Building a Decentralized Lending Protocol

Brian Friel sits down with Solend founder Rooter to hear how he built the largest lending protocol on Solana.

Show Notes
00:41 -  Origin of PFP                    
01:27 - Origin Story                     
03:26 - When did they settle on Solana?  
04:36 - The Journey of Solend            
08:19 - Dealing with the Whale and Governance Proposals
14:23 - Time Frame from whale discovery to governance proposal
16:06 - What were the proposals?         
19:06 - The third proposal               
21:31 - Thoughts on governance and decentralization
23:46 -  What would he do anything differently?
25:36 - Advice to other founders on governance
26:49 - Isolated pools                   
31:24 - Onboarding onto DeFi             
32:57 - Permissionless pools             
35:39 - What's next for Solend           
36:33 - A builder he admires        

Full Transcript
Brian (00:06):

Hey everyone and welcome to the Zeitgeist, the show where we highlight the founders, developers, and designers who are pushing the web three space forward. I'm Brian Friel, Developer Relations at Phantom. I'm super excited to introduce my guest, Rooter. Rooter is the founder of Solend, the largest decentralized lending protocol on Solana. Rooter, welcome to the show.

Rooter (00:26):

Hey, thank you, and thanks for having me.

Brian (00:27):

I think everyone on crypto Twitter will recognize you by your iconic profile picture. It looks to me like a aardvark that's shedding a single tear. Can you give a little background on what is that animal in your profile pic and how you came to rock that as your PFP?

Rooter (00:41):

So it's actually Pequenino, which is a fictional species from a book called Speaker for the Dead, which is part of Ender's Game series.

Brian (00:52):

Oh, I love it.

Rooter (00:52):

Basically, I was looking for a profile picture and this is one of my favorite sci-fi books. I found a frame in a comic and thought it was unique to have the character crying. Took that frame and ended up getting a custom artist piece made.

Brian (01:08):

That's an awesome backstory. I even seen some derivative pieces of that made now that you're a judge on the latest Solana Hackathon that's become a very iconic image in Solana lore.

Rooter (01:18):

Yes, it's funny.

Brian (01:20):

For folks who maybe aren't on crypto Twitter, would you mind giving us a little background on who you are and how you became to be founding Solend?

Rooter (01:28):

Sure. My background is in software engineering. Basically, I've been working as a software engineer for multiple years. Been doing smart contract development for a few years as well. Started off in the Ethereum space, did a startup in the past, which ended up getting acquired. And basically last year, I was looking at what to do next in crypto. And it was very clear that building something on a scalable version of Ethereum was the way to go because it would cost hundreds of dollars to do certain operations which would add up to thousands very quickly. Basically, at the time, Solana was one of the only alternative blockchains that provided good scalability and was actually live, since a lot of what was out there was just not live yet.

Brian (02:13):

And how did you settle on a decentralized lending protocol, in particular?

Rooter (02:17):

I was always quite fascinated by lending protocols. I read the Compound white paper when it first came out and was super excited about it when it first launched, I was a user. I always felt like there was something missing in the lending space, namely that not enough assets were supported. During DeFi Summer, there were all these different old coins launching like YAM, and PASTA coin, and whatnot. One thing I really wish I could do was short a lot of them or at least farm and hedge the risk, but there were no lending markets that wanted to list these coins because they're too risky and it imposes risk on the rest of the protocol. So I really felt isolated lending protocol was needed, which eventually came in the form. It was Arari launched Fuse. It was always like really want to see this in the market. The opportunity that Solana presented was perfect since basically, Solana was a fast-growing blockchain, and the ecosystem was rapidly coming together, and there was not yet any leader in lending. It just made sense that someone would need to lead that space, and why not be us?

Brian (03:27):

When did you guys settle on Solana as the chain that you were going to build on

Rooter (03:31):

At the very start, I just want to learn about different technologies so I was looking at Optimism, which hadn't launched yet. Polygon was live, aka MATIC, and I had dabbled around with that a bit. Basically, I want to mess around with Solana a little bit, see what the developer experience was like, and it turned out to be pretty good, which I guess some people might be surprised to hear. It wasn't terrible, terrible. We built the first version of Solend prototype during the Solana Season Hackathon, and we saw from that experience how powerful it could be and that there was a lot of potential. After that learning experience, we decided to double DAOn on it.

Brian (04:13):

That's great. And so you guys essentially went from a Hackathon project to today, one of the largest DeFi protocols on Solana, the largest lending protocol. You guys are facilitating hundreds of millions of dollars in loans every day. Walk us through that journey. What do you attribute that success to? And how big is the Solend team? You guys started, I assume, mostly just you in a Hackathon project. What's that journey like?

Rooter (04:37):

So in the start for the Hackathon, it was about four of us I think, and then we became five for launching the Mainnet, stayed around that size for a couple months. More recently we started growing so we're up to nine people now, mainly engineers still. And as for what I would attribute the success to? I would say, a lot of our success was on this pivotal moment which was when we launched our liquidity mining program and our token, that's when we saw a really huge spike in deposits. Basically, I think it was timed well.

Rooter (05:12):

Liquidity mining is just a really great way to bootstrap liquidity, which in lending you need that liquidity for it to be useful because if there's only a couple hundred thousand dollars of liquidity in assets, a lot of users just ... It's not useful to use it. If you're doing any size, you would be moving the rates too much. If you borrowed the rates would spike due to the algorithmic nature of it. I'd say that was definitely a pivotal moment since ... When we launched it I think we had $200 million in deposits, and basically, in one or two days we jumped up to $2 billion so that was quite a wild day for us.

Rooter (05:48):

Before that, there was a lot of anticipation to our token launch and our liquidity mining because I think we just did a good job with our branding and our UI. To this day, I think we have the simplest UI for a lending protocol that's the most easy to use and just straightforward. And, of course, coming from myself, take it with a grain of salt but I actually really do believe that. And we spent a lot of time going back and forth with a designer who is really, really talented, and coming up with this design that we felt was improving upon existing models. We tried to design it from the ground up, especially having used these products a lot. What are the pain points and what do we want to improve on?

Brian (06:28):

It sounds like you guys took a lot of your learnings too from DeFi Summer, as you mentioned that being a inspiration for you guys and how Compound's COMP token really kicked off that whole scene. I definitely agree that a well-timed liquidity mining event can really spur some pretty crazy liquidity in a bull market here.

Rooter (06:45):

And actually, I would add another thing, which is that we've built up this reputation of doing what's right by our users. Multiple times there have been issues, whether it's Oracle issues, where wrong prices are printed, and then there are some wrongful liquidations that happened. Or, at some other occasion, the Solana network went DAOn for quite a long time, which meant that some people couldn't pay back their loans when they wanted to, and as a result, they got liquidated so we've refunded people for all of these issues. And also, there was a vulnerability disclosure from Neodyme at one point, and we worked with them to get them a million-dollar bug bounty payout. No users have ever lost money due to anything that's out of their control really on Solend, and I think that goes a long way as well.

Brian (07:39):

I think that's a great segue. I think a lot of folks who maybe aren't users of Solend might have heard Solend for the first time back in June, this was around when much of the crypto market was experiencing cascading liquidations. Solend was trading around 25, $26 today, at the time of this recording in July, just a month later it's back up to $40. Around this time, there was a pretty big whale on Solend who was at risk of liquidation, and Solend initiated its first three governance proposals all in rapid succession. Could you walk us through what events were going on at that time in your own words? How the situation was affecting Solend?

Rooter (08:20):

I'll rewind a little bit. So basically, we first noticed this whales activity back in February when they deposited around $200 million worth of Solana at a time when we had around $2 billion. Or, maybe it was closer to one billion actually. Basically, they were a very large position but it wasn't anything super crazy, and also, they weren't borrowing anything at the start. And basically, what happened is, over time they started borrowing USD against their SOL position until it got to a point where they were borrowing $100 million.

Rooter (08:52):

At the time, it was almost 90% of the USDC borrowers in the main pool, and their so deposits were 95% of the deposits in the main pool. Over time, they gradually became an extremely large user on our platform. When we saw this, we tried to contact them so we did a couple of things. First, we went through our own private networks. Asked investors, "Hey, is this you or do you know who this is?" Didn't get any success there. The next thing we tried was sending an on-chain message and posting on Twitter a public announcement like "Hey, is this anyone out there, please reduce your position." I

Brian (09:33):

I remember seeing some of your tweets saying, "We're trying to reach you about your car's extended warranty jokes."

Rooter (09:39):

Definitely tried to use some humor to make it go more viral so that we could have a better chance of getting in contact with them. And the issue is that they're borrowing so much that their liquidation price was $22.30 cents, and SOL was trading at $26 and had just come DAOn from $100 just a month ago or something. If it continued to go DAOn, as was a trend, what we were worried about is that they would get liquidated which would cause such a massive amount of SOL to come onto the market in a sudden way that we were really concerned about the side effects of that. Specifically, on Solend, generally, the way that liquidations work is, there are bots that will liquidate someone and sell the assets on chain in one transaction. Basically, if that transaction is not profitable they're just not going to do it. If it is, then they will.

Rooter (10:37):

And typically this is fine because Solend has partial liquidations of 20% so only up to 20% of your position can get liquidated. And usually, positions are relatively small so it gets handled with ease. But this would be an unprecedented size of liquidation where basically, around $120 million would become liquidable in around $21 million chunks, which if you try to market sell $20 million, especially on DEXs on Solana, you're just going to crush the market.

Brian (11:09):

There's no liquidity to support that.

Rooter (11:11):

Exactly. If you try to sell one to $2 million, there would be 5% slippage. At 20 there's just no liquidity or tank it by maybe 50%, and doing this over and over again would cause a lot of issues. A couple of things there. One, it's a super lucrative liquidation transaction that Boss will be incentivized to spam so much to try to win, right. And two, it would create a very large arbitrage opportunity on the DEXs for people to ARVE between other venues. Both these combined would cause such incentive for a lot of bot activity that we were worried that it would potentially overload the Solana network. At the time, Solana was experiencing a lot of network issues, transactions would fail.

Brian (12:00):

This is the pre-1.10 release.

Rooter (12:04):

It's gotten so much better since then, thankfully.

Brian (12:06):

Was there any risk at this time to say just your average depositor in Solend, "I have a couple hundred USDC, I just want to deposit in Solend for a yield? Was I at risk here based on this mega whale's actions or failure to respond?"

Rooter (12:22):

So usually other users are not at risk, but given the size of this user and the fact that they could move the market with their liquidation, then other users were at risk because what could potentially happen is the Solana network could go DAOn, and in the worst case it would go DAOn for a couple hours. And in the worst case, the price of SOL would also continue to drop, and so liquidations would not be able to occur, or even if they did they potentially wouldn't be able to happen fast enough which would leave bad bet on the platform. Basically, when the collateral is not worth enough to cover the loan then there's a mismatch in liabilities and that would mean basically that some users who withdraw last would not be able to withdraw. Solend has an insurance fund to repay these bad debts, but there's a chance that our insurance fund of $20 million would get depleted, and on top of that users would lose a bunch of money.

Rooter (13:20):

Basically, what we were looking at is, do something about it and avoid gambling for the outcome where users would lose maybe $100 million dollars and Solend’s treasury gets depleted and basically spells the end for the land potentially, or do nothing and hope for the best. We're being proactive, and we looked at this situation and felt that it would make sense to do something about it. The solution that we came to was basically a very large market sale is extremely hard to do on chain but it's a lot more routine to do so OTC. So we made a governance proposal of this so then one, and basically the proposal was to liquidate these assets OTC, over the counter, to minimize flippage, get better execution, et cetera. Minimize impact to the market rather than doing so on chain. There was a lot of controversy around that.

Brian (14:20):

And I just want to set the stage here a little bit because I think this is a really fascinating case study for everyone who's interested in decentralized governance. I've been in the space for a couple of years now. Maybe thinking back to March 2020, but even then having governance issues that were so important. Solend being the lifeblood of the lending ecosystem on Solana, this being like you said, a potential existential question, but then also the fact that it's time sensitive. Can you just recap real quick how long you guys had from essentially finding out that this was an issue to turning out your first governance proposal?

Rooter (14:56):

So I think we first noticed this on June 15th, and a couple tweets trying to get in contact the 16th and around that. And then it was a couple days later that this proposal happened. And basically, we were looking at the sold charts and it's a very volatile asset, right. It actually dipped to $25, and I think in the high 25s for some time, which meant that only around a 15% drop would cause liquidations to happen, and doomsday scenario could happen. For SOL which dropped 80% in a very short period of time, a little 15% dip is not unfathomable, right.

Brian (15:41):

Walk us through some of the strategies here. You mentioned the first governance proposal involved taking control of the user's account in the event of a liquidation to handle this over OTC. The thinking being that OTC desk can provide better rates than the liquidity that's available latently on decentralized exchanges, there'd be less impact on users. Walk us through that proposal and then some of the subsequent proposals that followed.

Rooter (16:08):

That was essentially the proposal is to temporarily move the assets such that they can get liquidated over the counter and then move them back into the user's account. One common misconception is, people thought that we were taking their funds or stealing them or what whatnot, which is definitely not the case. Basically, they were just to get liquidated. And that money is still their own money it's just would be converted from SOL to USDC with the best execution that we could find. Better than what would be on Dex's. Of course, there's a lot of controversy around the method of doing that. Following that, there's a ton of controversy. And thankfully actually, the price of SOL started to recover which bought us some time. And re-evaluating the situation we thought okay, now that we have more time we can consider some other options potentially.

Rooter (17:01):

We made a second proposal, which rolled back the first one. We were listening to our community and the general community at large of crypto. It was definitely unpopular. Although one interesting fact actually is, it was definitely very controversial in the general sphere, but for our users that had money stuck in the platform and who were just waiting for a train to come and hit them, they really wanted us to do something about it. And we would get a lot of messages like F the haters, just do something. Don't listen to them. Just try to appease a bunch of people on Twitter and let us lose our money.

Brian (17:41):

And these are messages on Realms, the voting platform, is that right?

Rooter (17:44):

Realms and in our Discord just talking to users directly. One additional piece that I forgot to mention earlier was that, due to the utilization of the pool being so high because all of the USDC in the platform was borrowed out. That meant that people couldn't withdraw their USDC anymore because basically, the funds are not within the platform

Brian (18:08):

They're not available. Right.

Rooter (18:10):

Users were stuck in their position. They were frozen in and they couldn't exit so that was an additional thing that was exacerbating. The whale was causing real problems, it wasn't just potential.

Brian (18:20):

The borrow rates must've been crazy then.

Rooter (18:24):

I think USDC was at 60% APR and USCT went up to 200 something.

Brian (18:31):

Wow.

Rooter (18:31):

It was quite crazy. Some people were happy to collect that interest, but definitely, I think most people were sweating a little bit.

Brian (18:40):

Understandably so. Okay. You mentioned that the first proposal was around the OTC liquidations, in the event it came to that. Luckily, it didn't come to that, the price rebounded, then you guys initiated a second proposal which essentially gave your team more time to reevaluate, introduce the minimum one-day voting period. And then you guys also launched a third proposal which passed. Can you walk us through a little bit about what that third proposal entailed?

Rooter (19:07):

So the third proposal basically implemented new liquidation rules for extremely large accounts so basically, it capped the borrows of any account to $50 million. It would start off the cap at $120 million and gradually decrease it by around 500,000 per hour. Two things there. One is we don't want any single user to be a systemic risk for the platform and so there should be some sort of cap. I think this makes sense and there's pressing for it.

Rooter (19:39):

If you go to a bank and you ask for 100X leverage on a trade, or if you go to a crypto exchange like FDX or Binance, for a very long time you could do 100X leverage if you had say $100 dollars in your account. It's just a gamble, it's a lottery ticket or whatever. But if you go to them and you have $100 million and you tell them, "I want to get 100X leverage," they're going to tell you no, right. They're not just going to give anybody that crazy amount of leverage so you have to consider size. In the same way here, we don't want extremely large users until the platform can absorb it. As Solend grows, we can increase these caps.

Rooter (20:16):

And then the second part was this gradual reduction. Basically, the intent there was to spread out the liquidation over time so that we don't get a sudden $20 million sale that could cause chaos. We have them in much smaller chunks, such that the liquidations could get absorbed. If you sell a little bit on the decks it'll cause some slippage but not a crazy amount, and then it can get armed with centralized exchanges and other parties. It'll happen over the course of a couple days rather than just a couple hours or less.

Brian (20:48):

This third proposal passed as well, this was the latest proposal. Thankfully in hindsight, none of this actually had to come to the test. SOL rebounded, I assume the whale is no longer in imminent risk of liquidation. But I'm curious because you mentioned throughout this, one of your guiding principles is doing right by your users. You guys have done a number of different initiatives outside of that to prove that. And then you also mentioned your users were among the most vocal asking for you guys to take action here. I'm curious, where does governance come into play? What is a good time for enacting governance as a protocol founder? What decisions can you guys make as a team? What times do you actually need to bring in the community vote in situations like these?

Rooter (21:32):

One last thing to wrap up the whale thing. In the end, we actually were able to get in contact with them. Basically, they heard the news and someone had reached out to the Binance team ... Or, we had reached out to some people who helped us get in touch with the Binance team who then forwarded our message along to the whale. Shortly after proposing SLND3, we got in contact and talked about some mitigation strategies. In the end, they reduced their position on Solend and the price of SOL rebounded, as you mentioned. We were in the clear after that. Moving on to the question about decentralization. Basically, my thinking there is that it's better to build something worth decentralizing than to decentralize nothing. And it takes time to get to that point where you have something valuable that's worth decentralizing. My thinking basically is, if you're fully decentralized from day one it makes things extremely hard.

Rooter (22:26):

I don't know if you've participated in many DAOs, but if you have you would know how extremely inefficient they can be and how oftentimes there's a lot of this bystander effect where people do nothing, just stand around waiting for others to do something. I believe that there needs to be some spearheading entity that gets things done and then gets it to a point where it can be decentralized to the community and governed by the community later on. But the whole startup maze of discovery and pivoting to find product market fit and reacting to changing market environments, that's very hard to navigate as a DAO.

Rooter (23:04):

I don't remember who put it out at the beginning, but this idea of decentralization over time, and we're embracing that. We are working on a decentralization roadmap, which we're going to release soon. And basically, it'll outline what are the milestones on the way to decentralization? And what are the steps that we're going to take? I mean, even Bitcoin wasn't decentralized on day one, right. At the beginning, most of the hash power was owned by Satoshi and his associates and whatnot, his friends, and, it took a while for the hash power to be, as we would call it, sufficiently decentralized.

Brian (23:42):

Is there anything that you would do differently looking back on this whole situation?

Rooter (23:46):

That's a very tough question. Looking back in the situation that we were at, we were in between a rock and a hard place, and we were faced with a real-life trolley problem where one option is we gamble with our users' funds and potentially let them lose hundreds of million dollars and the life of Solend. And the other is doing something controversial that doesn't sit well with a lot of people. It was very tough and definitely hope to never have to make this decision again, but I think what we would do is we would always put our users first I think. Having users lose money is the worst thing that could happen so we would do everything we can to prevent that from happening.

Brian (24:28):

I commend you for your composure in responding to that situation. A lot of people like to chirp on Twitter. I think a lot of people who are working in Web 3.0 or founders of projects know what it's like to be in the arena there, and definitely don't envy your position but also, I think it's great that you guys have that principle of doing right by your users.

Rooter (24:50):

That was definitely tough. There was a lot of criticism directed directly at us as well. And the frustrating thing was, none of these people were users. None of them had ever touched Solend's, maybe never even touched Solana, but they were just criticizing from the sidelines. A lot of them had something to gain sort of. Maybe they were maxis or they have a platform where controversy and engagement drives metrics for them. It's an ugly reality of things.

Brian (25:23):

One last point on this. For any founders who are thinking about starting their own project in Web 3.0, generally, maybe listening to this podcast, is there anything you would tell them about governance as a whole or what you've learned so far in your time at Solend?

Rooter (25:37):

It definitely helps to be proactive and to set things up, and discuss things with people as early as possible. But at the same time, I think if you're just starting out a project, I wouldn't suggest to spin up a DAO immediately unless it's a DAO-specific project like X DAO or whatever. NOODLE's DAO or some fan club. If you're working on a startup, which happens to be crypto-powered, I would definitely suggest to figure out your place in the market first, figure out some product market fit before spending too much time decentralizing. Because if you spend all of your time and effort on the DAO piece and your startup ends up failing because it can't find product-market fit, then what's the point? That would be my advice.

Brian (26:25):

Well said. I want to switch gears here. I know we've been talking a lot about your guys' plans with decentralization over time. I think that's really exciting. You guys also have a number of other product launches though related to Solend. And recently, you guys announced the launch of your isolated pools product. Can you give us a quick overview of what these isolated pools are? I know STEPN has been a big example here. Can you walk us through how end user might interact with some of these pools?

Rooter (26:51):

An isolated pool is basically a separate pool of assets that can be cross-lent and borrowed. And this is in contrast to the main pool that we have on Solend, which was the first one that we launched with, and every time we list an asset it would go into that main pool. But the problem with that is, if you list an asset with low liquidity that is, therefore, more easily manipulatable, it opens up the entire pool to attack, not just that asset. To take one simple example. Let's say there's some token and an attacker was able to manipulate the price such that Oracle's believe that it's a million dollars a coin, then a user could deposit some of these tokens and borrow millions of dollars worth of assets against it. Millions of dollars worth of USCC, SOL, et cetera, and then just let the price fall back DAOn to say pennies and just walk away with the debt not intending to ever repay it.

Rooter (27:50):

Basically, yes, you need to be very careful about which assets you list. And everything in the main pool has to be of extremely high-quality, and not manipulable, and not just mintable willy-nilly. If someone could just print an infinite amount and deposit and borrow against it, that's also a big issue that we want to avoid. What isolated pools does is, we can set up these separate pools that we can then list riskier assets. And if there is an issue, it would be isolated to that pool. So we're not saying there's not going to be any issues ever, potentially at some point there would be issues, but users are opting into that risk when they enter into an isolated pool.

Rooter (28:30):

Some of the interesting things that we've done with these in the past is one, it's called the Turbo SOL pool so it's a pool with only SOL and USDC in it. And basically, since we have only these two really high-quality assets, we're able to increase the loan-to-value ratios for these assets. So rather than being 75%, so you can only borrow 75% of the value against your collateral, here you can borrow 95% which lets you get 10X leverage. So that's a pretty interesting use case. If you only want to leverage trade SOL versus USCC then you're better off using the Turbo SOL pool where you can get higher leverage or you can just have a lower liquidation price.

Rooter (29:13):

And then the other side is, listing long tail assets. One very interesting one as you mentioned was a STEPN pool. We have this pool where there's GST, which is the token that you earn from walking, and GMT, which is the STEPN governance project token. This has been an extremely popular one where it was growing a lot. The number of users in this isolate pool was our fastest growing for some time. And it opened up some interesting use cases which showcases the power of DeFi and composability. So basically, STEPN just launched their token, they don't have anything to do with lending markets or whatnot, but we are able to just launch that permissionlessly and provide a product for users to use to do a couple of things.

Rooter (30:01):

One is, they can hedge their entry to STEPN. Rather than paying $500 for a pair of shoes to start walking, you can borrow a bunch of GMT or GST and sell it to US dollars and then buy the shoe with that. And what that does is, if STEPN as a whole doesn't do well, for whatever reason, and the price of its shoes go DAOn, probably the price of GST and GMT is going to go DAOn as well so you end up saving that money since you didn't make such a big upfront investment so you can hedge your entry costs. And another interesting thing you can do is you can borrow GST to level up your shoes and pay it off later so it's level up now pay later. Buy now pay later.

Brian (30:49):

Buy now pay later but for STEPN.

Rooter (30:50):

Exactly.

Brian (30:52):

That's awesome. How have you guys seen this DeFi component fitting in with the general onboarding funnel for something like STEPN? I think STEPN's fairly unique. It brought in maybe a lot of users who weren't familiar with crypto for the first time. Do you see this as something that's more of an advanced feature today or could this potentially, in its own way, be a bit of a gateway drug to onboarding on the DeFi directly getting new users who might not have ever had a wall before to open up wallet and deposit in Solend because of the yields that they see?

Rooter (31:24):

It's definitely a more advanced feature. Users that use this have to be aware of their liquidation risk and they have to manage that which is pretty tricky and ideally would understand how markets work. What's an order book? Just those basics. I think a lot of people in crypto take it for granted because they've been just breathing it, eating it for breakfast every day for so long, but it does take some time to learn these concepts. The one beginner feature is you could just deposit your GST or GMT and lend it out for yield, that's pretty simple. It's good to understand the risks involved as well, but that's a much simpler product than borrowing against and managing liquidation.

Rooter (32:04):

And, by the way, the interest rates on GST and GMT were historically extremely high, something like 500% API which made it very exciting. And that definitely brought on a lot of people, which a lot of them were beginners to DeFi, hadn't really used crypto products much but were attracted by those very high yields, especially those that were holding GST anyway because maybe they were saving up for the next big purchase and why not earn 500% API per year?

Brian (32:33):

That makes a lot of sense. 500% though, I think that starts to ... Maybe some folks who aren't familiar with crypto, that starts stirring a too good to be true. For those who were around in DeFi Summer, that's small fry numbers. I think a natural extension of this concept of an isolated pool gets at something that you guys have hinted a little bit called permissionless pools. Is there anything that you guys can share about this?

Rooter (32:58):

As I mentioned a while ago, when I was using lending protocols and then DeFi Summer, there are a bunch of things that I wish there were lending markets for. I really wish that I could just list my own, but it's a lot of work to start a whole lending protocol, and get usage, and get network effects, and whatnot. As you mentioned, the natural next step from isolated pools is allowing anyone to list their own. And this has been something that I've wanted to get to for a very long time and so it's very cool to finally be seeing this come to fruition. Basically, quite soon, maybe by the time this pod airs, we're launching permissionless pools, which anyone can list whatever asset they want. I think it's quite a powerful concept since on day one that an asset list gets launched you can have a lending market for it so you can short anything on day one, you can leverage long anything on day one, you can use anything as collateral for a loan on day one, all powered by the community.

Rooter (33:57):

Going back to isolated pools, we've been launching one isolated pool about once a week, but even then we only have ... Right now we have 40 something assets and around 16 different isolated pools, but it's going to take us forever to list everything in Solana ecosystem, right, there's thousands of tokens. And at this rate of once a week, even though that's a pretty decent rate we're just never going to get to everything. I guess borrowing some lessons from Uniswap, they have permissionless listings and it made sense that they could list everything under the sun without having to spend any engineering time on it.

Rooter (34:34):

And in contrast to that, centralized exchanges like buying that through Coinbase, they have to spend maybe a week of engineering's time to set everything up, especially if they have legacy systems that are not designed from the start to be rapid listing machines. Coinbase, for example, as a startup, it was only for buying Bitcoin, and then they added Ethereum later on. And then now there's hundreds of assets but it took a really long time to get there, and it takes up a lot of engineering time. In this way, we're able to provide a market for everything on Solana, and we're removing ourselves as a gatekeeper and as a blocker for having these markets.

Brian (35:13):

That's really cool to hear. I do think that is one of the main selling points of a decentralized protocol like this is just capturing the long tail of all assets in a really efficient way. I think this dovetails pretty nicely too with what you mentioned with this hint at what your guys' plans are for decentralizing long term. I guess looking ahead, what are you most excited to build with Solend? What do you envision is the end state for how Solend fits into the broader Solana ecosystem?

Rooter (35:40):

One of the big parts that we're encouraging is developer usage. Solend at the end of the day is a platform. We have Sea Tokens that make it extremely easy to integrate with. Sea Tokens are just like any other SPL token that represent your deposit. And basically, we want Solend to be pretty deeply integrated into the ecosystem and be used across various different use cases. Lending is a core primitive of DeFi and it's a building block that should be used as much as possible. I really see ourselves at the base layer of Solana as a primitive that's used in many different other products.

Brian (36:21):

That's awesome. Well, I think you guys are definitely well on your way to being there. Rooter, this is a great discussion. Thanks for your time. One last question we ask to all our guests. Who is a builder that you admire in the Solana ecosystem?

Rooter (36:34):

I really admire the Orca team. Yutaro and Grace inspired me to build Solend right at the beginning because ... I knew Yutaro from his Orion days and I knew he was a very talented dev. Seeing him jump into Solana was definitely a factor for me to take a second look at it. If you haven't already had them, I would definitely recommend them.

Brian (36:58):

Couldn't agree more. We had Orion for I believe our third episode that we launched. They do a really great job of setting the UX bar very high in crypto, especially when it comes to DeFi projects, as you guys do as well. Well, Rooter, this is great. Thanks so much for your time. Where can people go to learn more about you and to learn more about Solend?

Rooter (37:18):

So for myself, you can follow me on Twitter, I'm at 0xrooter, most active there. For Solend, Solend.fi is the website and Solendprotocol on Twitter where you can find all of our updates. From there you can find all the links to everything else like our documentation and whatnot. Developer, portal, and whatnot.

Brian (37:38):

Perfect. Thank you, Rooter.

Rooter (37:39):

Thank you.