DeFi Builder Andre Cronje Isn't Going Anywhere (For Now): "Innovation Can't Stop Because of Fear"
Andre talks to The Defiant about the road to building yearn and YFI, and how he never anticipated the project's wild success.
Hello Defiers! This week’s podcast episode is with Andre Cronje, the developer of the yearn platform and YFI token. There’s this sort of mysticism around Andre: The solo genius builder who relentlessly releases code —even if it hasn’t always been properly audited or tested. The DeFi wizard capable of turning a token which he himself claimed to be valueless to be worth thousands of dollars in a few days.
Andre created yearn initially to invest his family’s savings. The platform allocates stablecoins to lending protocols that are generating the most yield. Last month, he released YFI, a token which had no pre-sale, no offering, and could only be earned in the initial distribution by using the yearn platform. With traders lining up to get their hands on DeFi tokens, more than $300M flowed into yearn, while YFI surged past $1,000 in a day and now trades at over $4,000. This isn’t exactly how Andre thought things would play out; his napkin math for what YFI would be worth had put its value between $2 or $3. The inflow of capital was so huge and unexpected, that it overpowered his own share in the platform, leaving him with just over 1 YFI token —but he’s not complaining and says he’s returning donations he recently got.
He talks about the actual paralyzing fear that comes with building money protocol, the huge responsibility he feels, and how finger-pointing has caused him to quit the space before. But he says the drive to continue innovating has helped him overcome that fear and keep building. He recognizes innovation and user safety are at odds in DeFi, pulling in opposite directions, and he’s not sure what the right balance is. At least on his end, innovation will keep winning that tug of war. But it’s not like he has some grand vision for DeFi. He says what drives him is waking up in the morning and being really excited to code and “add new things to this ecosystem.” As long as that holds true, he says, he’s not going anywhere.
You’re a paid subscriber, which means you get the full transcript below. Subscribers also get exclusive access to The Defiant’s Discord chat for the community, here’s a new link to join.
🎙Listen to the interview in this week’s podcast episode here:
🙌 Together with DeversiFi, a professional-grade, self-custodial exchange.
CR: Andre Cronje, he's been leading the charge in innovation in DeFi recently. He's a longtime developer and the founder of yearn, previously iEarn, we'll get into all of that and the YFI token and all the craziness around it. But first, I really want to start and just get to know Andre better. Andre, welcome to The Defiant podcast. Thank you so much for taking the time to talk to me.
AC: It's a pleasure. Thanks for inviting me. Thanks for giving me a forum to talk about a lot of this stuff. I like to preface by people like saying I'm the founder or whatever, I just like to preface by saying that I'm not. A big reason why I'm trying to remove myself, like the thing I messed up the biggest, sorry, this is actually a point that's been annoying me for a while, so just a good forum to say it.
Irony of a DeFi Founder
The thing I messed up the biggest was not being anonymous. I realized one of the key things about building something decentralized is being able to walk away from it. And at this point, people associate protocols so much with me, that it's almost difficult to detach the two. That's actually in the current narrative really annoying me, because you can't be decentralized and the founder, those two just don't go hand in hand. But anyway, we can unpack that a little bit later. It just popped into my brain for whatever reason now and it's been a bit beef.
“The thing I messed up the biggest was not being anonymous.”
CR: What's the right way to call you? What's the title you refer then?
AC: I am a guy that deploys smart contracts on Ethereum. That's my claim to fame that I've deployed a few lines of code onto a very well-designed, cool blockchain.
CR: You're being a little bit humble, but I’ll go with it. Let's start with your background and then I'd love to begin to the topics you mentioned just now, which is this aim of DeFi founders to really detach themselves from their protocols. I think is super interesting. But before we get into that, let's get to know you a little bit better. I know you're based In South Africa and you have Crypto Briefing also, but I'd love to kind of learn how you got into crypto in the first place. What were you doing pre-crypto?
AC: Well, how far do you want to go back? I'm one of the older specimens in crypto, I think. We could go back 20-30 years if you like or we can keep it crypto recent, so I don't know how far back you want to dig.
CR: Whatever is relevant to understand your enthusiasm and activity in this space.
AC: Well, then I'll brush through the sort of before crypto very quickly. Started off doing my LLB, originally to become an advocate, took a year off in that year, helped a friend was doing computer science, got really interested in the courses they were doing, so I ended up doing them myself.
I managed to finish the courses quite a bit ahead of schedule. One of the lecturers had resigned at the time, so they not being able to find someone to replace to fill the rest of the semester, they ended up offering me the interim job to actually be a lecturer so I took that. From there, started getting heavily into computer science. First job was with Vodacom in Tanzania and Congo. Vodacom is a telecommunications company, Vodafone, holding their GGSN, all of those things.
Stuck in Fintech
After Vodacom got heavily into mobile security, but this is back in the days of like J2ME and Nokia, Symbian development, it’s like Android and iPhone didn't even exist yet. That was card switching payments, so just back in payment systems for traditional credit cards and those kinds of things. From there into big data, Spark cluster development, lots of scholar work, very high throughput stuff that we were doing with again the telecommunications companies. Got a little bit into ML, Machine Learning, just for building some, like regression models and things, because we started going into loans and insurance with Shoprite Group, which is a big retail company here. Stuck there in the fintech for probably the three, four years leading up that technology department.
Then there was this tipping point where I was stagnating in that industry a little bit, there wasn't much more I could learn or do and I do get bored very easily. I don't manage to stay at companies long. I'm highly addicted while there's a puzzle, but when the puzzle has been solved, I very quickly get bored.
“I'm highly addicted while there's a puzzle, but when the puzzle has been solved, I very quickly get bored.”
At this time, my one mate, Antonio, who I've been working with for the last decade plus, he's a phenomenal human being, but he got married and he went on honeymoon. And I was alone at the office and I was bored and this was December and I was looking for something new to do. And this was, I think, either in 2016 or in 2017, I don't remember and that probably in 2017, because that's when the initial big hype cycle on crypto started. And I had been doing mesh networks and distributed systems previously with my Vodacom stuff and my machine learning stuff.
I got very excited originally on the distributed consensus side. One of the projects we were busy with at the time was localized decentralized consensus that we wanted to do for self-driving cars. Because we had this research theory at the time where self-driving cars right now, they still use normal optics, so they'll use a stop sign and then they'll stop. But what we wanted to accomplish is if you have a bunch of cars coming to an intersection, there's no reason for them to use an archaic system like a stop sign if they can communicate to each other in a localized network at high frequency, arrive at the decision even though there might be malicious actors.
This idea of high throughput and this was that phase when everyone was promising 10,000 TPS and the next one at 100,000 and the next one at a bajillion and like it just kept scaling like that. Coming from a fairly traditional like research and technology background, I started getting very excited about this, because here was this field that were promising that solves all of these problems and then that was phenomenal to me. Because on that research side, we've been stuck for 20 years without breakthroughs and all of a sudden, you've got 19-year-olds fresh out of school saying that they came up with a solution and they solved it. And at the time, I was hooked, I loved the idea, so I started digging into it more.
CR: And at this time, this was with blockchains like Dfinity promising that that level of scalability right at this time, like 2017, I imagine is that what you were looking at and were hopeful about?
Writing Code Reviews
AC: Looking back at the reviews I did, I think it was probably like Wanchain. Dfinity, I think was a little bit off to that, maybe a few months after. It was still a little bit before that. But so to learn, I started digging through their code and I started documenting what I saw and what I learned from it and I wrote that on my Medium. And at the time, it was exploratory for me, I wanted to learn more, so I figured there's probably other people that are going through the same thing I'm going, so I'll write it down. And the code reviews blew up.
For whatever reason, it became wildly used and then Han from Crypto Briefing reached out to me and he had the idea of formalizing it with their reviews, which the guys there are awesome and I love them, so I was like, okay, let's do it. I started giving them the rights to, I say rights, all of my stuff is open, anyone can use it. There's no rights, but I mean, they were the first ones to publish it on their Crypto Briefing website, CryptoBriefing.com.
Image source: Crypto Briefings
What I didn't like is, I wrote for people to learn, but instead, people started using it as a measuring stick for ICO qualities and like as investment advice and that really pissed me off. I started pulling a little bit back on that, because while I was doing it and what people were using it for were two very different things. I pulled back from my reviews and this was especially during that mega high period and I mean, people were doing private sales off of my reviews and things like that. And it really annoyed me so I almost completely cold stopped that.
“What I didn't like is, I wrote for people to learn, but instead, people started using it as a measuring stick for ICO qualities and like as investment advice and that really pissed me off.”
Then I went back to my roots, which was on the research side. Now, at this time, it’s I think about a year, maybe a little bit more later and I've come to realize that these really difficult problems have not been solved yet. But what's fantastic about this space is that there's so much money being thrown at research. In the traditional space, you have to go through so many hoops to try and get $100 grant to do something. Here, you write the same idea on a piece of paper, call it a whitepaper and then all of a sudden, you can raise $40 million, which is insane. But it created the opportunity for a lot of good research to happen.
“In the traditional space, you have to go through so many hoops to try and get $100 grant to do something. Here, you write the same idea on a piece of paper, call it a whitepaper and then all of a sudden, you can raise $40 million, which is insane.”
I started digging in heavily on that side. I got very involved in the asynchronous BFT research space, did a few papers there with a few different guys in the space, also did some formalized VM stuff with the guys in Sydney. Got really excited about all of that stuff, helped launch one or two of these blockchains to test in a distributed capacity and the results were good, but not as high as theoretically possible. That was probably about the middle area of 2019.
At that time, what was happening parallel to this was I'm now that guy, I'm the one that's in crypto and I'm the one that's telling people about crypto, so I'm that annoying douche that no one likes. But the counter effect there is all of my friends, all of my family, they're a little bit scared of this crypto stuff, but they want to test the waters, they want to dip their toe, but they don't want to figure it out themselves. All of them started asking me to invest like 1,000 bucks or 500 bucks or whatever it is, never big money, into like BTC or ETH and just hold that stuff there.
Now, during the 2019 volatility, I wasn't comfortable keeping their things in BTC or ETH or any of these other crypto assets that are a little bit more volatile, because I am not a trader, I don't understand the TA-FA side, I still think it's flipping a coin and guessing, I think it's formalized gambling. I wanted to get out of that as much as possible.
I started moving all of that stuff to stablecoins. Now, I've got this stack of stablecoins and I'm like, okay, but I still want to show them there's value here. It's not a 5X, but you can still make more than you're going to make by putting that money in a savings account that's going to give you maybe 1-4% if you're lucky, depending on your lockups here in South Africa.
“I started moving all of that stuff to stablecoins. Now, I've got this stack of stablecoins and I'm like, okay, but I still want to show them there's value here.”
At that time, I started doing a lot of sort of initial research, DeFi protocol work, but not yet on Ethereum. The one problem I still had a little bit on the Ethereum ecosystem was that there was this fight for liquidity and it's still happening, it's Uniswap is trying to capture it or Bancor trying to capture it or the DEXs are trying to capture it, but everyone wants the liquidity.
I had formalized this additional base layer DeFi protocol where the liquidity is shared between your AMMs and your lenders and these kinds of things and started building that a little bit off to the side. But the more I built it, the more I realized the power of DeFi isn't the tooling, it's not the lenders, it's not the DEXs; it's the assets. The thing that was huge for Ethereum during that time was that it had launched so much financial value in ERC20s, in these underlying assets, and it's the chain that has all of these assets that has the value. It's all about the AUM.
“The power of DeFi isn't the tooling, it's not the lenders, it's not the DEXs; it's the assets.”
I think it was around December 2019 or Jan, this year, I started shifting all of the theoretical stuff I was doing in terms of shared liquidities and AMMs and lending providers, all this stuff onto the Ethereum ecosystem, but I didn't have to do most of it, because we already had lenders there was dYdX, there was Compound, all these wonderful things. AMMs weren't that big, we basically had Uniswap at the time.
What I started to do —now, my belief system when it comes to coding is never code something you can't do yourself. You as a human need to be able to do steps and accomplish something and then I can code it, because that's the only way I can teach the program to do something. It has to do what I can do.
“… my belief system when it comes to coding, is never code something you can't do yourself.”
Now this is where the two stories tie in. I started using the stablecoin portfolio from friends and family and I started moving it between lenders to make sure that it gets the highest APR rates. But this is time-consuming and takes a lot of gas. I'm a coder. We'll spend a month trying to figure out how to do something to automate it that actually only takes us five minutes a day to do. But in our brains, it makes logical sense and we want to do it.
All APRs in One Place
I started designing my on-chain oracles. Originally, it was just so that I have a single place where I can see all of the APRs on-chain. That came with its own nightmares, because the one reports APY, the other one reports current utilization, so there's a lot of normalization and things you have to do there, but that's a different story.
But after trying to get this APR going, I'm like, “okay, but now I want smart contracts to automate this for me.” They can move my funds around without me needing to do it. I ended up deploying sort of the V1 iEarn system that the strategies that would check what the APR is and it could move the thing.
The thing about blockchain is it only does something when you interact with it. I can’t say in 10 blocks, do something, because it doesn't have that concept. What I realized is I need more people to interact with the protocol. Because the more people that interact with the protocol, the more often it checks where the highest APR is, which means the more often it's going to move, which means that increases its frequency of return.
That's why I opened it up so that people can use it, because the more people that interact, it doesn't subtract anything from me and it means the protocol itself becomes more efficient. Because every time someone deposits, now it checks hey, where is the highest rates? Oh, no, it actually changed, I should move my funds; versus if it was just me, I was maybe once or twice a day clicking on the rebalance button so that it checks and then moves. That's originally why I opened it up.
Then before that, during the V2 I was building at the time, that's just when Curve Finance launched. Their original, the very first pool was cDAI/cUSDC, just the two Compound ones. I started talking to them about the idea of using the now yearn protocol to do profit lending switching, because it's the same concept, it just adds a little bit of extra yield on top of it.
We ended up building the Ypool that even though everything blew up now with the YFI, I don’t know how people want to pronounce it nowadays, YFI token, I still think that was sort of the crown jewel moment was that yPool that we created because here you had the yCurve, the output token from that pool is something that totally, I'm very proud of. Because you have this token that's stable by representation. It's a basket of these four different stablecoins that get automatically adjusted as people trade in the market, so the basket itself is as close adjusted to one as possible, which means you have a lot more stability on that side as well. Underneath it, you've got the yearn protocol that's doing this profit switching.
This yCurve token to me is a new kind of stable, integrated, savings account. In traditional financial account, you need a savings account to put money in. This is money that has those properties inherently, which is something I was very excited about.
“This yCurve token to me is a new kind of stable, integrated, savings account. In traditional financial account, you need a savings account to put money in. This is money that has those properties inherently, which is something I was very excited about.”
CR: You've covered a lot in this space, so let's stop here and just clarify.
AC: Curve is 100% Mitch and team from Curve Finance, we just collaborated.
CR: Separately, they had launched their DEXs. And so, can you explain more about what iEarn was doing and how you complimented that with Curve? Because so you had invested your family's stablecoins in this program that was good at finding the best APRs across different DeFi lending protocols. That's what iEarn did at this point, right?
AC: Yes, that's the base protocol.
CR: And then separately, Curve had launched and they had optimized a way to swap stablecoins? That's kind of what their edge is versus other AMMs and DEXs. What you did there was create on Curve a basket of stablecoins to make this system of optimizing for APRs easily, is that kind of a good way to represent it?
yearn — Looking Under the Hood
AC: If you break it down, you think of Curve’s just like you would any of the other AMMs. We think of Uniswap and we think V1 Uniswap. And let's say you want to do DAI, you would do DAI/ETH, so you'd have 50% DAI, 50% ETH. Now, Curve does that as well. But you wouldn't have ETH, you would have DAI/USDT. In this specific case, you would have DAI, USDT, USDC and TUSD. Now, that's what the user sees, that's on top.
But now, if you look under the hood, then what's actually in that pool is not USDT, DAI, USDC and TUSD, it's actually the output tokens from iEarn, so it's actually yDAI/yUSDC/yUSDT/yTUSD. When I put in one DAI into iEarn, I get out a share value of yDAI. I say share value, because as the pool grows, one yDAI becomes more expensive algorithmically. Now it's $1.3, where I launched it, it was exactly 1:1, because there was an interest in the underlying pool.
That's just a share calculation so that when you put in money now, you can't take someone else's interest. You’ll start accumulating from that point so you need that conversion ratio. But so, the thing about the yDAI token is it's always redeemable for that underlying DAI. From an architectural point of view, if you look at the curve tool, that curve pool is actually holding yDAI/yUSDC/yUSDT/yTUSD.
Image source: yearn.finance
CR: And they are all shares of pools in iEarn?
AC: Yes. In iEarn, although for purposes of the conversation, I'm just going to start sticking to yearn. So, in yearn, you've got a TUSD pool, where if you deposit TUSD, you get yTUSD. The reason that I always build all of my pools asset to asset, because I never want to give exposure or risk to the user depositing that that might change. My very first design would actually switch between stablecoins for you to have the highest yield one.
If you deposited USDT, but it's so DAI was 20% higher, it would actually switch it over to DAI and invest the DAI so that you get that yield then instead of the yield on the USDT. But then you have to start doing calculations around slippage and trade loss and those kinds of things. So, I made a choice at that time that I'm never exposing depositors to that risk. If you put in the one rule to everything I'm building, is if you put in one DAI, you might not get back more than one DAI, but you will always get back at least one DAI. For whatever reason, you might not get an interest, but you will never get less than what you put in, because I wouldn't want that to happen so I'm not going to let that happen to our people. In yearn, I put one DAI I get 0.99 yDAI and it's the yDAI that actually goes into the Curve pool.
CR: In the Curve pool, what's the benefit of having all these kind of derivatives of stablecoins bunched together?
Bundled Stablecoin Derivates
AC: The major thing that Mitch’s team solved was and that's why they're called Curve is because of this bonding curve variant that they use, whereby Mitch had this idea of adding a single assumption to a Uniswap pool, that one of the one side is equal to one of the other side. This is the big in “assumption” that the Curve pools make. And that's why Curve pools are only for things that are 1:1. So that's why there's a pool that has renBTC and WBTC and probably ZBTC and stuff later and SBTC in it, because it assumes one BTC is equal to one BTC is equal to one BTC.
It's this assumption that makes that trade that you do between two different stablecoins so incredibly efficient. Because now in something like Uniswap, if you were to trade DAI to USDC, and you were to do 50% of the pool, XY Curves that they do, you would lose a large amount to slippage. But with this bonding Curve design that Mitch uses and with the assumption that it's one, it's super-efficient. What that means is the net effect of that is that the underlying capital actually becomes more efficient.
Now on Uniswap where you'd need $10 million worth of USDT, now you only need $1 million to do the same amount of transactional volume and trades with less slippage. I keep comparing to Uniswap just because like I mean, it's in its own right an absolutely phenomenal protocol, but it's for more volatile assets.
Like Curve, you can't use for volatile assets. If you put volatile assets in there, you will always be able to manipulate that against the rest of the market because the spread will be so much tighter between those two assets that are trading. But it's phenomenal for things that are stable. If you're comparing synthetic ETH with ET, or SBTC with renBTC, whatever, that's great. And that was the big breakthrough.
The awesome thing about Curve is now with a lot smaller amount of capital that you put in, you can get a lot more trades happening and the more trades you get, the better trade fees you get. There was nothing like this on the market at the time when Mitch released it. Remembering those first few days, the volume would be 10-20 times the amount of capital in the pool, which means it was getting 90-100% APR on any given day. The beauty of this is your risk exposure is still in that basket of stablecoins, which still keeps it in “relatively safe”.
“the volume would be 10-20 times the amount of capital in the pool, which means it was getting 90-100% APR on any given day.”
CR: This was just because people were attracted to the efficiency of the low slippage of trading stablecoins on Curve that they were using this pool to trade and because of that, you're getting these fees, which meant you were getting a huge return on your deposits?
AC: Correct. I remember in those early days, we used to track it against Binance stablecoin volatility, because when there was any kind of volatility on Binance, people would withdraw their USDC or USDT there, trade it on Curve and then put it back into Binance. We used to get such big spikes from people doing that and that's phenomenal, because it's more capital efficient. Which means now again, you can't do a $3 million stablecoin trade that keeps to its peg on Binance, you're going to destroy the order book. But within the first month, we saw trades as big as that on Curve with almost no slippage.
CR: At this point, and were you still just taking your family's deposits or was this already open to anyone who wanted to use it for?
AC: Yes. By now, it was open to everyone. The website front ends and stuff was up. We heavily advertised APR at that time, which in hindsight actually, I regret, I've learned my lesson there. We attracted for the time so much attention.
“We heavily advertised APR at that time, which in hindsight actually, I regret, I've learned my lesson there.”
CR: When was this?
AC: This was Feb-March, before the real DeFi boom hype, whatever you want to call it that happened. DEXs didn't really see volume then. I mean, if we did like a million dollars in a day, that was huge. It was awesome to see. I actually can't believe how much it's changed from then to here in such an incredibly short amount of time. This is an evolution you see in in traditional industries over two to five years, not in six months.
CR: It has been crazy. Definitely want to get into this evolution of DeFi. But I want to address something you mentioned which is you regret advertising the rates, why is that?
AC: One of the big reasons why I like DeFi is because it, in my mind was, and I might be naive here and I accept that, but I came from traditional banking where your money is not yours. When you put your money in that savings account, that money becomes [your bank’s] and we are going to lend that out, we are going to fractional reserve that stuff until there's only $0.50 left of what you deposited and we are going to make as much money as we can with your money.
In the DeFi space, you had the switch of, it is my money. I'm the one that decides where it sits at any given time. I can see where it sits. I'm the only one that chooses where I want it to go. To me, it's a beautiful concept, but it comes with certain responsibilities. It means you need to understand what you're doing with your money.
“In the DeFi space, you had the switch of, it is my money. I'm the one that decides where it sits at any given time. I can see where it sits. I'm the only one that chooses where I want it to go. To me, it's a beautiful concept, but it comes with certain responsibilities. It means you need to understand what you're doing with your money.”
Now, the more we advertised the high returns we were getting at the time, the more people stopped listening to that and they would just hand over their money and this is something that used to always annoy me. Today it’s still the same, which is why I've switched gears a lot from now versus back then. Because back then, I'd advertise APR, I’d share audits with people. I gave them a false sense of security, which I don't like. Nowadays, I don't do that anymore. Even if the stuff is audited, I don't share it's audited. Even if I've got all of my money in there and it's been safe for the last month, I don't tell people that, because I want them to always feel at risk. You should never feel safe in this space. Because you need to always know what's going on and you need to always be aware of your money.
“I want them to always feel at risk. You should never feel safe in this space. Because you need to always know what's going on and you need to always be aware of your money.”
I don't advertise APR anymore, you are more than welcome to ask someone and they can validate it for you or you can deposit and you can validate it yourself. But don't, just because I made a tweet that said 60% APR, throw your life savings into that thing. Because you didn't validate if those numbers are correct, you didn't verify if it's safe. One of the beauties about this and it's one of the things I hate because I've made the mistake of being a fairly public figure, but it's also one of the best things I think about this industry is everything is publicly verifiable.
We as a community shout wolf very quick. If something even looks mildly off and everyone starts screaming at the top of their lungs, which having been at the receiving end of that, often when something was not wrong, has definitely driven me away from this community a few times. That did happen back in March. But at the same time, that's one of the good things, because that's what protects everyone else. Because that's the whole concept of blockchain. It's about, I don't trust that you move the money. With my node, I'm going to verify that you moved the money.
“We as a community shout wolf very quick. If something even looks mildly off and everyone starts screaming at the top of their lungs, which having been at the receiving end of that, often when something was not wrong, has definitely driven me away from this community a few times.”
The fact is that that we've started moving away from that ethos. Because I've seen projects advertised how many audits they've had. That is not a metric, that is not something you should be sharing with the rest of the world. I'm very glad you had it audited, but don't use that to give a false sense of comfort to other people…
Andre temporarily quit DeFi in March. Image source: CoinTelegraph
CR: What's the right way to communicate whether a person’s project is safe or not? You do kind of the opposite and maybe that's been a lesson that you learned from what you're describing now. You very clearly say “this is not safe,” I test in production is on your Twitter. But you know, at the same time, users are craving that, because not everyone is technically able to sift through the code. What's kind of the right way you think to communicate them?
Convenience or Safety
AC: The way I see it where we're in this highly experimental phase right now. If you look at other technology booms, the first people on the scene are the ones that take the risk. And they take that risk for the first two to five years before they start rolling it out and making it more accessible to others. In today's digital age, we just live in a space where information is so readily available that everyone has access to everything the whole time and something like building on Ethereum makes it worse.
And I've had this pain multiple times, where I can't deploy and test in a small scale. Because on the one hand, I can't stop people from interacting with it and on the other hand, if you test into a smaller scale, you don't attract enough eyes. Then when you get big enough, you get the eyes and then you backfire and you also don't want that. You almost want to be big enough at the start so that you have the eyes and then you can check it. But back to the original question.
We're in that discovery/innovation phase, where it's dangerous if uninformed actors to interact with the systems, because a wrong click can send your funds to the wrong place. I've watched how some of my friends and other people interact with MetaMask, they don't look at what that approval of your entire life savings in your wallet, to what smart contract is it actually going, and what is that smart contract doing with it.
The approval system was originally designed on ERC 20s so that you didn't give full control over to someone else. The contracts allowed to take 100 bucks, but it's not allowed to take everything else. Nowadays, everyone uses infinite approval. When you're on any website, it's infinite approval because they don't want to inconvenience the user.
But there's this awkward balance between security and convenience. Now, I agree over a little bit more time, we should get to that convenience spot. Like blockchain, Ethereum, smart contracts eventually will be at the point where you do not know you're interacting with it. You'll have an app on your phone, and there's a lot of guys already doing phenomenal work there, InstanDapp, Zappier, all these guys are doing phenomenal and they should keep doing it. But they've got the experts that look at and test with these smart contracts and then make it available to their users.
All of these guys, I've got group chats with all of them. When I launch something, they'll first look at it, they'll review it and they know what they're doing. And then if they're happy, they'll unveil it to their users. But there's a lot of users who don't want to wait, they don't want to wait until it's in those apps, so they'll just directly start interacting with the smart contract.
That's one of the reasons why and I've said this in a lot of my previous Mediums and tweets is I don't want to host the websites. I don't want to have people to have accessibility to these platforms. I think, low-level platforms should be accessible via wallets and these wallets have the engineers that can validate it. I think that's a much better process flow than having someone that doesn't really know what's going on, but they know how to do an approval on MetaMask, interact with a contract via EtherScan.
CR: You think the right way for mainstream users or nontechnical users to interact with DeFi is through these apps, these like apps at the higher level than the protocol level?
Frontier Instant Apps
AC: Dependent on the user. I mean, if you have that technical level, by all means, interact with the protocol directly. But I think it's a lot safer if you put another layer in between. Because as much focus as any of these companies might have on security, this is a highly experimental space. There are hacks and exploits we know about now that we couldn't conceive would exist a year ago. And I feel in a year from now, that statement is still going to be true. The amount of money that's flowing into these protocols, my own included is terrifying for a builder.
“There are hacks and exploits we know about now that we couldn't conceive would exist a year ago. And I feel in a year from now, that statement is still going to be true.”
I was so scared the other day when I launched those yVaults, because I did my precautions, I did my reviews and just praised whatever deity is appropriate, but just thank you that Samczsun exists and that he is a good guy. I released those vaults and within half an hour, there was half a million dollars in there. And that terrified me to the point where I was second-guessing if I should release more products. Because, I mean, you should never lose anyone's money. But if a mistake were to happen and a mistake will happen, then I'd rather be it on 500 bucks which I can reimburse someone than on half a million dollars which there's no chance I can help those people.
“I released those vaults and within half an hour, there was half a million dollars in there. And that terrified me to the point where I was second-guessing if I should release more products.”
CR: It's interesting to hear you say that, because, watching you, it seems like you have a much more kind of cavalier attitude to building. It's like what I had got from you was I'm building this, nobody trust me, I'm putting this out there. I can only imagine how scary it is to actually be attracting that amount of money and the level of responsibility that you're feeling.
Getting Over the Hump
AC: I stopped building in the past. I took a long break after March, when I had to rethink a lot of stuff about this industry, specifically because that fear stopped me from innovating. And it took me a long time to kind of get over that gap. Because I realized if the innovation stopped because of the fear, then just nothing more is going to happen. You have to get over that hump, but it's still terrifying to me. Because I mean, I know the whole test and prod and all that stuff is a bit of a meme, but it exists because I lack alternatives. I can't afford a $20,000 audit every time I launch a new smart contract and I mean that was the cheapest quotes I was getting, they were as high as 60,000. And I don't have that kind of money lying around to actually get these professional audits done.
“I realized if the innovation stopped because of the fear, then just nothing more is going to happen.”
But at the same time, I don't want to stop innovating, because there's still so much to do, there's still so much. I like saying there's so much money being left on the table, because we're not capital efficient yet. These systems aren't leveraging each other in intelligent ways by sharing liquidity, which is why I still believe everything should be tokenized into the nth degree because it gives you access to the lower level stuff. And some of the new stuff we're working on that has tokenized leverage and access, that's the stuff I'm getting really excited about, because that's when you start becoming extremely capital efficient and that's where we want to be.
But at the same time, it's a jewel movement I've noticed and I understand where it's coming from. But a lot of the hate is towards the builders, because it's their responsibility, they should have checked, they should have audited it. But you know what, as a builder, I will categorically state there will always be mistakes, there will always be flaws, there will always be faults.
“But a lot of the hate is towards the builders, because it's their responsibility, they should have checked, they should have audited it. But you know what, as a builder, I will categorically state there will always be mistakes, there will always be flaws, there will always be faults.”
It doesn't matter if you're Google or Apple or if you're some guy working in his garage, mistakes are made. The one thing that I've switched to and this is why I'm very “don't put your money here, I'm not going to explain it to you, you need to do your research yourself,” is because I want the people that just randomly put their money into these protocols to start taking a little bit of that ownership as well. I'm not saying it's all on them, but it can't be all on the builder, because if we keep putting everything on the builders, the builders are just going to go away.
CR: It's a delicate and hard balance right between innovation and safety. And I don't know. I assume there's people in this space who think builders shouldn't be putting anything out there without an audit, because in the end, you are holding people's money, it's a big responsibility. But at the same time, I do see your point that if that happens and only people with really deep pockets who can afford $60,000 audits for products are the only ones going to be shipping and that kind of goes against all of the innovation that we've seen in DeFi. I assume people in this space wouldn't want that to happen. I don't know what the right answer is.
AC: No, don't worry, I have no bloody clue at all whatsoever. What I’d like to see, one of the things I try hard in doing is just, let's decrease the finger-pointing. And that's why again, I'm going to come back to Sam, who has been such a pillar for me lately. Because, when he found that, he didn't go shouting at the top of his lungs. He was like, hey, this is how you can fix it. That change in mindset is so big to me.
“What I’d like to see, one of the things I try hard in doing is just, let's decrease the finger-pointing.”
Because a lot of the times, I remember this distinctly back in May and I remember this distinctly when I relaunched the YFI token as well, is there's a lot of people and these are very public, well-known people that go into private conversations and will tell people about like the private key thing. But for example, where that was shared in private groups, but it was never discussed with me. And I get it comes as a double-edged sword. Because now, let's say I was a malicious actor and I was planning on stealing all that money, then if they were to come to me and tell me about it, then I probably would have exited at that point, because people are going to realize and I want to steal as much money. Again, there's a delicate balance and I don't know what the answer is. I just actually don't know what the answer is. But we need to find that little bit of a balance of it's not you and me, it's us. And I'm seeing us getting there. I'd hoped it was a little bit faster, because in all honesty, things haven't changed much since Feb-March. But you just have to keep doing it. It's hard. But if you keep doing it every day a little bit, then it gets easier.
CR: I guess, in general, what's always advisable is for users to be extremely careful with wherever they're putting their money and taking responsibility to some extent for what they're doing with their money as well.
Okay, so we haven't even got into YFI which I definitely want to ask you about. YFI is the governance token for yearn and it just has prompted so much activity and fueled this very ardent community in just a couple of weeks since it launched. I want to ask you how the idea for this token came about and also for its architecture, which is I think is the first fully community-owned token?
AC: I always like this part, because there's a lot of people that assume it's altruistic, I did it for decentralized reasons, a bunch of stuff like that. And to be blatantly honest, I was interacting with yearn one day and I did a rebalance because a lot of people had just sort of started leaving their funds there. There was like 8-10 million AUM, but these are people that have just left it there for months, so like there, there wasn't as much interaction as I'd like, so, I told the system to rebalance itself. And the gas fee was $108. I was like, “that is insane,” I cannot keep paying $108 and I mean, it's getting worse. And I realized I need other people to interact with the system as well.
And that means, setting up, helping fine-tune settings, which I've mostly thumbs sucked and if they work, they work; if they don't, I change them. But like it helps getting more eyes on it, more people involved, more people interact in with the system.
I started building my little governance module, which just hooks into all of the different systems and is capable of updating the variables. And then I needed a way for people to make decisions on those actions on this smart contract layer. Originally, I thought about a DAO, but I would be like, okay, but who would be my signatories and for how long would they be interested? You know, what happens if they leave in a week from now and they're no longer interested in working on it, then the whole system shuts down?
I realized, let's launch a token with the idea that I stressed it so much in that initial articles, that I did not want it to have financial value. Because as soon as something gets financial value, then people start looking at that more so than the protocol. As much as I love the YFI community and everything they're doing and it's phenomenal. I am blown away by what has happened and I could not have guessed it in any prediction model I design.
But at the same time, all of the conversations are on YFI, like there are very few conversations around the protocol. What should the fee and the reward share be? What should be the next product that we launched? Should I be focusing on ySwap, yTrade, yLeverage? I mean, there's so many other tools, which also sorry, we haven't even touched those in the conversation, because there's a whole suite of stuff that's yearn, it's just one component.
YFI Governance Module
Anyway, back to the original point, is that's why I did the distribution. And that's why the initial distribution was for people that are in the yearn system, because their money is there, I wanted to give them that control so that they could make those decisions. Because if your money is already in the system, you're already vested to do what's best for your money. You as person that has your money in there would want the yield switching to happen as quickly and as fast as possible to optimize it or you would want the fees to be at a level where you're compensated for the work you're doing, but not so much that you are subtracting from the interest to the point where another protocol is cheaper. Because DeFi is also a zero-sum game. It doesn't help if you build in fees and structures like that, because eventually, someone's going to fork it and remove your fees and do it for free, so you just have to accept that.
That's 100% the only reason why I decided to launch a token. Everything else that happened after that was a beautiful sequence of, I don't know what else to call it other than just coincidence and luck.
CR: Before I get to this question, how much YFI do you own?
AC: Currently, 1.48 or 1.58 tokens. I had just over 2, but unfortunately, I had to sell some from gas, I actually feel very bad about that, but I’ve all on my account, so it's not like I can hide it or anything. I wish I had more, but the capital that flooded in so quickly overpowered my position that I just couldn't compete. I don't regret it. In all honesty, this is how I pictured it played out.
$300 Million Overnight
I picture I start my rewards contracts and people who are already using the system, so you know that like $8 million AUM, let's say like 20% of them hear about it and they think, “hey, why not, there's nothing to lose” and they'd stake and I'd be staking with them. And we distribute it in that little pool and then I would now have like earned, let's say, the majority of these tokens. So, then instead of needing to release new tokens, I would have actually taken the ones I earned and then incentivize the next week and then incentivize the next week. And I would have just kept cycling that until the token was distributed enough to the point where it was enough of the protocol users’ hands that I could then move away from there.
CR: But obviously, it all happened way too quickly for that, like you got $300 million overnight.
AC: Yes. That kind of completely ruined all of my original plans and theories. Still, I’m very happy about this phenomenon. But it was definitely not the design.
CR: You said this is a valueless token. But you must have thought it would have some value, I mean, did you ever imagine it going to like over $4,000?
AC: I calculated the value for it actually. Because I knew what kind of rewards the systems were generating and I know, sort of the yearly prediction of it. That gave me a baseline of you know, if I were to earn this token and I were to then leave it, like, how much would I make over that year? All of my prediction models, all of my designs based on this reward structure priced the token between $2-3.
“All of my prediction models, all of my designs based on this reward structure priced the token between $2-3”
CR: Oh my God, that's crazy. And now it’s 1,000 times more than that.
AC: I was slightly off lightly.
YFI price chart. Image source: CoinGecko
CR: Slightly. And is that because of the amount of liquidity that flooded in that it has raised the value of the token?
AC: I can't answer that question. I don't understand how people are valuing these tokens, because now this comes back to like I said, originally, I don't understand the volatility side of the crypto market or the TA-FA sign. I know a lot of people talk about TVL which TVL, I think only matters if you have inherent like percentage-based fee system that's fixed into the protocol. But like I said, you can't really have that because of the zero-sum game and all that stuff. So, then I sort of understand that model, but I also don't understand that model.
A lot of people have said that it has the cyclical because TVL was increasing, people assume, they compare it with another token that is of X market cap and thus for it should be Y value. And that feedback loops kept cycling in, which means higher TVL, which means higher valuation, which means more TVL, that's I think, the predominant theory.
I try and not give it too much attention. Reason being is I have a fairly vocal and longstanding history in this industry of saying tokens are shit. I have a strong belief system in base-layer tokens. What I mean with that is, ETH or BTC or a native chain token, because it needs a security budget that rewards people that secure the network. Any other token is actually just rent-seeking and I mean, I feel bad saying that. Because I love a lot of these companies and I work closely with them and it's phenomenal. But in the back of my brain, I'm always like, why didn't you just use ETH? You could have done the same thing there.
“I have a fairly vocal and longstanding history in this industry of saying tokens are shit.”
CR: Why is it so different for your own token?
AC: Because I couldn't afford ETH to distribute for people to then govern. If I could, I probably would. And I mean, it could be fractional, it could be way of ownership. But then there's the whole question of security versus governance rights. But that's again, why I'm trying to stay away from whatever the crazy crypto market says that thing is worth. Because as far as my protocol is concerned, one token is equal to one token. It doesn't matter if someone else attaches the value of 5 or 500, because it's not going to change your rights in the system.
That's a very important focus to me as well. Because I find when you start getting entrenched in the noise of token value and profits and all of those things, like, I really hate trading, because there's always a winner and a loser and we glorify the winners and that's fine, I respect it, because you don't want to talk about the bad stuff, I get that. But there's always someone that bought in the top. And that makes me miserable. I don't like that part.
I understand 100% is part of our industry, it's probably never going to go away. But when anyone asks me about the price or what I think about what's going on or its movements, I tell them, look, whatever the value is that you connect to it, that's up to you. Because for all of my purposes, I only care that that one YFI token is equal to one YFI token
“Whatever the value is that you connect to it, that's up to you. Because for all of my purposes, I only care that that one YFI token is equal to one YFI token.”
CR: Otherwise, I can only imagine how distracting it must be to be holding a stake in this token and have it blow up like that. It must kind of blow your mind a little bit, so I think it's a good way to frame it.
AC: But so, the other reason why I also consider it a bit of a bad thing is everything that happened around the token put it in the hands of people that don't care about the protocol. Now, that has meant that I have now had to change certain governance designs to accommodate the fact that not everyone that has the token might be a participant in the ecosystem. Sometimes that's a detriment.
Actually, I've written a few articles on this as well back in, I think, mid-2018 probably, where I tried to explain to people how speculators ruin tokens, because it's just tokens have a certain system efficiency, where, if the token is above a certain value, that's good, because it adds, let's say it's a base-layer token. If it's above a certain price, it's good because it adds security to the network. If it's below a certain price, it's bad, because then it means that the people that want to provide that security aren't going to provide the security because it doesn't cover their base costs. But now, if you add speculators into the mix, they ruin this price dynamic to the point where that band that it was designed to live in no longer makes feasible sense. They make the token so expensive that they can no longer do their job.
“But now, if you add speculators into the mix, they ruin this price dynamic to the point where that band that it was designed to live in no longer makes feasible sense. They make the token so expensive that they can no longer do their job.”
CR: Is that happening right now with YFI?
AC: No. The only problem with the current high value is that it means it's in the hands of speculators. I mean, there's nothing that hurts me more when I go into Twitter or I go into Medium or Discord or Telegram or I don't care what the medium is and I read about these people that say they're holding, but they have no intention of being part of governance. Like that to me hurts. Because it's designed for the singular purpose. It has this one mission that it's supposed to accomplish and you're now keeping it out of the hands of people that want to accomplish that goal.
CR: And I think that's been kind of the danger of a lot of these governance tokens. I mean, they have been so speculative after what happened with COMP and how it shot up, now everyone wants the next DeFi governance token basically because it gains in value and not necessarily because they actually want to participate in governance. YFI has been a victim of that if you want to call it that. But going forward, how do you think this space can mature and evolve from here where governance tokens have become so speculative? Do they become the thing that they were actually built for?
Victim Of Its Own Success
AC: Honestly, I don't think we can. The beauty of something like the current DeFi ecospace on Ethereum and as you know, you've got so much liquidity and interaction between the DEXs, Curve, Balancer, Uniswap and more coming out on a daily basis. Like that there's so much immediate availability that the traders will control this market.
I actually don't have as clean answer for that one. Well, I guess it probably comes with maturity. We're still in very much that teenager phase, I want to call it, and in a few years, maybe we'll reach that maturity stage where things are less volatile. Because I think, if the volatility stabilizes, then you'll probably see a difference in those systems.
But I mean, we've got the same problems than in traditional stock markets. I mean, how many people that own Facebook stocks are actually voting or actually contributing to board-level decisions or shareholder-level decision? I was probably just naive in all honesty. I think I was just naive.
CR: Fair enough. It's so hard to predict where these tokens will go and nobody thought any of them would rally in the way they did or that they actually be so effective in attracting liquidity, which I guess was the purpose for some of them. They have been successful at least in that way in attracting activity and liquidity to the space. Maybe in the long term, they will be more like shares, even though nobody wants to talk about that, because regulators. But I think that's how it kind of they're acting right now. It's like sure, some people will participate in governance, but other people just use them as a way to gain exposure into the platforms that they believe in.
AC: Also, true. I understand you've got money in the platform, so why wouldn’t you want to have a vested interest in the thing that governs that even though you might not be a participant? And I respect that as well. My ideal person that holds one of these tokens is A, I have some funds in the protocol, so I care about is it making me the most yield possible and B, I participate in governance. Short of not having the I participate in governance, at least give me I actually have funds in the system. But if you're that third case that just has it because for speculative reasons and you don't care about the protocol or governance, no offense, but you shouldn't have this thing.
“Short of not having the I participate in governance, at least give me I actually have funds in the system. But if you're that third case that just has it because for speculative reasons and you don't care about the protocol or governance, no offense, but you shouldn't have this thing.”
CR: Makes sense. Awesome. We were going a little bit over time, but I'd love to just end by hearing your bigger vision and long term view for DeFi, where do you see this thing going?
Dog Chasing Cars
AC: I don't have an answer for you. I really wish I could say something inspirational and based on some kind of visionary statement. I have this quote from Batman that I always tell people, that I'm just a dog chasing cars, and I wouldn’t know what to do with one if I caught up to it.
I think there's this convergence between people having more control with their own money while making money programmable, which I think is two very awesome concepts that are busy colliding. I think that will give us a more efficient world.
“I'm just a dog chasing cars, I wouldn’t know what to do with one if I caught up to it.”
I remember, I was asked by someone if I think that DeFi is the future of finance or the future of banking? I don't remember exactly where they phrased it. And I never liked that, because I don't see blockchain destroying banks or changing the current status quo. Banking is for bankers; blockchain is finance for geeks. You know, it's for the geeks, the coders, the technocrats that want to have that little bit of control, so it's more of a new paradigm. But I have no bloody idea where it's going to go.
I could not have projected in any given timeframe looking back where I was going to be or what was going to happen in the next three months. This is something I try and stress on my Twitter as well. I know we're in this hype cycle right now and everyone is excited, but none of that is sustainable and people shouldn't think it's sustainable.
But on the other hand, we have these wonderful “savings accounts”, things like yearn or, Curve or a Uniswap LP, it doesn't matter what you do, that's earning you 10-20% a year which is phenomenal rates. I would like to see more focus on that side and less on the unsustainable side. Because the problem is you focus on the unsustainable side, and when it stops and it always stops, then people say oh, but it didn't work, see, we told you so, none of this can work. Versus if they focused on the sustainable side, they stop saying that. Because in a year from now, that still exists and it's going to be those things that in a year from now still exists, not the 10X moon, now all of a sudden you're DeFi. But in the next two months you're gone again. Like all of the big names you'd look at now are guys that have been trudging through this since last year.
Any new protocol that’s launching today, they're not doing it for the sustainability reason, they're doing it for the hype reason.
I don't really have a answer, a prediction or even a vision. All I know is I wake up in the morning and I'm really excited to code and try and figure out new stuff to add new things to this ecosystem. As long as that holds true, I'm going to keep doing.
“All I know is I wake up in the morning and I'm really excited to code and try and figure out new stuff to add new things to this ecosystem. As long as that holds true, I'm going to keep doing.”
Hope you’re enjoying The Defiant. If you are, spread the word!
The Defiant is a daily newsletter focusing on decentralized finance, a new financial system that’s being built on top of open blockchains. The space is evolving at breakneck speed and revolutionizing tech and money.
About the editor: Camila Russo, is the author of The Infinite Machine, the first book on the history of Ethereum. She was previously at Bloomberg News in New York, Madrid and Buenos Aires covering markets. She has extensively covered crypto and finance, and is now diving into DeFi, the intersection of the two.