Ajna: Lending and Liquidations Without Oracles

0:00 Welcome back to the Permissionless podcast brought to you by the team at Guardian. And of course, here we're going deep with D5 founders. We're figuring out how did they find their way into DeFi? What got them excited initially? Why did they start building? What are they building towards? And of course, what keeps them up at night? What are the struggles that they're going through as a DeFi builder in today's market?

0:22 And today we've got an a really exciting conversation with Ian from Ajna who broke down all of the interesting design decisions and their thought process when they were designing the Ajna protocol. And Ajna is introducing Oracle loans and oracleless liquidations in a completely permissionless way. And as you'll see in the conversation, we actually got to a lot of interesting use cases and functionality that otherwise can't be achieved or or can't be achieved in a decentralized manner as easily that they almost sort of stumbled upon when they were building out this protocol in the way that they designed it.

0:58 So, it's got a very interesting design and and really a novel approach to borrowing and lending that we got deep on and we got to learn a lot of the considerations that they had to make in that process. But of course, we'll go much much deeper on that in just a second. My name is Owen and over three years ago.

1:15 I founded Guardian and we're driven based on the premise that DeFi will engulf trillions of dollars of assets in the coming decade. And so we set out to build the smart contract security and the full stack security solution that would allow teams to ship confidently, consistently, and within a reasonable time frame. In this environment, we use two competing audit teams and a smart contract fuzzing engineer to perform advanced invariant testing and give teams a systematic process to reach production with as high confidence as possible.

1:47 Many projects get an audit and then proceed to get hacked months or even sometimes years afterwards. And the Guardian model is here to put an end to this. Today I'm beyond proud to say that we have achieved the standard that is necessary to onboard trillions in TVL into DeFi. And at this point really we're here for DeFi to catch up.

2:08 And now without further ado, let's get into the conversation with Ian from Ajna. But yeah, dude. Well, I mean, first of all, appreciate you taking the time to to hop on and looking forward to learning more about your experience in DeFi. what you see as is as as being you know the cool stuff that's being worked on right now and learning more about some of the biggest problems and challenges you guys have at a but yeah I mean really first of all interested to learn like what was your your pathway into DeFi?

2:39 How did you get into DeFi? What did that look like for you? Yeah, like all people a friend introduced me to to crypto and and to Bitcoin. This was a long while ago back when the smart contract ecosystems were just kind of getting their first start and Ethereum was getting its first start and so got incredibly interested fall down fell down the rabbit hole just like everyone does.

2:59 And he got me really into Maker. So just kind of started learning and learning about Maker and better understanding it. And that's kind of where I started my my DeFi journey and crypto journey generally was working out of Maker. So I got picked up by them and and started working on the market maker keeper which is like a open-source marketmaking software that was originally used to assist and and support the peg before like other instruments came online like the peg stability module and stuff like that for maker and yeah I've kind of fallen in love with DeFi and crypto ever since.

3:34 I mean, I think it is incredibly powerful. Technology, the composability, all those elements of writing this software and using it is is super addictive and it never stops. So, so yeah, it's kind of been my experience. I've been mcking around in this space now for six years or so and had the fortunate opportunity to work with some some great people along the way.

3:53 Yeah, dude. There's something about mastering Ethereum, the book, that's like so addictive. I don't know why or how they've like specifically engineered it, but exactly the same. Yeah. Yeah, when I first started reading like mastering Ethereum and playing around with smart contracts, it is for some inexplicable reason very addictive and it's hard to shake. We're here like half a decade later still working on this stuff.

4:13 So, they got something right about the developer on boarding there. But Maker is a tough protocol to to start out with as like your first experience. What was the onboarding like there with, you know, of course they've got all their jargon. What was that like? yeah, it it's tricky. the the Taiwanese the pill to swallow is really tough.

4:33 You know, it's I think it's a good sign when you get a really opinionated developers who write code and decide to use language like that in the sense that it it means that they're really passionate about what they're building and that obviously just results generally into better software. Yeah, there's a lot of contention on whether or not it's actually beneficial generally for for the protocol and people understanding it, but it's also a bit of a right of passage as well.

4:55 I think the protocol itself just being one of the first stable coin protocols or I think it's I was actually the first on on Ethereum I forget that sets a great tone for just getting to better understand the different tools and things you can play with. And then of course the the team at Maker back in the day and even now I'm sure although I don't follow it all that closely now was just huge heavy hitters.

5:18 And a lot of people who kind of had really interesting ideologies when it came to how to best use this software and kind of the ideals they believed in in terms of what it allows humanity to do and how it frees us from some financial institutions. So it was kind of more of the pie in the sky idealist group back then when I was there.

5:39 Things have have obviously kind of changed over time like all projects do in morph but a great great introduction nonetheless and you know you get to play with with some really interesting code and a great protocol and then you know writing wrapping software around that. So working on keeper software I worked on the auction system as well helps you just better understand like the importance of design when you build these systems and how to make them easier to interact with and the different honestly the amount of work that's required to to support smart contract systems is was really eye opening.

6:16 Yeah. And it it suited me well. Yeah. I think it's really cool how they almost encoded at least a part of their initial culture into the codebase and into like immutable smart contracts. Kind of cool how that has sort of played out and obviously it makers continues to to use the same sort of approach and and build on top of the same vernacular.

6:35 So, it's cool to see that sort of persisting even though of course, you know, we've the maker team has had, you know, a number of people leave or they're working on other stuff now, but new folks are picking up the reins and introducing new words, which is cool to see. But yeah, I mean, I was also interested, have you guys I mean, obviously this adds like a speed bump when it comes to onboarding developers and maybe even like external auditors.

6:59 What's the experience been like? because I don't think any other protocol really has like a similar sort of approach. So is there anything and obviously I don't think you guys use anything similar at so you've seen both sides of the coin is like having you know what most would consider a normal vernacular versus maker's glossery. What kind of outcomes did you see?

7:20 How was the experience for like onboarding people audits etc. you're saying specifically into Ajna you mean? Well, I mean maker versus Ajna like Oh. Oh. Oh. If there are any versus Yeah. So, I didn't work on the contract code itself at maker. So, I can't really speak to the difficulty auditors had. I think just based off of being around those people in in kind of chats and exposure.

7:45 I would say there's always a steeper learning curve to using Taiwanese or like a very customized specific vernacular to your codebase. And so that initial ramp up period is more costly. However, once it's saturated in your brain and it clicks and you're can think about the repository, it results in significantly simpler code because your variable names are always I think it's like four or five characters.

8:11 So it just results in really clean code that's easier to think about and work through. So it's a bit of a trade-off, I think. in and the trade-off is like cleaner code, stronger understanding, but then that ramp up period is more costly. So, in thinking in context of audits, like I would assume a contest would probably be less helpful for that codebase because you're getting a lot of new people who are originally exposed looking at the code for the first time have to undergo that hurdle of learning.

8:39 Whereas if you have, you know, a larger auditor that you've paid contractually or maybe even have a retainer on, you know, that brain has already been saturated or those sets of auditors have already been saturated and know the codebase. So their ability to navigate in that in a more simple codebase could be beneficial. So I would say that that's kind of the trade-off.

8:57 Yeah. No, that definitely makes sense. It's almost like kind of more purely translating the code into like a a mathematical formula almost where it's like like fraud is X or you know stuff like this. So it's also nice for forcing people to think abstractly in a sense like releasing all of their preconceptions about what a liquidation is or something like this.

9:19 So I think in a sense that's that can be kind of a plus. It's like you're thinking outside of the box. Yeah, I agree. From auditors and also from Yeah, it forces you to leave your your preconceived notions behind you. Yeah, absolutely. I mean, even though it is kind of ironic because there were no preconceived notions when Maker originally created what they Yeah, of course.

9:40 But I mean, speaking about like, you know, liquidations and all this stuff, obviously you guys have also thought outside the box and and sort of rearchitected and re-engineered what lending and liquidations and stuff like this can look like at Ajna. I mean what was the catalyst for this? Was it from your experience at maker? Like what was really the seed that became Ajna?

9:59 Yeah. So I think gosh it was like two two years ago or three years ago now that kind of the Ajna journey started. The landscape was a little bit different back then. So I think back then you know morpho didn't have its foothold or it still had optimizers which were a great a great innovation but I don't it didn't have its Morpho Blue base protocol wasn't released yet.

10:22 So the idea a couple novel ideas were kind of running through our heads. and again it was just the team that I was working with at Maker who was working on the the auction keeper software and Markmaker keeper software. We kind of split off and said well you know there's a lot of oracle manipulation that's happening anytime people try to introduce new collateral types.

10:41 We don't love that and there's also a lot of shared risk that occurs in a lending protocol. And so we had to kind of think from first principles and try to figure out ways to build a lending protocol that is oracleless, so it doesn't have a price feed. It's permissionless and that any type of collateral or quote token in our is how we describe it or a debt token can work together.

11:01 So an isolated risk pool and it's immutable so the code can't change or there aren't governance to step in and adjust parameters inside of that protocol. A lot of that inspiration came from like a Uniswap v2, you know, Uniswap v3 kind of designs where we just said, "Wow, this is great. These protocols get to access the Lindy effect and they're around for such a long time and if they're, you know, straightforward enough and other people decide to build on them, their security is always their assumed security is always greater over time."

11:37 And we really wanted to capture that. And so Ajna was kind of the result of that. And the design decisions we made in terms of how lenders interact with borrowers in terms of auctions, how profits are distributed to the token hold the governance token holders. All of those things were kind of out of those those fruitful discussions and brainstorming.

12:01 And then we decided on to head out on our way and and and build the thing. and and ash knows what we came up with. Yeah, I mean well first of all I mean obviously there's a lot of operations at Maker and obviously you know lending protocols like a I mean there's just a ton of resources that go into risk parameterization you know they've got entire firms that are like multiple firms at one point that were you know both co-working on like risk parameterization and stuff like this so I mean to remove that and kind of just well first of all put it back on the users in some sense back on the lenders and just remove all that complexity.

12:38 It's definitely a a step forward or you know a step adjacent to what the typical like approach to lending is. So that's a really cool adaptation that you guys have made. Do you think like okay so it's been out for a little while now. Do you think that that kind of approach has stuck to what you guys were expecting?

12:59 Like how's the reception of that been in terms of the goals with which we wanted to meet and the purpose of building Ajna the goals we set and we executed upon Ajna fulfills those goals I think the tricky part with Ajna and you just hinted at it is some of that complexity is offloaded to the user and that's something we're working actively on right now to try to kind of better better resolve I think as we move forward in time these applications should their user experiences should I ideally just be far simpler and more delightful for the user and have forcing them to to learn or invest a significant amount of time to to learn how a you know the risk grammarization works in a lending protocol or how it interacts with price fees or

13:45 Whatever is kinds and so we've we've acknowledged that and I think that that's kind of been our primary focus is really hunkering down and figuring out different ways to ideally resolve that by building other software on top So there's another team who we've worked with in the past and and help us did audits. Their name is Protote.

14:05 They're a bunch of Xmaker smart contract guys. They have actually been working on another piece of software referred to as like an arc. And I think that hopefully here that'll be that'll be coming out shortly. But but that essentially just totally simplifies the lending experience. So we don't force people to have to undergo that that treacherous sometimes treacherous process of learning Ajna's ins and outs as deeply.

14:28 You can just deposit and it's kind of providing more of a set and forget vault-like experience. So that's been our primary focus. I see. So would that be almost sort of akin to like you've you've mentioned Morpho like they have sort of like you can choose which lending pool you want to go to based on you know who's setting the risk parameters.

14:45 Would it be sort of like a more managed approach like that? Exactly. Yeah. So Morphos's design has those managers and then the managers get to determine what collaterals the depositors deposit into like which pools subpools. ARC's design is that the collateral types are set on deployment. So you don't have to worry about the manager making a decision that you might not agree with initially, right?

15:11 So the manager will have their reasoning. They're going to deploy their their pool and then you as a depositor get to jump in there, but you don't don't have to worry about them all of a sudden switching it up on you. So to think about it from a risk perspective, you're you're trusting their original analysis. You can read about their analysis if you wanted to.

15:29 And so you use the user set and forget and you trust them. But then from that point forward, you don't constantly have to stay up todate on their analysis if it is changing, right? If it does change, they could just as easily create a new type of vault and then users can now opt into that. So it's again hopefully trying to access that Lindy effect, right?

15:50 You're kind of like this doesn't even necessarily directly relate to Lindy. doesn't have to directly do a security, but you're putting your risk trust in one person and then the the ground doesn't move underneath you, which I think is just generally a really important principle when you build these kind of apps and and applications. Yeah. And it's so easy to stray away from like what DeFi was originally intentioned to to provide with applications and especially with lending applications where you know it starts to become like with some with some protocols it's it's almost like okay well basically this is just a hedge fund.

16:23 Yeah that happens frequently. You start to see those trends for sure. I yeah it's it's hard to resist like the entropy of just falling back into a hedge fund which you have to actively like work but yeah I mean I think that's a an excellent way to you know of course we need to have some kind of automated oversight otherwise a becomes a a protocol that's really only palatable to like institutional managers right and I think with this move you guys are becoming much more applicable to like the everyday user that just wants to come and lend their assets.

16:57 Maybe they want to limit their exposure to like a Chainlink feed manipulation or something like this, but they also they still want to, you know, be able to lend out their assets and and earn interest on that. Yeah. And of course from the the borrower side as well, you guys have, you know, a great value proposition there.

17:13 But yeah, I mean, I think that definitely makes sense. And have you seen anybody else like trying to sort of attack the problem the same way that you guys have? I I don't know that I've seen any, but are you guys aware of any? I mean, I think Morpho is closest to us. I think the big caveat there is oracleless lending.

17:27 Yeah, I have not seen that. Yeah, and and I can opine on that a little bit as well. I think oracleless lending is is really tricky. I like Uniswap v4's approach as well. I think when you start with the core component of an amm and then you add hooks, you can do some really interesting things. I think that we're going to see some interesting use cases there as well.

17:47 I think when you operate without an oracle, you're trying to like Ajna's use case was we're trying to figure out how do borrowers and lenders interact with a protocol in such a way where we're determining the price and then we use that price in terms of the lender interactions. so doing that just by nature adds a lot of complexity and so figuring that balance of how do we limit complexity still provide that utility of you know totally getting rid of price feeds and the reason why we even set out to do that is we wanted to be the first to grow that liquidity pool.

18:25 So liquidity often works as a moat in these D5 lending protocols or any DeFi protocol generally and so the first to build that liquidity pool can offer better rates. it can offer more diverse pairings, all of those different things. So, if you don't have an oracle, great, you're the first who can start that liquidity mode. But then the caveat there is you need to ensure that once that token if it does grow into being, you know, a blue chip or grow into start being a blue chip and does have a price feed that people trust, can you still offer a lending experience that's relatively simple that they can still interact with and understand in order to compete with lending protocols that do have Oracle price feeds?

19:09 So that's kind of a bit of the the trick. I think a is headed in the right the right direction there. But it's again a very difficult problem to get right. But I think Uniswap is an interesting example of of how you can kind of combine the AMM elements with a lending protocol. And actually kind of does that as well.

19:28 You can actually swap inside of the protocol too, which I could explain, but that allows the price to be formulated by actors interacting inside of the protocol versus having to use an external price, which is manipulated or can be manipulated depending on external liquidity. Yeah. And of course, I mean, just thinking about it, would would you guys explore like using a UNIV v4 as like a backend?

19:49 Because I almost imagine you could instead of like what you guys are you're doing now is you have an auction and you you let people arbitrage it to basically liquidate positions. But I mean almost that would be the same thing as if you had a UNIV4 pool spun up for the asset and then you just let people come and arbitrage it and once it passes a tick then you just get a call back and then you liquidate positions based on that.

20:11 Is that something you guys I mean of course you've already deployed the protocol and it's immutable at this point but would that be like a V2 for you guys that you're maybe considering? Yeah, we haven't really been we've been kicking around some ideas for V2, but none of it's really set in stone. I think that there are other complexities that comes with hooks like from a security perspective that make things a little tricky and there's just a lot of stuff to manage inside of that.

20:37 But admittedly, I haven't dug very deep on hooks. I just know that something there's some similarities there in terms of how the two two models look. So, no, I can't say that we're, you know, thinking about building on UD4 or anything like that as of yet. It's a definitely interesting idea. There's something there and people will take advantage of that.

20:54 Yeah, I mean, just off the top of my head, you would probably run into issues where like obviously right now you require a bond to be put down for liquidations. Obviously, if somebody swaps in the pool and like sends it to Tech Zero, you can't just liquidate everybody and does like a flash swap. But yeah, we've seen some some pretty interesting Uni V4 integrations and really I think in the coming years we'll see how far hooks can be pushed.

21:16 Yeah, it's an interesting model. It's almost like you can take that sort of like if you look at what the UNI V3 model is is it's kind of similar to what you guys have which is an accounting system where you want to spread out liquidity or you want to spread out assets to be borrowed across some range in certain buckets and UNIV4 kind of allows protocols to just like borrow the back end of Uniswap for that accounting and totally abstract that away.

21:40 I mean abstracted to a certain degree there's still like a lot of details to be studied about hooks and if you do integrate Uniswap v4 it will kind of dictate the architecture of your application but yeah I think that's just in general I think we'll see start to see a lot of new v4 integrations that sort of just leverage that as an accounting system you know beyond even just like normal swapping and stuff like that which should be pretty interesting to see but you mentioned like this is something I'd never considered like you want to get to the asset before it gets big and before it gets listed by oracles and before all these other lending protocols can realistically support it.

22:16 So that's kind of a like a keen like strategic like tactic to to get the liquidity before anybody else can like technically even support the project. So you guys have that capability, you have that edge. Do you specifically target like longtail assets for that reason? Well, yeah. I mean I that's that's why Ajna's design is the way it is is because we set out to support Longtail.

22:39 That being said, the TVL that Longtail offers is near zero. So it is very much about being the first to the market and then growing with the protocols that decide to incorporate Ajna or the collateral types that want to jump in and help support pools. That's a success case is it really grows along with the protocols that that decide to kind of put their focus and and energy towards providing liquidity through a treasury or anything like that.

23:08 Yeah. Yeah. So you want to attract like find those hidden gems and then or ideally they come to you and they just know that is the place to to spin up a borrowing pool for those assets and and then you want to retain that liquidity. Yeah. And and obviously like the best rates and everything like that that comes along with it.

23:25 So yeah, that that definitely makes sense. And I guess what would you say where is the next goal for Ajna? Like where are you guys heading over the next few years that you're sort of steering towards right now? So right now I think the main focus is has always been just how do we really simplify things from a user experience perspective and when the next version you know comes out or or as we work through it I think the big the big focus will the buck will start and stop there.

23:53 If you can't make something that's that's really usable or simpler for people to use then you just don't get the same set of adoption. So that's our primary focus. and I think again a lot of that will be accomplished with this this release of ARCs which I'm super excited for. I think that it'll offer a lot of that utility.

24:10 That's where a lot of our focus lies to be honest. I was just going to ask have you felt like the road map or the vision has it changed or like what's the difference when you started Ajna versus going into ARCs now? Ajna's initial scope was a little large. For those who don't know, can do a lot of things, right?

24:25 So, you can you can borrow and lend. You can actually perform swaps. So, at each tick price that lenders deposit liquidity, they act kind of like a mini vault is the best way to think about it, right? So, it actually works as a mini pool. So, people can come in and swap debt token or collateral token inside of each Unis swap price bucket.

24:43 So, it does swapping. Another thing that isn't widely understood is we actually support NFT lending as well. So you can actually lend against NFTTS and you know we have the auction system built in. We also have a a fairly sophisticated governance system as well. And so I think we just bit off a lot. I think rep prioritization is how you get back to the basics and deliver like good strong utility to a lot of users which is our primary focus.

25:09 And I think that ultimately comes down to just simpler experiences and kind of like refining what you have. So picking that one core insight and driving deep on it and then just trying to deliver as best you can something users love and and love to interact with. Definitely. Yeah, makes sense. Yeah. And I think you guys also a big feature set that you guys enable is like shorting these tokens.

25:30 Do you guys see a lot of people utilizing that? Yes. Not as much as we like. We thought that that was an incredible use case actually. It's very much dependent on the liquidity in that pool. So to give kind of a a quick example, right, each pool is is risk isolated. So we have one collateral token, let's say that's ETH.

25:48 We have one debt token or quote token. Let's call that USDC. You know, you could use it traditionally, which is people lock up that ETH, they draw debt as USDC. You could also flip it around such that the collateral token is USDC and the debt token is ETH. That's an interesting use case there where people can actually just lock up that USDC as collateral, draw debt as ETH and then go do whatever they want with it, go sell it into the external market and then, you know, you can then short tokens if you if you wanted to.

26:20 A lot of really interesting use cases there as well, which, you know, again, there's so many of these tokens out there that go to zero very quickly and here's an interesting way to interact with them that you don't see in a lot of other places. And then same thing, right, that one of the other really really cool features about Ajna is it also allows you to do an interest earning limit order.

26:41 So if you're a lender and you want to lend out USDC at a certain price, I can then just set it and forget it. I can stick it in that a pool, walk away, do whatever I want to do. I'm earning interest because borrowers are borrowing from it. Yeah. And then if I arrive back and the external market price has dropped below that bucket, boom, now it's already swapped out.

27:03 And so now I have the collateral that I wanted to trade into. So it's another really interesting use case as well. So again, a lot of really cool base features that exist inside of the protocol. It's a matter of accessibility to those features, which is I think what we're trying to drive home. Yeah, that's really cool. I never actually thought about that.

27:19 You could totally do a limit order. I mean, I guess the only requirement is that everything you put up gets lent out, but I mean, if that happens, then that's exactly what you have. So yeah, I mean it definitely sounds like you've got so much capability in the contracts and like the back end of the system. It's about exposing it to the users and finding like what is the best product market fit for all the features that you guys can support because yeah, I mean you it's kind of like the Swiss Army knife of DeFi.

27:50 And the best part is is it's like as decentralized and trustless as really you could hope to get at this point. Yeah, I think that's that's incredibly important. And the other thing I I also want to touch on is auction support. Yeah, we do have, you know, we're open sourcing auction keepers as well, so people can show up and and interact with those auctions as they so choose, which is great.

28:10 Yeah, I mean, in order for the space and ecosystem to grow, I mean, it's it's relying it's all about software composability, that's what makes blockchain applications so powerful is you could think of every method on every smart contract like an API that's just open forever, right? So being able to call those methods and and and leg block on top of them is what makes this whole system valuable and you cannot have the ground moving underneath your feet.

28:36 So I really think that permission or immutable protocols are here to stay. Granted the simpler they are the better as that building block. They offer a lot of functionality. You can really build anything on top of them. Yeah. and you don't have to worry about your lunch, you know, your lunch getting eaten or, you know, the application underneath you, you know, deciding to do something that threatens your business.

29:00 And then there's also that amazing and and mo most important security benefit that you just don't get from mutable protocols. Yeah. No, I love the way you put it because really it's like the stronger the base that we have and the more like the less trust assumptions are baked in from like the building block level, the further we can sort of build into the into the sky like like if you want to build a building, you got to go deep and you got to build a solid foundation.

29:28 And there can't be, you know, even the slightest deviation and then you can go and you can build up from there. So I think building that way from the start from the the groundwork is is really important and I mean like you said you know some of the Lindy effect with stuff like Uniswap V2 and and these protocols that are still being built on today even though there's a V3 and there's a V4 it just goes to show like the timeless like life cycle of software which has really not been seen before because if you worked in any kind of like SAS company you know in web two like man we were rewriting like areas of our codebase like every every year or every two years it's like oh well we scaled up 10 times

30:06 So now this part this flow doesn't work anymore now we got to re-engineer to use radius and like we need to put a like a some kind of an async Q in front of this so we need to like redo all this logic but I mean stuff on on blockchain if you go and you disintermediate everything to the extent that you can you remove all of these external dependencies it just works and it just works for for years which is really incredible to see so I think we'll continue to get compounding effects like in the industry as we start to get more of these building blocks and then people can come in and they can with confidence put these things together and they can build and orchestrate their projects and protocols on top and it

30:46 Just gives us more and more leverage. So yeah, I mean that that's really exciting and and obviously huge kudos to you guys for building, you know, what what I think is probably going to end up being like a very useful building block in that sense. But yeah, I mean it's it's exciting to to see more of this groundwork being done and and we're figuring out more cool ways in which we can build composable really like you said like composable derivatives.

31:09 That's what DeFi is. And hopefully we can we can start to bring on more institutions and start composing more TVL and and showing all the cool stuff that DeFi can do. Yeah. I guess speaking more about like immutability, trustlessness, decentralization. What's your experience been as a a DeFi operator maybe like a few years ago versus now thinking more from like a regulatory perspective?

31:32 Like do you think DeFi is is poised to grow more now with the way things are or how's that been? Absolutely. I think anybody who who says counter to that isn't isn't really paying attention or maybe wasn't here two years ago or a year and a half ago. You know, admittedly there was some caution when building this or building any protocol, right?

31:51 And I think and now a lot of those fears have been largely alleviated, which is amazing. You know, we're still kind of waiting to see, right? Things are still gray, but I would say they're positively gray. You know, like a lot of the lawsuits have been dropped against, you know, other large DeFi protocols kind of just sets the tone.

32:09 Yeah. So, I think as a builder, it's an amazing time, right? We get to work in this space on software that anybody in the world can build on top of publicly which is just an insane amount of leverage because so much of humanity no matter their location or language or whatever it be can show up and build and then we all get access to that.

32:30 I think a lot of it for me when you compare DeFi and Tradfi, you know, you get TRDFI is is a bit like these intranets that existed back when the the internet was was first get first getting born and people thought that Wal Gardens was going to be the way to go. Yeah. And I mean, I just don't see at all how that's possible just due to the sheer amount of innovation and and brain power that's getting thrown at these applications.

32:56 Yeah. And the truth of the matter is is we're gonna have so many different variations or use cases per financial applications that just don't exist anywhere else in Tradfi. Yeah. Because it's permissionless and because anybody can build on top of anyone else. And so that compounding effect is just insanely powerful. Yeah. open source it's it's very disinflationary in so far as like we can build a financial application we can build a structured product we can build you know any kind of financial use case and it immediately becomes accessible to everybody in the world who has you know a private key and and can operate a a wallet in some sense or you know soon enough doesn't even have to operate a wallet if the UX can't beed out there but that's the goal yeah so

33:43 I mean like building something that immediately 7 billion people have access to is like the ultimate software distribution platform. So yeah, it's very disinflationary in that sense and and giving folks access to finance that is even higher quality than you know like the highest echelon of our society have access to right now. So yeah, I mean very clearly the potential is there.

34:08 What do you think is the number one thing that's holding DeFi back from getting to that point to that singularity moment? I think time is the answer to that question. The truth of the matter is there's a couple large hurdles. One of which is just general adoption through saturation. You need to eventually have people who become like the idea of thinking about a token just a simple token becomes more widely understood in public culture.

34:36 Yeah. So that's that's just a core element of all of this, right? People used to build internet applications that did tons of things, but so long as email was made fun of, nobody understood what email was, right? Or or or how it could be useful. And I think the same thing applies here to just standard DeFi tech like tokens and vaults and things of that nature.

34:54 Like it just kind of has to saturate into culture a little bit. I I'm not saying that everybody needs to know how to, you know, look at code or read code or even handle a wallet directly. You know, ideally someday all of that's abstracted away, but the idea of, you know, here's a token. I can send it anywhere.

35:11 This is a version of money that I respect and understand even at a high level and that the culture of the world now trusts. humanity trusts generally because the understanding has gotten to a certain level is necessary for there to be, you know, wide amounts of use cases and wide amounts of users and monies flowing into the money capital flowing into these systems.

35:33 And then the other thing is what everybody always says which is listen UX needs to get better. I think that that's true as well. You know you got to make interacting with these apps and wallets as intuitive as using other applications and the the more complexity you add the the the lower adoption you gain. So, it's a it's both of those factors I think need to need to come together and regulatory stepping back and giving builders room to get in there and and and not be afraid when they go and they build these applications is a huge part of that as well.

36:07 I think that that's been holding a lot of people back. It held me back as well. So, that's that's another big puzzle piece as well is letting people just dive in and and make mistakes and learn from those mistakes as they build, which is a big part. Absolutely. I really like the word that you said, respect. I think that's kind of exactly at the pinpoint of like what we need and and what just takes time.

36:28 There's really no way to speedrun gaining the respect of, you know, really everybody in the world in terms of their the financial system. So yeah, I 100% agree. I think really it just it's going to take time for us to prove our track record and also prove the usefulness. And I think you know the larger institutions are starting to say like oh like this is actually useful and they're starting to respect it and I know you know for a fact that a lot of these big institutions have been spinning up divisions for blockchain and they're looking into it.

37:01 So it sounds like that respect is growing and we're we're starting to reach that threshold where we'll start to see more of this stuff coming on chain. And actually I'm interested. Have you guys interfaced with any kind of like traditional financial institution? Maybe as like if anything like a manager for something like these ARC lending pools. Has there been anything on that front that you can share?

37:23 Of course. Not yet. I think that there's still that hurdle of it's actually interesting because it's a bit of a trust thing as well. So, if I'm a user and I'm and I'm deciding to offload some of my risk decisions to a trusted third party, I kind of want that third party, at least in crypto right now as a semi-experienced user, I want that third party to have to kind of be birthed inside of DeFi already.

37:48 So, you know, Gauntlet is a is a great example of that. I wouldn't trust a large financial institution right now to step in and manage my vault and understand all the risks. Honestly, that will change over time. And I think that's me as a as somebody who's natively been in crypto for a very long time saying that.

38:05 I think somebody who's more normie or getting into crypto this year or the last two years may feel differently about their financial institutions. And there's also insurance and other factors that that that play a part on those risk decisions. But no, no, Ajna hasn't had any had any talks yet. but but we'll see. Who knows? Yeah. Yeah.

38:25 When Vanguard managed Ajna funded. Yeah, that's right. Yeah. And I mean, sort of building on that, I know you guys welcome integrators and you welcome people like building on top of Ajna. And I actually I was talking to somebody I forget the name of the protocol, but yeah, I was watching the Super Bowl and and somebody who actually came over to our house was like building something on top of Ajna.

38:46 I forget exactly. Oh, awesome. Yeah, it was just totally like out of the blue. I forget exactly what they were building. I wish I remember the name of it, but and I know you guys have this like grants program. So, what kind of stuff do are you guys looking to incentivize people to build on? Yeah, thanks for asking that question.

39:05 That's kind of our main drive right now actually is trying to figure out how do you leverage the ecosystem and get other players to to step in and want to interact and build on top of the platform. We have ARCs which will largely handle deposits which I'm excited for, but I think the borrower side of things could be picked up.

39:23 So, you know, there's still, we alluded at it kind of earlier, but the managing your borrower position can be a little tricky because you have to kind of understand like your collateralization ratio relative to what a price might be. And I think a lot of that complexity can totally be obiscated away by other thirdparty actor you know thirdparty actors using a price feed not at the core level of the protocol itself but just as a UI in the UI for help to help people reason about the prices that they're interacting with.

39:59 You know, again, we kind of had that point there made earlier where you start the liquidity pool, the collateral just doesn't have a price feed, and then you grow into a market where that collateral does have a price feed. In order to have Ajna be able to compete with third-party or with other lending protocols in the space that do use oracles in their base protocol, providing a price feed in the UI, just the UI layer, would help users reason about what's happening in that base in the in the base protocol and help them better check and assess the risks.

40:33 So any type of borrowing, you know, manager or or or or a system that allows users to better interact with their borrowed positions, I think would be hugely helpful. Yeah. And something that I would love to see in grants. And we already have some auction work going on now that will hopefully be published here shortly and users can can jump in and interact with that.

40:53 But yeah, I mean Ajna is amazing in that it gives we give away such a large portion of our treasury to people who do want to show up and build the governance system. So yeah, if you are a builder and you are interested like please jump in the forum and also our discord and you know any any and all suggestions are great and we'd love to give them a shot and get them funded.

41:15 Sweet. So if anybody's listening and you know wants to build something on Ajna or maybe they already are is the process to just sort of reach out to you guys on like the forum or is there any specific place they can go? Yeah, so they could go to the forum and I'm forgetting exactly the URL for that but a finance has the links there.

41:34 They can go to the forum you can see what other people have put forward as proposals. just generate a proposal and then you drop a address to the wallet you want the Ajna sent to if you win or if your proposal gets passed and then if it's passed you receive the Ajna tokens. Any builder that shows up with something already built is far more likely to get funding versus showing up with just a hope and a dream.

42:00 Yeah. So that's kind of what we ask people to do is obviously be well thought out and anything you've already built would be fantastic. Yeah. You just go through the governance process directly. Absolutely very cool. Hope to see some more cool integrations coming across with Ajna people like building contributing to the UX of like bringing all these features to light.

42:20 That'll be very cool to see. But yeah, I mean I guess one thing I I I just am interested about and I always I like to ask every every DeFi founder that I talk to is like what's one thing that keeps you up at night? What's the number one like thing that is on your mind these days?

42:36 I think for me it's a bit split. The primary thing for us right now is UX as I found it on. Yeah. So that's that's really what we're wrestling with is how do you take a very complex system distill it into its core components? Y both for another version but also for users right now to interact with it and what systems can be built on top of Ajna right now that provide that.

42:58 I think the other thing that I wrestle with lately which is a bit of a side note to Ajna or how Ajna kind of distributes things is governance systems. So I'm personally incredibly passionate about Futarchy. Yeah, that provides some great protections to minority holders, token holders, governance token holders. Our system is quadratic voting, so it does that as well, but in my opinion, not to the same extent.

43:22 So I'd be curious to doing more research there as well and trying to better understand those systems and see if there can is anything that could be incorporated. But yeah, those futur is wildly addictive if you guys haven't read about it already. I think that that's going to be a huge shift in terms of how governance is done generally in crypto applications and then maybe even real world organizations as well like any sort of shared community good but that's obviously very very far down the road.

43:51 Yeah, maybe if you could I'm sure it's complex but give us like a a TLDDR on on futarchy. Yeah. So Futarchy is a system or it's a market. It's an information market that was devised by Robin Hanson. And what you're allowing users to do is they can actually trade proposals. So in existing majority based governance systems right now, you know, the minority token holder just isn't protected.

44:16 So they place a vote. Let's say you guys are also governance token holders that have large positions and you guys collude to steal the treasury. as a small position holder, I can vote against you, but it just materially doesn't matter. And then I'm the tokens I hold are now impacted by that decision of you guys stealing the treasury and I'm out all of the money.

44:37 What this system does is it creates two markets, both a yes market and a no market for the proposal at hand. And there's a market expiration. And so these two markets trade. So one market being yes, the proposal should pass. Another market being no, the proposal should fail. If the yes market is ever, you know, some x percentage above that no market, then the proposal passes.

45:05 So what does that allow us to do? That's really powerful because you guys would be colluding in this case. You guys would be trying to buy up that yes market, trying to make sure that proposal passes. But me as the little guy, I then get to sell my governance tokens into the yes market at a advantageous price and profit.

45:25 So because you guys are pushing that yes price up, the likelihood that it's higher than the external market price outside of this voting system altogether is very likely. So now I've actually made money as somebody who is trying to vote against something that I think is bad for the protocol. So it's it's tricky because we're introducing a new idea for a market, right?

45:47 Just like any other DeFi application, there's that onboarding period of having to understand what the market is just like Polymarket helped us out with prediction markets. You know, there's that hump you have to get over and understanding what's actually happening and what people are trading. But both through you know simplification of UIs as well as people just understanding better through saturation.

46:07 I think that we could see these decision markets come about where people are actually trading the proposals themselves. Yeah. And then it it creates some really interesting scenarios about like risk decisions. It could be like how to which projects to fund and then again people are are actually making money by executing influence on their protocol that they care about or sometimes even just like Polymarket right correcting biases.

46:34 So if someone jumps in and is boosting up a market to inject bias like you guys would be trying to steal a treasury I can come in and correct that market to its actual real probabilities of passing and I've actually made money in the process. It's a different type of market where people aren't just speculating on prices of assets alone or whether number go up, number go down.

46:57 I think that this gets to the heart of how we can profit from making better decisions and that's why I think futarchy and decision markets are pretty interesting. Yeah, the the game theory of that is super interesting. I think it could solve some of the governance problems we were just talking about. Yeah, absolutely. Yeah, that's very interesting.

47:14 And would the vote exit be only available to folks that are like in the snapshot or is that like open to anybody on the market who could like come and arbitrage buy up tokens and then sell them at the inflated price? Anyone in the market who's holding that governance token? Okay, that is interesting. Yeah, I mean it arbitragees to what the outcome of the proposal is likely to be.

47:38 That's very interesting. So then it sort of makes it to the extent that you have liquidity for your token, it makes it harder to get these sort of like collusion proposals through. Exactly. And then the other thing that's really interesting is when you think about it in terms of risk decisions for lending as well is, you know, if if if it was an a let's say or or or compound, what have you, and you're voting on what assets get in or out of the protocol, you want markets that actually incite bias injection.

48:09 So you actually want markets, you call them sheep and wolves is how Robin Hanson describes them. So you actually want sheep. You want people who show up and want an outcome even though it's not the one that's actually best for the protocol. Why do you want that? You want that? Because the sophisticated actors like the hedge funds or eventually someday, you know, sophisticated AI models that are going to be trading these liquid markets, you want them to be able to make money.

48:38 And how they make money is if there is bias in the market. So I think you could get some really interesting scenarios with like for instance deciding what collateral types to whitelist in a protocol or something like that for lending that we could foresee happen here shortly too. Yeah. It's almost like I mean how arbitrage bots arbitrage across different venues today where assets are mispriced.

49:02 If you have a decision that is misdecided an AI can come and like arbitrage that. So that is very interesting and it like you are taking the economical value of decision- making and making it available to the deciders. So that's a pretty cool unlock and and especially with the advent of AI, we're going to see a lot more of this stuff start to become possible which is very cool that we're able to combine all these things in the next couple of years.

49:29 But yeah, I mean in terms of actually building this stuff and and executing on all these cool things. Obviously, we've got limited time and limited resources, what is like the number one bottleneck for you guys on designing some of this stuff and and getting it to market right now? I think our bottleneck is that distillation of complexity, right?

49:51 When you go out to design these DeFi applications, you have this weird job of kind of combining these different inputs into your your reasoning, right? So, you want to talk to users, you need to get that under your belt. You also need to think about the trade-offs, right? This all of these applications, these D5 applications are all structured around trade-offs.

50:12 And as a good DeFi builder, you have to consider the trade-offs and ideally make the decisions that benefit you for the pursuit of the goals you're chasing. In Ajna's case, doing an oracleless protocol, you got to for instance, we made the decision to do price buckets such that a price could be formulated in the protocol. So, I think the hardest problem that we're working on now or kind of chugging towards is just that distillation process of breaking down what trade-offs do users really care about deeply and figuring out how to still keep some of those unique insights and features while distilling away the complexity is a very tricky endeavor.

50:55 Yeah, that's kind of what we're what we're tackling. Yeah. And of course with the mutable development, you're trying to like hit the hit the target dead on from the first try, which is always very hard. So of course all of that like iteration and stuff that you would typically have in like a normal web 2 software world gets kind of pushed back to the ideation phase and testing it against, you know, lots of user feedback and stuff like this ahead of time.

51:20 And I think that's still an art that has yet to be perfected in the DeFi world is how do we do product management and make product decisions before we get to immutable code on the blockchain so that we don't have to like migrate everybody's funds and upgrade to a v2 every year. But yeah, so I think there's there's a lot to be explored there and obviously you guys are probably experiencing some of those pain points in your development process right now and and probably in the past couple of years as well.

51:47 But yeah, I mean really to wrap things up, is there anything else that that you want to add? Anything else you want to discuss? No, no, I think I think it was a great conversation. Yeah, hit on hit on the main notes. Obviously excited about what's coming down the pipeline. I think I think Ajna still has some really interesting use cases that haven't really fully been taken advantage of and people aren't aware of.

52:07 Yeah. So, I'm excited to see kind of some of that stuff ab or again the complexity abstracted and then being able to deliver that to to users and help them out. So, yeah, I appreciate the conversation. It was a great chat. Yeah, dude. Absolutely. And I mean, likewise. I mean, I discovered more use cases that I didn't even know about Ajna just from the past hour.

52:27 So, hopefully this will be useful for for folks listening and and hopefully they can go out and try Ajna, perhaps start building on top of it. It sounds like there's a lot of value to unlock with all the the core features that could be supported there. But yeah, dude, I appreciate you hopping on and and it was a great combo.

52:43 Yep. Thanks for hopping on. It was a great chat.