September 20, 2022
Rafael Cosman is the CEO & Co-founder of TrustToken, the core team building TrueFi into the protocol for global lending. TrueFi lets portfolio managers launch uncollateralized lending opportunities on-chain to a global base of lenders and borrowers. Since November 2020, has completed $1.5B of origination and +$1B of repayments with no defaults. Prior to TrustToken, Rafael helped build StreetCode, a non-profit that teaches East Palo Alto youth technical skills, and worked at Google Brain, Palantir, and Kernel. Rafael graduated from Stanford University with a Bachelor’s Degree in Computer Science. In his free time, you’ll find Rafael reading or surfing.
Julian: Hey everyone. Thank you so much for joining the behind company lines podcast. I'm here with Rafael Cosman CEO and co-founder of trust, token trust, token, freeze money to move to whatever it creates. The most value instantly its main project right now is, is true five, which lets portfolio managers launch uncollateralized lending opportunities on chain to a global base of lenders.
And. Rafael. Thank you so much for being on the show. I'm really excited to dive into not only your background, but also what you're doing with trust token. And just to jump on in, what were you doing before you started your company?
Rafael: Wow. Before I started my company, that was it's about six years ago. I was working at Google brain doing AI research.
And then before that I was at Stanford setting computer.
Julian: I love that. And, and describe to me this, this experience at Google brain. I chatted with another ex Googler. They were working on this, this whole other experimental project that I still don't understand. What, what were you learning with Google brain and what were you [00:01:00] researching that you can speak
Rafael: about?
yeah, so of course I can't share technical details, but we were doing research into reinforcement learning, which is, it's a type of machine learning. That's. You know, how can an agent in some kind of environment take the optimal actions to maximize some reward function that it's trying to achieve.
And we were specifically focusing on smart ways that you can balance exploration and exploitation mm-hmm , which is a very fundamental problem, both in machine learning, but also in life, something that all of us face, you know, should you go to that restaurant? You know, you love, or should you try something?
Yeah.
Julian: Yeah. That's, that's fascinating. I feel that that is my day to day issue is what to eat and making the, I think it's what is it? There's a phenomenon where it's like, you have too many choices and you're overloaded by the options and, and it, you know, it leads to the inability to choose, but less on that and more so on, on your background I [00:02:00] learned a little bit more about your experience and one experience I wanted to chat about was you.
Helping kids learn programming and giving them the opportunity to learn, you know, different languages. Tell, tell me about that experience. I'm very curious about anytime someone's giving back, especially when it terms of technology.
Rafael: Yeah. So Julian, this was at back when I was at Stanford as an undergrad, I helped to found a nonprofit called the street code academy, which is.
Basically like a boot camp or tech school in east Palo Alto, specifically focusing on some of the underprivileged youth that are growing up there and may not have access to the same kind of high quality tech education that I, and many other folks that went to Stanford, other great universities got access to so teaching them, you know, just programming languages, other tech skills.
Getting them into computer science programs, getting some of them internships at local tech [00:03:00] companies and really trying to make sure that every community in the bay area is getting the benefit of the huge tech boom, which has happened over the past decades. Not just some subset. I love that.
Julian: And describe to me, you know, what inspired you, you know, transitioning to trust token, what inspired you to to build trust token, and then now create the different products that, that are under the
Rafael: portfolio?
Well, I've always been interested in startups, you know, I thought about just staying at Google and pursuing the path of, you know, being a researcher and discovering full new technical things and stuff like that. And that's, that's very exciting to me and it very much calls to me, but I think, you know, for many people who are listening to the podcast, maybe thinking about a similar question, you know, is it better to, you know, take a stable job that you can be working on?
Very interesting work, get paid well versus start your own company. [00:04:00] And for me, my analysis was just, Hey, if I'm smart and young and I don't have a lot. Responsibilities, you know, I don't have kids, I don't have a wife and so on. Then that's a time in my life when I can take a lot of risks. And if I go try a startup for a couple of years, if it doesn't work, then worst case I can go work at a big company again and climb through the ranks and go on that path.
But if it does work as it has worked pretty well, then, you know, it opens up all kinds of opportunities for you.
Julian: I love that. What, what in particular around web three and, and blockchain technology. What got you excited about that? You know, you felt inclined to, to pursue the startup entrepreneurship because of some internal.
And I feel like that's, you know, time and time again, the answer that founders give me is, is like, there's just some kind of itch that you need to scratch. But in terms of blockchain, in particular, in trust token, why web three? What in their, what in that technology did you see that, that got you excited in.[00:05:00]
Rafael: so my former business partner and I really got into blockchain, full-time focused on it in 2017. I'd been following it for many years and actually first encountered it when I was an undergrad. It's at Stanford back in 20 13, 20 14, but it was in 2017 that. Crypto was growing massively. I mean, it's, it was nothing fair to what it is today, but at the time we thought, oh my gosh, the market cap of Ethereum, you know, is skyrocketing.
There's huge opportunities here and specifically the first major product that my company launched in crypto, which we saw is just a great startup opportunity is a product called true us. Which is a us dollar back stable coin. It's now over a billion dollars in circulation, and we just saw it as if, if folks listening are familiar with tether, you know, there was a, there's a currency [00:06:00] called tether.
It was growing very quickly. It's this Fiat back stable coin, but it really hadn't built a great product and hadn't built a lot of trust with its users. And so we just saw a huge opportu. Where we said, look, they've already proven out the market that there's a tremendous demand for this product. We just need to build a high quality version of it.
Really earn trust with users, build something that people are gonna trust and love, and that could easily grow to a billion dollars or more. And that proved to be trip. So that was our first major product in crypto. True. S D has been very successful. And then a lot of the stuff that we've done since then, such as true fi have really built upon that.
Julian: That's amazing. What, in one door, I have two questions. What how do you build trust within your product and in your coin in particular, and then two how do you build a high quality product? What goes into building you know, a higher quality currency.
Rafael: So it's different for every. [00:07:00] Our company is very focused on trust.
That's why it's literally called trust token and our two major products. And we've got true fi which I'm sure we'll talk about in a little bit, and then true us D which is our first major product. So a lot of trust and true in our product. And part of that is because you know, our company is a bit different than a lot of other crypto companies.
And I know you've had many crypto and blockchain C. On the podcast before, but you know, some crypto projects are pure technical. Like they say, Hey, we're going to make the sleekest fastest L one blockchain that has ever existed with some crazy technical innovations. And you could be a pure tech team and get a project like that done.
And I love that. I'm glad there are teams like that. And I think that that's. Probably the majority of the work in crypto is that kind of very highly technical stuff, which is great. What our company focuses on is [00:08:00] somewhat different. We've always had part as part of our DNA being this bridge between tra and DeFi, traditional finance and decentralized finance.
And we've found that there are very few folks that really know how to play at that intersection. And it's a place where we can differentiate. So true USD, the Fiat back stablecoin, it's a perfect example of that because you can't build. Fiat back stablecoin with a pure tech product. You need to actually work with banking partners.
You need to have, and know your customer policies in place that you can comply with regulation, you to accept millions of dollars of bank wires. There's all kinds of things that you have to do to interface with traditional finance and with the traditional legal system to make a product like that. Well, also having very solid technicals, such that your stablecoin can get integrated into DeFi protocols and exchanges and all kinds of other places.
So [00:09:00] it's really about being that bridge and creating that trust because some of these things like a fi pack stablecoin are difficult to make completely decentralized. So how do we avoid being like some of these other products, like tether that have had some issues with trust in the marketplace? How do we build something that is super transparent?
Super trustworthy. So that users know, Hey, even though this may have some centralized components to how it works, that's inevitable, but we still completely can rely on it as users.
Julian: That's awesome. I love this concept, that, or the product that you build Tru with this this idea about lending. I'm curious about because it's an uncollateralized uncollateralized lending platform.
How, how do you lend without collateral? I, I was talking to another founder, just a little anecdote and they had created a lending platform where they would stake their either NFT or. I think there was one company that did like a physical Rolex at a location [00:10:00] that then mentioned an NFT and then they state that to then lend how do you do it without collateral?
Rafael: Yeah. So and Julian, can I actually project my screen since we've got video? Go ahead. I think it might be interesting folks that, just some of this stuff. Yeah. Cool. So there have been several waves of innovation in Def. And one of the first ones was with over collateralized lending protocols. So can you see my screen?
Alright. Yep. So, you know, one of the main, one of the most famous over collateralized lending protocols and a protocol that really pioneered what DeFi is today is a protocol called compound. That I think many folks have heard about which we are. It's a protocols, a smart contract running on the Ethereum blockchain, where you can do over collateralized lending.
You can put up a dollar and 50 cents of Ethereum with a smart contract and then borrow a dollar of stable coin. Or you can put up a [00:11:00] dollar 50 cents of stable coin. And borrow a dollar of Ethereum. You always have to put up more collateral than you're borrowing, and that's what keeps the system safe. If you don't pay back that loan, they can always liquidate your collateral.
Does that make sense? Right? Yep. So that was kind of part of the first wave of DeFi protocols that came out compound. There's another one called Ave. It's pretty large. It's got about 10 billion of liquidity in it right now. It's a very significant. There's another one called maker. And so these are all major DeFi protocols, really you know, first wave V one of how lending works in DeFi.
Now it's a cool system, but this kind of over collateralized lending is only a very limited market. And if you look at the entire world of lending and how it's evolved, Actually millennia cause lending is one of the oldest forms of finance even before equity and [00:12:00] lots of other things were invented. This kind of over collateralized lending is just a very, very small market within all of lending and the vast majority of lending doesn't just take into account collateral.
It also takes into account the credit. Of a borrower. And that is sort of the second wave of DeFi lending protocols that our company and Tru really help to pioneer and define. So the idea there is a protocol like Tru fi you can put capital in, and there are portfolio managers who are experts in different areas, and they actually lend that capital.
Sometimes taking some collateral, sometimes taking no collateral, but none of the lending that happens Onfi is over collateralized. The way that compound and Ave are over collateralized. It really does rely on high quality underwriting to make sure that most or all borrowers are gonna be paying back those loans.
And that the lenders who are putting their capital at risk are [00:13:00] ultimately gonna achieve a good risk adjusted. so if I can actually just show you briefly how that works. So this is our site app dot TruFit IO. I hope everyone's listening can go play around with this later and get a sense of it for yourself.
So this is the first and still one of the largest uncollateralized lending protocols in DeFi. We launched it about two years ago in November of 2020 at a time. Uncollateralized lending on the blockchain seemed like a crazy idea. Now it is very well accepted. It's very quickly growing. It's one of the most exciting hottest areas of DeFi.
And we think it's really gonna push DeFi forward in the last two years. As you can see here, we originated just over $1.7 billion of lungs. So this has been a very successful model with that being said this is still a startup. These are still early days. Our ambition was not just to get to 1 billion.
We [00:14:00] want to get to 10 billion. We wanna get to a hundred billion. We eventually want to get to a trillion dollars because the global lending market really is that large. And we think that this technology is a fundamentally better way to do lending. And so we think that a generalized lending protocol, like what we have with Tru.
mm-hmm, unlike some of the more niche over collateralized lending protocols like compound and Ave and maker, a generalized lending protocol, like Tru that can take into account credit and collateral can really address the entire global lending market where at least a very large fraction of it and make all of it more transparent, more open and more efficient.
So that's ultimately where we're. We've got a very good start. We've got several years of lending under our belt. Very, very strong track record and good results so far, but hopefully several orders of [00:15:00] magnitude of growth ahead of us in the coming years, as we go from just, you know, a cool, exciting, fast growing product to truly changing how finance works at a global.
That's incredible. How, how
Julian: does credit work? You, you describe, you know, credit score as a main function, credit score and collateral really giving you the ability to borrow from, you know, the lenders that are on the platform. How does that work? How is that calculated? And what are the benefits of having now a credit score on, on the, on chain?
And does it communicate with your score off
Rafael: chain? Good question. Okay. So the way that credit works. In the true five protocol. And in most other uncollateralized lending protocols, these days is you've got portfolio managers who are experts in a certain area, and they do all the underwriting for loans in that area.
And it's a tricky problem because there's so many different areas of lending. We [00:16:00] can't be experts in all of them. We have a platform and we bring on folks that are experts in these different areas. So a couple of example here is a portfolio that recently launched with this group called cars that does lending with financial technology companies, FinTech companies in emerging markets.
And they really know emerging market lending and emerging market FinTech. And so they are very well qualified to deploy capital in that area. You know, there's many other types of portfolios with different strategies. We have a portfolio. That's actually in partnership with another crypto protocol called woo.
That lends to folks in the woo ecosystem. We've got portfolios, some of our largest portfolios that lend to crypto funds, crypto market makers, folks like winter mute nibi Alameda research. And so on. We also have portfolios [00:17:00] that actually just have one borrower. And where you can just deploy capital with a single borrower that you may have a very strong brand name that people really wanna work with.
And so here's an example of one of those. This is a seven and a half million dollar portfolio that is exclusively lending to Alameda research. And so we really have a variety of different opportunities that where you can deploy capital on Tru fi. And our goal is. You know, provide high quality infrastructure, a high quality platform that can support a whole variety of different things that lenders may want to deploy capital into that that's really our model.
Julian: Yeah, I know that that's incredible. And, and back back to the, the credit score question, how is, are the underwriters then evaluating your company and your ability to, to borrow and then giving you a score. And does that score build over time based on your you know, your, your repayment on that initial [00:18:00] principle or is it kind of net zero and then starts to build up at any given time?
I'm curious about how that initial score is. Is it mirror traditional finance or is it something that's slightly
Rafael: different? Good question. So a lot of those things do mirror, traditional finance mm-hmm in that, you know, people will be building up a larger and larger credit record and credit score over time.
Be able to. Borrow more capital with more flexible terms. But the big difference with traditional finance is how much more transparent it is. And if you look at part of what's happening recently in crypto, we've had some major implosions companies like Celsius, like three arrows that have either defaulted or are not able to repay customer funds and are going through some significant issues.
You know, part of the issue there is that users who users or businesses who [00:19:00] were lending capital to some of these firms they didn't have transparency about what risks those firms were taking with the capital that users were providing. And, you know, there's, every lending business is gonna have some risk, right?
There's no such thing as lending or invest. Without having a risk of losing some or all of your capital. But part of the point of DeFi is, Hey, we can do finance in a way that is much more transparent about what those risks are. So let me just show you as an example with Tru fi just like with many DeFi protocols, it's completely transparent.
You can see every single loan that has ever been made by the true five protocol. All 1.7 billion. You can see all the terms of those loans. The protocol was founded two years ago. So here's the actual analytics page app. TruFit IO slash loans. You can go check this out yourself and all this data it's public on the Ethereum blockchain.
So you don't really have to trust me. You don't have to trust our [00:20:00] website. This is completely transparent. So you can see we've got a bunch of loans that are active right now. And, and you can just scroll through here and just. You know, dozens and dozens and dozens of loans that have been repaid and you know exactly.
When was the loan applied for? When did it go out? When was it repaid? Who borrowed it? What was the rate? What was the term? What was the amount? What was the currency? Everything you can see all this data completely on chain. And part of what's interesting about this is as you've pointed out, the borrowers are building up more and more credit data.
That's public on chain as they borrow. Yeah. So here's just as an example, here's Alameda research one of our first and largest borrowers, as well as. One of the main investors in our company, they've been a great partner to help us grow. You can see that they started their first loan. Well, this is a little test loan of one true USD, just at the very beginning.
Their first real loan was 3.5 million for 30 days, [00:21:00] right? At 11.5% interest rate, they repaid that they got a loan of $5 million, then 6.5, then 10, 10, 10. and you can see as they grew, they pretty soon got access to at some point being able to do even 50 or 70 or $80 million loans with the protocol.
And that's how many of our borrowers go is that, you know, they'll start doing small loans and they'll big build up a large track record over time. And for Alameda research, they can now show, Hey, they have repaid a total of $484 million of. Loans to the protocol, you know, over dozen and dozens of loans.
Yeah. This is all completely public, completely transparent. It's on chain. So it's not, it's not them saying, Hey, you should trust us. We are creditworthy. Or even at credit bureau is saying, Hey, you should trust these folks. Their credit worthy. You can actually go on chain and you [00:22:00] can literally see the loans that they've been taking for years.
So that's a very significant track record and the, and that they've been repaying those on time and with interest every single time. So that's just a level of transparency that we've never had in traditional finance, nothing like this to my knowledge has ever existed where you can, someone can directly prove their entire credit history, and you're not even trusting a third party to report it.
So, so Julian, I would say that just like many things. Crypto. And in DeFi, there are many things that are analogs of what's happening in traditional finance. Yeah. You know, credit is always gonna work in certain ways. You know, you have, you have to build up history, you know, show data for why you're trustworthy, et cetera, etcetera.
Some things are not changing, but these transparency and the openness absolutely is. And that is where I think crypto and DeFi [00:23:00] has a tremendous amount to.
Julian: That's incredible. I, I can see so many just improvements to the, the credit process because, you know, there was a huge, you know, housing crisis in 2008 and, and with parts of the recession where people have to default on loans and it affects their credit score and it takes, you hear countless stories of it takes them an incredible amount of time, almost an, an absurd amount of time for them to rebuild after so much history.
And there is that lack of transparency. There is that lack of trust. They can't see through to, to what's going on behind the scenes. And now with the ability of QFI, it's incredible to see that other lenders can identify those, those individuals and look through their history and really make a decision based on actual data.
And, and without like this overarching. A shadow score that is like the wizard of Oz. You know, you don't really know what's going on until you uncover behind the curtain, but now you're able to see everything. And that's, that's incredible to see. And, and it seems as though, and I'm curious on your take on this.
What does it [00:24:00] have in effect on, in terms of the rates that that, that you'll be seeing as the market grows and progresses? Because I think with a lot of lending, especially, you know, the fed. the, the interest rate on, on mortgages. And so a lot of people are, you know, purchasing houses less so because of the increase in that rate does this offer a more fair marketplace for rates to be at a position for the community to actually borrow money?
How do you view the, the effect on rates in this
Rafael: ecosystem? Good question. So there are a lot of things that are affecting rates. you know, as borrowers build up more history, their rates tend to go down. Yeah. They can borrow at longer term and at larger size, but as the macro environment changes and the fed raises interest rates that is gonna affect crypto when we have been seeing [00:25:00] interest rates affected in crypto as well.
And most importantly many of the defaults And market issues within the crypto lending world right now are also causing rates to rise. So there's a lot of factors. Yeah. But you can actually just, you can literally go and look at some of the, the recent loans that are being originated on Tru fi on many other protocols and you can watch rates going up and down in real time in response to macroeconomic changes.
Yeah. Crypto market changes and so on. And Julian, just to respond to what you were. About credit and how things are developing. We do actually see folks from the traditional credit world getting very, very interested in using this kind of technology. And we've recently been talking with S and P wow. The company that makes the S and P 500.
Yeah. They do bond ratings in just like Moody's and many other folks. Right, right. They recently did their first [00:26:00] rating in crypto. Of compound treasury and they're interested in expanding that business. And that's something that we're talking to them about is ways that we can help them. And other folks that do ratings in tra to come and rate some of the products that are in crypto.
That's the kind of partnership that we're very excited about, where we can bring some of the things that have been developed in tra and give consumers and businesses confide. When they're investing capital in tra bring that over to crypto and start to create that same sort of confidence that they are high quality products here.
Cause we really think that there are, but right now, you know, some of these issues around credit in crypto are still, you know, creating a lot of fun in the market and making people hesitant to come to deploy capital, unless they're really comfortable with the technology.
Julian: Yeah. Incredible. What's some of the biggest risks that, that you face that your company faces.
Rafael: risks that you face. We don't
Julian: have any risks. No. Is that
Rafael: right? [00:27:00] so wow. A bunch. A big one is one that, the one that I just mentioned about how to bridge this gap to Tradify institutions, a lot of our growth. So far, the 1.7 billion of loans we've originated in the last, approximately 24 months since we launched the protocol.
A lot of that has been within crypto, you know, crypto borrowers, crypto lenders, you know, who, who understand crypto protocols and are comfortable with this kind of technology providing capital. And that has been our beach head market where we've gotten a lot of growth so far, but that's not where we think that the next billions of dollars of growth for us are gonna come from.
You know, we think we're, we're always gonna have some foothold in that market. And we think it's an important market and it's gonna continue to grow as crypto grows. But for crypto and DeFi to really grow, we do need to build products where [00:28:00] traditional investors, traditional institutions are gonna feel comfortable deploying capital.
And that is where a lot of our focus has been over the last year or so. And we've been getting. Good traction and good response there. And that is one of our biggest things we're focusing on going forward. Yeah. But it is definitely not easy because there's just so many, you know, marketing and compliance and reputational hurdles sure.
For crypto and blockchain to get over before it can get more mainstream acceptance. Yeah.
Julian: Yeah. What's the long term vision for trust, token and Tru and, and all the technologies that you're. What, what are you most excited about? In
Rafael: the future, our long term vision is to build the protocol for global lending and to change how a lot of finance works to make it more open and more transparent.
That's where we wanna [00:29:00] get. And that means that we get to the point where, you know, you may not even realize it in the future, but you'll be using like a banking. and under the hood, true fi and other DeFi protocols are gonna be plugged in. Yeah. That's where we think that we're gonna be. So we don't necessarily need true fi to become the biggest consumer brand that everyone knows about.
But we think that it's has the potential to become very critical, very central infrastructure for how a lot of global finance works. And that's really where we see DeFi providing the most.
Julian: that's incredible. I'm excited to see where the technology goes and, and how it becomes more foundational in not only the consumer products that we, as, you know, regular consumers.
And it sounds like businesses are already taken advantage of it, but it's exciting to see that it's gonna have this effect. That's gonna be, you know a really. Transparent, trustworthy, and also give everyone the accessibility bonus question. I like [00:30:00] to ask all my guests for, for research on my, on my end.
But also for the community here that listens what books and what people influence or what people influence you the most.
Rafael: Oh, good question. Okay. So let me start with books. That's an easy one. So getting things done by David Allen is like my, I think that's the Bible of productivity. I've never found a better productivity book than that.
It's so fundamental. There's a book called the 15 commitments of conscious leadership. Hmm. Which is just an amazing book about how to live and lead. A company in a very thoughtful, conscious, productive way and build a great culture. So that would be second on my list. Those are probably the two, two biggest ones in entrepreneurship and productivity books.
I also read a lot of fiction. We can save that for later. [00:31:00] Cool. And then people in crypto, specifically, there is an investor named Ari, Paul mm-hmm , who's one of our investors that I just, I met years ago and I've always thought is a brilliant thinker. He's got a phenomenal Twitter. And so if you're looking for just some of the best thinking about how to work, how to invest, how to exist in the crypto world, Ari Paul is definitely the guy.
And. In the broader startup world, you know, it's classic, but Paul Graham. Yeah. I mean, his, I just, I grew up with his essays on startups. I feel like those were just like my bread and butter. Yeah. And if you're a, if you're a young entrepreneur, you know, just read everything that Paul Graham wrote, read all of his essays and that's just like, that's the bedrock of how startups were.
And there's a lot. Other stuff out [00:32:00] there. That's great to learn, but he just, man, he just hits the nail in the head with some of those things.
Julian: Love that. Love that. Rafael, thank you so much for, for being on the show. I'm excited to see where trust token goes and, and really excited about you know, researching the, the information that you you sent.
Last little plug. I always like to ask my guests, where can we find you? Where can we support trust, token and yourself? Give us your LinkedIn LinkedIns, your Twitters, your your all, all, all your handles.
Rafael: you got it. Okay. So my Twitter is at Rafael Cosman R a F a E L C O S M a N N. And our Tru Twitter is at Tru fi da T R U E F I D a O.
You can find all of the Tru links. TruFit IO. I would love for folks to come join our discord, get involved in our community. Part of what's exciting about DeFi protocols is that if they're decentralized as Tru fi is, then they really are community governed. [00:33:00] And so people that participate in the protocol can literally shape the future of where the protocol is going.
What get, what gets built, what gets shipped what ultimately is gonna go to users and affect the. So, if you're interested in being a part, both of our community or of our team, please go engage with us on discord. Let us know your thoughts, let us know your ideas. We would love to chat with you.
Julian: Awesome. Rafael, thank you so much again for being on the show. I hope to catch you next time.
Rafael: Thank you, Julian. It was a pleasure.