blake [01:10:02]: You could like take it to the exchange and like turn it into your local fiat. We want to have like, what are the exact costs? How much are you losing on that to make sure we get the economics, right? And we're not like missing random fees. And then like, okay, now like pay it back a week later and let's make sure the interest got calculated sort of the way we thought it would. And all that kind of stuff.

But that was just for like a thousand dollars. Right. And we just did it ourselves. Didn't tell anybody about it, but we got that full end to end, which is like, we had a bunch of great learnings.

Sina [01:10:31]: Right. And then what happened from there?

Blake [01:10:34]: Then we just, we were like, okay, now let's do like a V one. And that V one didn't have any notion of tranching in the pools like backer stuff wasn't there yet. And like we had actually even, we knew we wanted to do that, like that was still like part of the design. But we were like, oh, this is just going to like, be so much to build. Like we just don't have time for that. Like we got to build something that's going to work for someone before that. So that was a V one that was just really simplified. And that one, we got that audited, right.

Like we've always said, like, if we're going to put real money through it, especially other people's money, but like, it has to be secure. That has to be really the top priority. And so then, when you have that constraint the way you shift things around is really about the features you build. And you think hard about what extra complexity are you adding and is there ways to like reduce the amount of change in the system. But we can't budge on the security aspect now, but you move fast by picking the right things to work on.

Sina [01:11:30]: You do this upfront scoping of like what is the level of complexity we want to take on? What are the features? You book the audit, which I presume is going to be scheduled like three, four months later,

Blake [01:11:42]: Yeah. It's so annoying -

Sina [01:11:43]: and then you start like working on the feature that you kind of have this deadline that you need to hit. And it's basically this internal process of developing, like dogfooding shipping to like local Testnet. Like actual Testnets and then doing the security audit at the end of all of that. Like, is that how the flow goes?

Blake [01:12:03]: Yeah, that's pretty much exactly it. We do a lot of unit tests. We do a lot of Mainnet forking tests too. So we actually don't use Testnet that much. I find Mainnet forking actually be like generally more useful. But we do Testnet on big deploys and migrations as well. And we poke around on our front ends and make sure all that stuff works. And yet the audit is kind of the last step in that process.

Mike [01:12:23]: But in a weird way, the audit is the first thing that gets scheduled. So like, we have a general rule, which is like every line of code that we submit basically and deploy needs to be audited. But then these audit firms they're like booked up six months in advance, it's crazy. And like, we're planning what we want to build, like two, three months in advance. So it doesn't work. But what we did was we booked out audits like once every two months. I think what for the next year Blake basically? Yeah. So like we just know we're going to [01:12:53 cross-talking] basically.

Sina [01:12:54]: So that's what - book them with your like shipping cadence? You know what the deadlines are?

Blake [01:12:59]: Well, yeah. Which I don't love.

Mike [01:13:01]: I would agree.

Blake [01:13:02]: Yeah. It's really not great. Because things just sometimes you want an extra week. Or you have this part done, and you'd like to get that audited a little earlier than the actual audit, but it's very annoying. This is like a huge problem in the industry, I think. But luckily, I'm seeing more audit firms come up, which makes sense because they're like making a boatload of money right now.

Mike [01:13:21]: Yeah. So then our process, now that we have these days for the audits, and we know that every single line of code has to go through an audit before it's going to get sort of actually deployed to Mainnet. Then like Blake was describing, it's this iterative thing of like from a feature standpoint, the philosophy is like, what's the very next thing we need to learn? Or need to test out on the protocol? Like what's the minimum version of a feature that could do this, and then what can we build and what can we build before the next audit date? And then basically just build that. And so we're not sort of minimizing the security there, but we're minimizing like the actual feature set that we do with each iteration.

Sina [01:14:02]: Seems kind of nice in a strange way, right? Like it's the constraint breeds creativity, and it like brings more discipline into like actually choosing what you're going to work on. Prioritizing. Having a deadline is kind of nice and, again, in a weird way.

Blake [01:14:19]: Yeah. I'll take that as a silver lining.

Mike [01:14:21]: It's a silver lining. I mean, it's really hard to adapt quickly to things like when we see something, and we want to like change something. But like, oh, we only have a month before the next audit, we were planning to do this thing. If we want to like build something else before that audit like, it's actually pretty challenging. And so it can makes it really hard to adapt even if we have things sort of planned out. But it is nice to have the cadence and have a forcing function; that means, okay, we got to get this code out and audited before the next thing like forces you to iterate.

Blake [01:14:52]: Yeah. Maybe the ideal there would be if we had the cadence, and I could do like plus or minus two weeks.

Sina [01:14:57]: Yeah. Right. Cool. Yeah. So the next and last kind of big bundle of things that I was hoping we could talk about was you kind of opening governance and launching the token, which happened relatively recently.

Blake [01:15:13]: Yeah.

Sina [01:15:13]: I'm curious, first of all, just from a personal, like emotional going through this crazy process of having a token out there on like CoinGecko now and like entering this wild world of crypto, what has that been like?

Blake [01:15:32]: So it was like so much more work than we thought it was going to be. It ended up being like a five-month project. I think we were talking to you about this. Like a few months ago, we were like, oh, we’ll be done with it like another month or two. And no, it was like another like three months. But so one, it was just a bunch of work, and I feel like it was a huge milestone for the team. Everyone was just like, so happy to finally just like get all this stuff out there. That was like so much material that had been written both from like on the front ends. The smart contracts, docs, announcements, wrangling media, right. Like just everything right. There was like so much stuff, and I'm not sure I've fully appreciated the like complexity that a token adds to a project.

I'm just sort of coming to recognize that now like post token launch; I'm realizing that like, there's just like multiple layers of complexity that weren't there before. Specifically, the stuff I think about, there's kind of this like community politics layer that comes into play now. Because if tokens have been delegated to one group or another, there's like this implicit sort of like ranking of like who's important and like not important and like how much you should give. And like as new features need to come out, there might need to be adjustments to those token allocations.

And like all that stuff is like super important to people, right. It's money now. It's like a lot of money, actually. So that politics angle was something I hadn't really fully appreciated, I think. And then also thinking about the economic complexity that comes with this and like the second-order effects of token price, right. Because token price can affect the APYs that are seen on the senior pool or as backers and like that stuff can be reflexive like up and down.

And then thinking about, oh, well, do you have like vesting or not vesting? And if there's vesting, then it'll like reduce token supply. Maybe that keeps price up, which keeps APY up. Or that if you reduce prices, maybe that increases capital in that you can use to lend out. But also, maybe it puts more tokens on the market, which can decrease the price, which can then affect the APYs and reduce the amount of capital that comes in. So like there's this like very self-reflexive and complex,

Sina [01:17:30]: Monetary policy.

Blake [01:17:31]: Yeah. And of course, all this stuff, as you're saying, it's now like a community-run protocol, right. And so token holders can vote on changes to the protocol. Any change to the smart contract stuff is going to be voted on by the community. We're using quadratic voting to do that, which I'm like super excited to be one of the first protocols to be able to really do that with fully on-chain data because of UID. We've already actually had some of the first votes have gone through, and it's really cool. So, yeah. I don't know. I think those are just some of the key learnings I'm thinking about is just man, the token adds a lot of complexity. That's, it's hard to wrap your head around.

Mike [01:18:05]: I mean, there's the complexity of the economics of it. Second-order effects and part of like the sort of politics thing, and working with the community. We were talking about how difficult it is to plan the work that we're doing just around audit schedules. And around like product iteration and trying to adapt quickly. And then we add a whole new access here, which is like everything that we want to do from like the Warbler Labs team. Like, we're just one among the whole community. So now, not only do we want to try to like do things that might be improvements for the protocol. But we have to plan through, okay, how do we propose this to the community? And how do we make our case? And how do we get them to be excited about it, so they vote it in?

And so that's like a whole other element that I'm also starting to learn like, oh, okay, like the process on governance is like you make a proposal and you open it up for discussion for a week. And then, after it's discussed for a week, there's like a three-day vote. And so then, okay, now there's like a week and a half. If everything goes perfectly well, that we need to incorporate into our planning for just general community discussion of things we want to propose to the community. So I'm just starting to sort of internalize that aspect of the complexity, too, of like shipping new thing.

Sina [01:19:21]: Yeah. It's a lot of complexity. Yeah. How has the community evolved since this token being out? Right, like with people having financial stake in this thing?

Mike [01:19:34]: In the last two weeks, I haven't seen a big difference except less questions in the discord about like when is there going to be

Sina [01:19:40]: When token?

Blake [01:19:43]: Yeah, less of that. They now know. But I mean, the community has been pretty engaged for a while now. And I think a lot of them are excited about the mission of Goldfinch. I think that's pretty much remains the case even after the token. And hopefully, it grows from here.

Sina [01:19:57]: Yeah. Another thing I was curious about was liquidity mining and like these incentives for growing the protocol. And you also alluded to other potential ways of doing this when you do have KYC, right, where you could do affiliate programs which is a totally unexplored territory. Which I think you will probably be well-positioned to explore. But I'm curious how the liquidity mining stuff like has gone. What sort of an effect has it had on the growth of the protocol on the sorts of people who are engaging with Goldfinch? Because it's always like a big variable to turn on these sorts of things. So I'm curious what like your actual experience has been?

Mike [01:20:43]: Yeah, it's been super interesting. So let's see, as part of launching the token, the protocol added liquidity mining, which is it allocates the GFI token to folks who provided capital into the senior pool. And it's this like interesting kind of a dynamic rate where the protocol sets to target. So right now, the target is at a hundred million dollars. And then, as the protocol reaches its target, the actual distribution rate declines so that the protocol isn't like quote-unquote wasting tokens on a lot more capital than it actually needs.

And such that if the senior pool were to reach double its target, like $200 million, as it gets towards that $200 million target, the protocol decreases the rate to zero. So, in theory, if there were $200 million in the protocol, the protocol would just stop distributing GFI altogether. And so that would be like, oh, there's way more capital than necessary. It just saves tokens. So that is to try to get to kind of like a market equilibrium of like the right number of tokens to get the right number. Right amount of capital relative to, the target actually wants without wasting tokens for like a lot of excess capital that the protocol isn't putting to use. And so we're pretty excited about this general dynamic and launched it. And what we're seeing right now is before launching the token, there was $40 million in the senior pool.

More capital came in. $65 million are now in the senior pool. And the overall APYs there, including the token, are in the range of like 60, 70 to 80% APYs. And our early predictions were that the APYs would be lower than that, like that the capital would be seeking lower APYs, but we're seeing this kind like higher APYs. And so it's interesting. So we've definitely seen more capital come in, which has been, I think, overall like a success for it. But we're looking to see like how do the LPs sort of compare to the APYs here, to other protocols?

And one, like we've been basically trying to reach out to different LPs to understand how they think about it. And one thing that I found interesting is the way that it's implemented on Goldfinch is it has this sort of 12 months unlocked period. Which is that you, need to leave the capital in the protocol for 12 months before you can claim all of the sorts of tokens that you've sort of received through liquidity money which is different from other protocols. We think this is a direction that protocols are headed in in general, but it means it really needs to attract like longer-term capital into the protocol.

And so I like my personal hypothesis, like that was one of the reasons why the required APYs are so much higher. But then, when we were talking to some LPs who have supplied, they were saying we were like, oh, do you think this vesting thing, or like this unlock schedule is a problem? And they were like, it might be reducing the amount of capital that comes in now, but they were like, I like it. It makes me feel really good to know that, like all the capital in here is long-term oriented. I personally like that there's that vesting schedule. Because then I know that I'm participating in this sort of long-term mindset alongside other providers of capital that have that long-term mindset. I thought that was like,

Sina [01:24:02]: Yeah. More stable, I thought to build on.

Mike [01:24:04]: Yeah. So I thought that was pretty interesting, which is like,

Sina [01:24:06]: So wait, what does that mean? So if you withdraw like a month in, what happens?

Blake [01:24:12]: It means you forfeit that percentage of your rewards. So if you were to like withdraw halfway in, like six months in, like 50% of the rewards you would have earned are forfeited. If you withdraw nine months in,

Sina [01:24:26]: Got it.

Blake [01:24:26]: 25% of the rewards you would've earned are forfeited. If you withdraw 12 months in or anything after that, there's no forfeit.

Sina [01:24:33]: Got it.

Blake [01:24:34]: So yeah, you can withdraw at any point in time, and all your underlying capital is available to you. But the reward part is what will be reduced if you withdraw quicker than 12 months.

Sina [01:24:45]: Yeah. Makes sense.

Mike [01:24:47]: Yeah. My kind of like hypothesis as we've been talking to folks is that when it's implemented this way, what you see is kind of like a bit of like slower steadier growth, but like sort of much more stable capital that comes in. So like, for example, if you just recently with the markets declining, like pretty soon after the tokens launched the capital, very little capital was withdrawn from the senior pool Goldfinch. Even when like overall sort of TVL in the space fell quite a bit on other protocols. And so I think that kind of speaks to when there is this unlocked schedule, the capital coming in is slower said year, but also just more stable and more long-term oriented. And so, yeah, we kind of like saw that in action during that broader crypto market fall.

Sina [01:25:39]: Also, I mean, speaks to how Goldfinch is uncorrelated with the rest of crypto. Right. Because the borrowers aren't getting liquidated, but the price is falling. Like, if anything, it's making Goldfinch seem more appealing because you are getting yield from something else.

Mike [01:25:55]: Yeah. The base interest rates on the USDC haven't changed at all.

Sina [01:26:00]: Yeah. One question was how, yeah, how do you think about Goldfinch in the context of the potential like macro environments? Like all of the crazy stuff that's happening?

Blake [01:26:14]: One of the things Mike and I have talked about a lot is just seeing, when you talk about this broad macro-environment, is these kinds of like megatrends. When we think over the next ten years where like one is that you have, basically investors we think are just going to start demanding new sources of yield. I think you have like bond yields are going to be super low. I know that sucks about raising rates, but like, I don't think they're ever going to do that much. And when you combine it with inflation, I think the net yield you're going to get on bonds globally is still going to be like super low. And you're going to have all this capital, like looking for new ways to earn yield both in like institutions as well as individuals.

And so I think that trend is just going to continue, right. And technology, of course, is going to make it easier and easier for people to get access to yield too. So that's going to be one big trend. And we also see just a ton more economic activity starting to move on-chain. And we were kind of talking about that earlier. There are still a few kinds of missing pieces. So like for that, be really ready for prime time, but we feel pretty confident it's going to happen. And when you have all these economic activity moving on-chain, that means like all of those transactions are now programmable through the magic of the blockchain. And you have all these investors now looking for yield in new ways. That means that those investors could be like lending against these transactions that are happening on-chain. And basically, every transaction in the world is now programmable and can have a loan against it.

And so, when that's possible, we think like you're going to need a marketplace to organize that activity. That's really what we see Goldfinch fulfilling. It's fulfilling that decentralized marketplace to coordinate all this debt activity for all these lending transactions around the world. And we've even talked about that as being basically like an open-source [01:27:49 inaudible], right. All these transactions are happening. Smart contracts can be written with different strategies of how to lend out to different people and institutions. And people can be then doing wholesale lending into those smart contracts. When you have this full life cycle of funds on-chain from like, person borrowing and paying and their income is on-chain. And like, so there's a lot of pieces that are getting there, but like we think that's where this is headed.

Sina [01:28:10]: That's a really exciting and big vision. Super excited. Well, yeah. Thank you. We've been talking for a long time, so we can probably bring it to a close. I'm, yeah, super, super excited about Goldfinch, and we'll be following closely.

Blake [01:28:24]: Thanks so much, Sina. This was a ton of fun.

Mike [01:28:25]: Yeah. Thanks for having us on.