"People Aren't Earning Anything in Their Bank Accounts; Let's Bring Them to DeFi:" Aave's Stani Kulechov

Kulechov says Aave's flash loans launched last week will enable an explosion of new DeFi products and users

Hello Defiers! Happy start of the week to everyone. This week’s Monday Interview is with Stani Kulechov, the founder and CEO of decentralized lending platform Aave. After much anticipation from the DeFi community, Aave launched on mainnet last week with a series of innovations, including “flash loans,” which are issued and repaid within the same Ethereum block, without the need of collateral, and fixed rate collateralizes loans.

Stani says he envisions flash loans as a tool for developers to build mind-blowing tools for the end user and enable entrepreneurs to continue building new DeFi projects without requiring as much capital up front. He believes the DeFi narrative will change from value locked, to how this value is being used, and that flash loans are one way to use this locked capital more efficiently. He also talked about what’s next for Aave, with governance at the forefront in terms of profuct, while also a Series A round with VCs coming up down the line.

You’re a paid subscriber so you’re receiving access to the full interview. Read up!


On the Aave Launch

Stani Kulechov: We were kind of ready already in Devcon, but still wanted to just improve a bit then. Then we had like a second trial before the holiday in December, and we still wanted to kind of tweak and polish it and eventually we decided to do it now. I think this is better than anything else. It was very tiring process. But now you finally see it and it's amazing.

Camila Russo: Congrats! A little bit more on the process. How long did it take to build, how many people were in your team? What were kind of the biggest challenges?

SK: We released the latest version of EthLend last year in January or March and we understood that pooled models were a very good way to arrange liquidity, and then we started building the Aave protocol. So it took quite a lot of time and effort.

We are not that big a team but we aren’t small. We're roughly 20 people here. Half are basically developers and related to product and everyone else is a bit kind of like support and research. The skill set that we have is pretty nice. Everyone has done great job and have grown with the process. We feel as a team and as individuals feel we actually know a lot about what we're doing. And that wasn't the case like a year and a half, two year ago. We are based in London and we have another office in Switzerland.


Image source: Aave website

Background and Crypto Red Pill

CR: Before we get into more of the details of the platform, I wanted to go back and ask about your background. Can you tell me a little bit of what you were doing pre crypto and what got you into crypto?

SK: To be honest, what got me into crypto was pretty much Ethereum. I knew always that there was Bitcoin around. But the thing with Ethereum was that with smart contracts you can create more complex transactions, and remove custody, and have code that can't be changed. That's super powerful. In college I started to try to explore a bit more what you can actually create. And back in the day there wasn't DeFi, there wasn't actually anything there. I think there was like prediction markets with Augur. There was a time when cryptographic tokens started to have a secondary market in centralized exchanges and a bit on dexes, but not that much.

And I was thinking, what could be cool is you could actually use this value as collateral and one of the interesting use cases for it could be lending. And I wanted to create something that's trustless with some value between the parties. That is where the collateral comes in. And then we just started to experiment. EthLend was the first iteration. And it was funny because back in the day when we launched, people were asking in a big Reddit thread and post why would someone borrow with collateral if they can sell or trade. And today there's half a billion dollars in DeFi.


Image source: Medium

CR: You said you were in college when when you started getting into Ethereum and building EthLend. So, you had no traditional financial job or non-crypto job before Aave?

SK: No. Well, I used to do some development when I was in my teen years, a few years. My background is legal, so I did some law-related work too. But I've always been interested and fascinated by finance, even in college I was more interested in finance than law and in general.

When you look at finance today, it's is pretty ineffective. What you have in your mind about what you want to do with finance, it can be very innovative. But you can't do any of those things in practice in the financial system. And for me that doesn't make sense. If I send funds to someone and someone needs to wait two, three or four days to receive those funds or any other kind of similar inefficiency. At some point I fell in love with financial instruments. I did a lot of research and tried to think about how we could actually reproduce those different kinds of financial instruments and remove all the inefficiency. I had this thought that you could have an internet of smart contracts that removes the base of trust between entities. And that was the vision.

CR: To round up the background part of your story, you said you were studying something related to law?

SK: Yes, I think I was in third year or so when EthLend was founded. I'm not a dropout, but I never graduated. I should graduate this year. Aave and EthLend basically came in the middle of my life and have opened a new world and it's just so exciting that it's difficult to do anything else, you know. It's not the first love, but I think it's like the true love, what the whole Ethereum space is for me. So it's just difficult to do anything else other than what we're doing now. I have an excellent team of people thinking similar ways to try to improve the financial ecosystem. It feels it's like a once in a lifetime opportunity and I just can't just go away from that.

CR: Can I ask if it's not a rude question, how old you are?

SK: I will be 29 on Monday [today]. This is the biggest present ever being able to launch before my birthday. I can't wish for anything else to be honest.

The Aave Idea

CR: Can you take me through the process of coming up with EthLend and how that led to Aave?

SK: We kind of kick started a bit the whole DeFi lending space. And after us there's been a lot of innovation and other different kinds of interesting projects out there improving what we did. What we are doing with the other protocols is basically improving what's there is now and not just improving but also giving new options for people. Because the thing is that we can't be too concentrated on doing something very similar because if you want to get more people into DeFi we need to have more offerings and more alternative offerings, different kinds of ways to establish interest rates, different kinds of risk models and different kinds of ways to build the protocols technically. When we have this diverse ecosystem, then we will be able to onboard more people and there's more trust there and different sorts of solutions.

With EthLend, is we decided to look at what's now working in DeFi, what others have been building, and let's try to figure out what's missing. And one of the main narratives was, we need to get yield for people. People are not earning anything in their bank accounts. We have to get those people to DeFi and, and show that there's yield here and other cool stuff that you don't normally get with your bank, like the non-custodial relationship and also transparency.

We also understood that there is now the assets locked. So if you look DeFi Pulse, what everyone is always talking about, which is like the narrative, how much assets are locked into DeFi protocols. But we tend to forget is that we need to always look a bit to the traditional finance and basically see how they're utilizing assets because otherwise we won't be able to compete. And it’s hard to compete, because they're doing it with a trust-based model.

Undercollateralized Flash Loans

To compete we need to actually utilize the assets in a better way and that's how we came up with the flash loans. So what we are trying to do with the flash loans functionality, which is basically, anyone can borrow undercollateralized loans from the protocol, do all kinds of transactions in the internet of smart contracts and return to funds and get extra yield in the protocol. We reuse the collateral. That's the next step in the DeFi narrative. We can't be just looking at how much we are locking, but how we are using those locked funds. How are we pulling out that liquidity?

CR: Can you go a little bit deeper into how flash loans work? I understand it's you take out a loan and you repay it on the same transaction, but what's actually going on there? And how long does the whole process take?

SK: Flash loans are pretty simple. There's a function in our smart contract called "flash loan" and basically when you call the function and you borrow from the protocol, there is a condition that says that if the loan is not returned on the same transaction with the fee, the transaction gets reverted. And because of the logic that we have, that basically means that you can do all kinds of different transactions. So it doesn't need to be just that you borrow.

The service that we are doing is ensuring that you can borrow and return and in between these, what we call circuits, you can do all different kinds of stuff. So for example, you can open a trade with that same liquidity in Uniswap, close another trade in Kyber or one of the most interesting use cases for me is for example, you take a Dai flash loan, close a CDP -- for example if you have a high-interest rate CDP but you see a lower rate somewhere else, you take the Dai, close the CDP, take the ETH, send it to the other lending protocol and draw the Dai from there and basically return the flash loan, everything on the same transaction, which basically means that you have just refinanced your client or yourself if you're doing it for yourself.

And you can also do it even more complex, like adding more things into the circuit. So you could actually swap currency. So let's say you are in a loan with a USDC in Compound. You could do the same thing in reverse, but also you change the USDC to basically Dai. And that way you just don't do the interest rate swap there, you are also doing the currency swap. This is what fascinates me. But I'm pretty sure that there's other people thinking way more complex or even more simple and workable things. That's the kind of logic how you utilize these loans. You can actually take as much liquidity as there is in the smart contract. So there's no cap in that and that can be done in every Ethereum block.

CR: Oh, nice. Okay but one block takes just a few seconds right?

SK: Usually in Ethereum each block is confirmed every 13 seconds, but the seconds are not that important. The block is. So whatever you do, you have to do in one block. Each Ethereum block basically settles the transactions in a certain way. So you can do all kinds of things, but there has to be finality inside of the Ethereum block. .

CR: Okay. So anything you do with the liquidity you take with a flash loan you have to do in 13 seconds.

SK: Yeah. Whatever you do, you have to return what you borrowed on the same block. So there is some different kinds of transactions that you can actually do and borrow liquidity. It's like flash liquidity in a way because you are just taking it to do some sort financing method.

So take the two young guys from India, who participated in the ETHIndia hackathon some time ago and built an easy way to do, for example, CDPs, with one transaction instead of the all the five or eight involved with InstaDApp. Those guys did a bridge where they enabled people to close their CDP and refinance your loan to Compound. And to be able to do that they had to borrow from Comopund, off-chain, and ask, I think it was between 300,000 and 500,000 pounds to do this operations, this refinancing.

These guys don't have that kind of money because they're just young developers with the vision is to build products. In a similar case, they could have used flash loans to borrow for this particular transactions without actually needing the capital. So if you think about the big picture of what we're trying to achieve is that we can get more developers building financial products without the need of the actual capital and that lowers the threshold. That’s how we can increase the types of DeFi offerings, and even more so with the huge composability there is.

CR: Amazing vision. What I still don't get is, you have to build some kind of bot or something to actually execute whatever you want to do with that liquidity within 13 seconds. Like no human can actually do that, right?

SK: Flash loans are we could say directly a developer tool. Because a developer can create a refinancing tool, or an arbitrage tool or just to liquidate CDPs for example. But at the end of the day it's always the end user who is benefitting.

Let's, let's say I'm a developer, I create a service and I'm using flash loans. Say the service is called liquidation service, and say you have a CDP open and it gets into collateral call. So when this collateral margin call happens in the Maker system, you're losing 13% as a liquidation fee. Say you took the CDP because you are long on ether, but you needed the Dai to spend in paying bills. So you already spent that Dai, and now there's a liquidation coming and you're losing 13% of the value that you have locked into Maker. Now what you can do as a developer, is build a product and says subscribe to this service and if your CDP is getting liquidated, a flash loan is taken and the loan is repaid before the liquidation happens. So then you save 13%, or maybe the app will get up 1%, and you keeps 12%. So the developer is building this product, but you could be the end user.

CR: So you envision flash loans being a tool for developers to make DeFi easier for the end user.

SK: Yeah, exactly. And not just easier. I would say they will also expand DeFi use cases because you don't need that much capital. And it also helps make things cheaper because it lowers the cost of the transactions. But of course that will take a bit of time and experimentation and we'll see how it plays out. But yeah, that's the core of what we're trying to achieve. So basically more DeFi products, more diverse DeFi products and basically more yields to the depositors and hopefully lowering the borrowing fees as well because there's more income coming into the pools. So that's the overall big picture of what we're aiming for.

Fixed Rate Crypto Loans

CR: I'd like to get into your other, other features too. You also have this ability to switch from a fixed rate to variable rate. What was the thinking and the innovation behind that?

SK: I'm glad you asked about this because, we want to change the narrative in the space and also give some attention to the consumption side. So we have all these people depositing funds earning interest, but we also need to acknowledge that there's other people that are utilizing and using those funds. We are trying to increase is the demand as well, and to increase it, we need to give more options for the end users.

We can't just offer variable rates, where let's say you are paying now 5% now, and during the night you might be paying 7%, depending on liquidity. And if one big entity withdraws liquidity the rate goes up. That's unacceptable for the end user, especially when we go to the retail side. And that's what we are trying to achieve is give retailers more options as well. And rate switching allows us to do that, allows us to give another alternative to the variable rates, where you have certainty and stability on your interest rates.

People who are uncertain on taking a DeFi loan because of the varying rates, now have more certainty. We expand a bit the scope on how we can onboard people. And you can always switch. Let's say you jumped into a variable rate and the rates started going up, you can jump very stable rate then and be there or if you see that the rates are going down, you can jump into a variable rate and you can switch on block by block basis as well.

CR: Is it always the option of the borrower to switch or does the system sometimes require for stable rate loans to upgrade to variable rate, or raise the interest because the whole market is increasing? So I get the idea of changing from stable to variable because rates are falling, but is there a case where it doesn't make sense for so many borrowers to have a stable, low rate when the whole market is increasing rates? Would the system force the borrowers to raise rates in that case?

SK: Yeah, actually, that's an interesting question economically because usually when you try to provide stability in rates, you usually offer fix rates within a certain period. Let's say, for one or two years. Our idea is not to fix loans to a certain period of time. The system works in a way that you always have a variable or stable rates. So if you take a stable rate and the market changes, the stable rates changes as well. But once you take the offering, you're fixed into that rate. So the stable rate might change, but for the next guy who comes into the platform and takes the next loan.

There's two exceptions, where we can rebalance the borrower. In case there's a system where the deposit earning rate goes higher than the stable, we'd basically rebalance the user to the current new stable rate. And this is to prevent an abuse where you borrow and deposit because you are basically getting cheap liquidity and earning more. But it also works both ways. When there is a lower rate stability fee than what the person is currently by 20%, the person is rebalanced down. And anyone can call this function to smart contracts and basically we were going to do a bot that actually automatically calls. So if there is a drastic movement in interest rates, the person is basically rebalanced down. But these are edge cases. I'm not sure if they will actually happen ever, but we never know.

CR: Okay. And that's in the case that the deposit rate on Aave itself, on your own platform, is higher from the borrow rate.

SK: Yeah, exactly. Something else to point out. You just deposit and you start earning basically from that particular moment. And that's why the system is so cool because you don't need to order match for example, like on EthLend where you had to match borrowers with lenders. This is an automatic system, you juts deposit and and start earning and borrowers can go fund themselves and get liquidity. So that's the beauty of pool models and of course because of the utilization rate there's basically always more capital in the pool, than there's actually consumption. So there's always enough liquidity there. If the utilization rate gets into 100%, you still have the aTokens you can trade in a secondary market if there isn't the possibility to withdraw.

Interest-Bearing Tokens

CR: Finally I wanted to talk about, I think maybe the third major piece of your platform, which is aTokens. So I understand the main innovation there is the way their value is represented to users, with their value pegged one-to-one with the value of the underlying assets. So can you, can you tell me more about why you wanted it to be this way?

SK: The whole idea of aTokens is we looked at how other DeFi projects have built the tokenization of the deposit position. Usually you have a token that grows in value, but it's very difficult to actually calculate how much you are actually making. And we wanted to have a token that’s a goos representation of what you deposited and the principal and interest accrued. If you deposit 100 Dai, you get 100 aDai.

And when you look at your MetaMask or Coinbase, wallet, whatever you're using, you actually see the balance of that aDai, increasing all the time. Then you see from that particular representation what you are actually earning. And then you take that aDai balance and you put it back into the Aave protocol and you get to your Dai principle with the interest and coolest part is that you can even redirect that interest to someone else. The exchange rate approach for me is more complex to understand. The end user really wants simplicity and clarity. And with aTokens they can understand what they're getting without actually going into let's say the platform or going to a Discord group and asking about the exchange rate. It's a small thing, but it's pretty cool thing that I improves the user experience.

CR: Yeah, it is easier to look at it that way. And, and you said there's an easy way to redirect interest. I didn't know that. I didn't see that.

SK: We built natively a way to redirect your interest on aDai. So now we or someone else has to build is the user interface.


CR: What are the next steps for Aave? If you can tell me a little bit of your pipeline.

SK: The Aave protocol is open source, which basically means that the protocol code itself can be usable by anyone and they can do whatever they want with the code, they can fork it, and that's the difference with Compound, which is proprietary technology and they own the source code. Open source is important because if you are building something very meaningful, you want to make sure that everyone who wants to can participate. In the same way, when you have big decisions, you need to have democracy, because then you are ensuring that everyone has a right to participate in the protocols' decisions, and ensure that's the finance that you want to have.

In the current status quo in finance, what happens is that other people are deciding for us what, what kind of products we'll have and that's what we are trying to change. Because of open source we can actually get the community to govern. So we are now building the governance function. All the holders of our token can basically vote on different changes. We want to have a democratic way for people to decide how the protocol works in the future, what kind of changes there will be and, and how, how all those will look in the future.

And of course now is the time to listen to the end users and hear how they feel about the current protocol and trying to come up with a good list for the next year and what kind of improvements we want to do and what kind of assets we want to add. But for the next couple of months it's all about governance, all about decentralizing the whole thing and basically involving the community as much as possible. And of course, focusing on developers and helping people to build on top. That's why we were participating in hackathons this year again. And we are excited now that we have a very nice product now.

Fundraising and Business Model

CR:I wanted to know what's going on with kind of your fundraising, like are you aye, are you getting money from, from VCs or like how are you bootstrapping this? Is it all from your own funds?

SK: Back in 2017 when I was still in college and we created EthLend we fundraised in the form of an ICO. Capital is very constrained if you think about it. So if you go to San Francisco, New York, Singapore, there's a lot of capital. But when you look at Nordics, South, South America and these kind of places, it's actually pretty difficult to get any capital unless your idea is very, very convincing. And basically that particular ICO gave me the opportunity to continue building a product and ensure that we actually have a team building EthLend and Aave. We raised $16.2 million and are still suing those funds.

Our next step is to do a round A with VCs and basically because we have a token, our goal is just to sell tokens from our supply to VCs that want to participate in the governance, have a say. And on top of that we want to do is we want to engage also traditional finance. We are betting on the smart contract-based money markets and DeFi, but we want raditional financial institutions to participate into the governance and be able to have a professional say. But end of the day, it's all about not having concentrations anywhere.

CR: Going forward, beyond fundraising, what's the business model that you're planning for the platform?

SK: Currently we have different fees. We have the origination fee and fees on the flash loans. So we collect some of the fees but we don't collect them directly to us. Instead, we are burning the platform's Lend token. We want to grow the Lend ecosystem. And when Lend is used in governance voting, different decisions-making proposals, it captures the value for the protocols and those who are early in the protocol, they'll hopefully see the incremental value in the token.

And it's pretty much the same that others are doing now. For example, Kyber, Maker, 0x, are good examples. It's actually very good because after the ICO boom people were thinking that tokens are done but now we're seeing that actually they’re not. Useless tokens are done but in some cases there's real usage there. If you have smart contract-based protocol where you can actually vote, that means that you have direct influence and that's very important for people.

CR: Okay. So basically you're planning to have use in the platform result in an increase in value of the Lend token.

SK: Yeah, exactly. And not just using, but also enabling people to have a say in the future of the platform. It's a governance token and that is something that's very powerful and that I really believe in. The general overview is that it has to help the protocol and be part of the decision making and that we don't have any power. The token is a nice way to decentralize the whole thing. The more we give out our power from us, the more value the token has.

Future of DeFi

CR: Looking at DeFi and where the space is going, what excites you the most? What's keeping you in Ethereum and DeFi?

SK: Immutability is part of it. The very simple concept that you can deploy code that won't change. It solves a lot of issues. It solves trust issues and creates opportunities. The biggest thing I'm bullish on is the assets. When I see different kinds of assets coming in and different kinds of protocols, whether it's real assets tokenized, or just cryptographic assets with different kinds of functionalities, it will be interesting to see a world of hundreds of thousands of assets that could be used as collateral or could be lended out. And in that kind of scenario, what is important is that your protocol is, is as democratic as possible.

And I think that's the key of the whole governance discussion. And I'm very bullish that the DeFi space will go into that way. And the more there is assets, the more there is opportunities and reasoning to create new DeFi products. That’s what I’m most excited about; the tokenization of everything and that everything will be on smart contracts. I know that sometimes it might feel that it will never happen, but I think personally that we are closer and closer to this ideology.

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. Sign up to learn more and keep up on the latest, most interesting developments. Subscribers get full access, while free signups get only part of the content.

Subscribe now

Click here to pay with DAI.There’s a limited amount of OG Memberships at 70 Dai per annual subscription ($100/yr normal price).

About the author: I’m Camila Russo, a financial journalist writing a book on Ethereum with Harper Collins. (Pre-order The Infinite Machine here). I was previously at Bloomberg News in New York, Madrid and Buenos Aires covering markets. I’ve extensively covered crypto and finance, and now I’m diving into DeFi, the intersection of the two.