Spark 🛠️ defi is cool

[Music] hey everyone uh thank you for having me here um let's dive in into the different offerings from spark uh and what you're able to build on top of them um little uh house cleaning uh spark is part of the maker ecosystem uh our main goal is to um basically encourage and grow the D ecosystem and we're doing that through building different infrastructure so as of now we there we go uh we had three main offerings um one being SP conduits this is essentially uh liquidity for your protocol then we have spark land which is a lending marketplace where you're able to borrow die at a predictable rate and I'll go a little bit in depth into that afterwards and finally s die the savings die or the yield bearing Sil coin based on D awesome okay so spark conduits this is something that is currently in development it will be launch next year in Q2 um but it really is BS down to makeer providing liquidity directly to different protocols um think about this as a non-bs uh LP uh inter your protocol uh and also adds very very large scale uh in essence you can think of Condit as adapters uh it's essentially part of this bigger picture which is the maker allocation system um and then yellow that would be the adapter um you would have an adapter per protocol and it'll be a from the point of view it will be a singlet ton contract meaning that the different subd Dows will be using the same uh conduit um to actually provide liquidity to different protocols um another way of thinking about about them is a deposit box at the end of the day the only um comp like strict requirement that you have uh when building on the interface is having a deposit and a withdraw function that is the base man LEL from there on you can build either logic uh with on the the conduit itself or you can have all that logic push into the protocol the liquidity pool that the conduit is looking at uh in a very simple terms a saow and for the sake of the hackathon you can mock this as a simple EA the subow will make a decision of putting 10 million uh into the conduit and this conduit will be looking either connected directly into your liquidity pool let's say you have a lending Market uh which is the case of spark l uh the subd will be providing die directly into the dipole uh earning that yield uh and then at any point the subow will call the withdraw function uh and take it back um if we think about more complex examples you would look at maybe something with real assets related where the moment that the die goes to the protocol it actually gets offchain so there's a bit of like delay in terms of conversions and movement of funds so that point she needs to T kind of can take care of logic uh of how you can do this of ramping how can the saow claim back the money uh or how the protocol can communicate with a subda in terms of like Capital requirements um or any of that sort uh okay I'll go a little more into depth uh in depth into conduits later on in case there's any specific questions mainly on the documentation uh now we can jump on to spark land so spark land is a landing Marketplace very similar to all the ones that are out there the very important thing is that you're able to borrow die at a predictable rate meaning that really the rate is not based on utilization so you essentially you can borrow on a limited amount of time um and it's very importantly you the rate changes based on a vote from governance at maker so you have a minimum of two weeks notice uh for the rate that will change and also uh what rate it will be in the future uh this is kind of uh kind of works towards like the better you user experience when you're borrowing uh stable coin we have two sites we have uh die and now we also have usdc and USD uh they you're able to provide us um on a silad mode um but also you have eth and state e and uh R sorry and Rocket e um so you're able to run emote on on either side and again it's based on the pool based on the pool construct um you can pretty much build anything I think the important thing here is take advantage of the predictable rate on di Market uh but you can build on top of it or you can also build tooling related to The Lending Marketplace and also uh Works a good example of this is liquidation Bots this is something that we're always very interested in encouraging open source uh work on this uh lastly we are on Main net uh there's a version on girly that is couple not fully up to date uh we're also on nosis chain uh we're not on the cheat on the on the testest net for nosis chain uh this is we're trying to encourage everyone to use local Forks rather than running on public test Nets um you'll still be able to do everything you want on girly uh if you need die on girly I can I found some I can send you some um but also if you want to actually run our local test uh test net uh let me know and I'll help you set it up uh we couldn't be using tally it's been working out pretty nicely um yeah a little bit just just to Shield a little bit on the predictable rates this is kind of what you're going to be seeing uh so this for a lot of Integrations that you build on top of lending markets this actually kind of takes care of a lot of your issues when having to deal with like spikes and uh borrow rates all right s die or savings die the final product this is a yield bearing stable coin based on D um is D in the DSR module so let's start with first with that section uh at maker the ding rate module has been there since the Inception of the protocol it simply hasn't never been used because it usually is to encourage more demand of the token um maker has actually been struggling for the most part with uh meet creating enough Supply to meet the current demand therefore it hasn't been used uh Meer has finally figured out the supply s uh of the system therefore now they're able to increase the the rate and actually offer this to users the the yield is coming from the revenues of the protocol but said the profits of the protocol right now 2third of it are coming from treasury bills and a third of it is coming from eth volts uh from the maker CDP system or the maker volts classically this um users have to deposit die into the DSR uh this is a very old school contract so you actually get no receipt uh back when you deposit uh so committee members have uh traditionally created rappers uh very old school one is called called chai and now the new one based on the standard 4626 is s die um big important caveat here is um the standard doesn't uh impose transfer event emissions at Burning on minting uh a lot of tokens have used it either way you duplicate a bit of gas but at the same time it's easier to integrate with s die uh we don't have that meaning that if you're going to be um basically depositing basically if you're going to be your protocol is going to be involved in minting or burning s die you're going to be looking at those two events down there deposit and withdraw that's the only major difference everything else is pretty much standard um couple other things uh it's so it's a token so it doesn't rebase it simply ACR interest in time therefore basically its value uh goes up in time um when it comes to a hack specifically it's very easy to integrate a token in your project uh but we are very uh critical of the being a very special use case for a yield bearing stable coin itself B easy Li limus test on this is um if you swap S for a complete different token does your project still work does it do what is intended to do if it's a yes then it the integration is not meaningful enough just to give you an example uh for Meaningful integration lossless donations so the user deposits 100 die uh the protocol swaps it to S die and then diverts the build to uh NGS and then when the user withdraws it's a still 100 100 die therefore haven't lost anything only the opportunity cost um but on this please come reach out let's talk about the ideas that you're working on uh happy to jam on them and figure out like the best fit uh for this um I think as di that should be complete there's not much uh to it um just a reminder of The Spar the bounties and now let me if there's any is there any questions on this point if not I'm going to talk a little bit about the conduits themselves no takers cool I'll get a little bit into this okay so conduit still a very ethereal kind of concept uh we're kind of working on it as well so um if you have any suggestions on the of how to explain it better very very thanful for that um the way I like to think about it is a direct pipeline into your protocol and it's liquidity coming from maker so rather than you go and um doing liquidity mining or marketing to get users to deposit uh you basically going through an assessment process at maker where uh the protocol builds the conduit or the adapter into it you're connecting the maker allocation system to your protocol uh you go through a series of Assessments again not part of the hackathon uh but once it's live any subow can deposit funds into your protocol uh under this maker allocation system where maker governance essentially has a admin control or the whole system um the main thing here is the so there's two things here in documentation we have a overview of what conduits are and then this is the sparkl conduit specific implementation this is not live yet sparland does get a essentially a credit line from maker today but it's with the old school system that we call the d3m d3m simply says it looks at an external contract says uh based on this uh rate on spark land will provide x amount of liquidity and it basically runs on that the in the future conduits are more of a um journalized approach so you can build this same logic into it but you can also build a bunch of different Logics as well as just have no logic whatsoever and it's simply a very manual process of depositing and withdrawing based on some governance decision for example uh so this will be a real overkill for the hackathon mainly because this basically takes into account the entire process of the the allocation system uh for example the buffer registry the roles it's all part of this all this you can just simplify it into an EA and you simply say hey sub Dow B send money here and then you can take care of the rest uh of the product um quite important um is if your protocol won't be able to provide in the next block liquidity uh let's say you're lending Market um you the subow provides liquidity into you and you're at 100% utilization uh the question then comes in okay how can uh the subow maker be able to pull the liquidity uh so cabat into there are kind of what kind of differentiat can differentiate your project from others uh when it comes to conduits um just to give you an example this is currently being scrapped and reworked on but uh we the team here for sparland had a essentially um a look at withdraw but this is not necessarily the best thing but essentially uh a request of funds function that it would call this function to the conduit and basically if there wasn't enough liquidity in the market it would send a message to the to sparland to increase interest rates to be able to recoup enough liquidity to then send it back um Turned out there's a lot of like edge cases that weren't well created for um treated for that um but this is kind of the idea behind this you don't need to add too much loic into the cont itself but you do have to have the scenarios where what's happening in the worst case scenario what will happen uh how what would um gone essentially okay um I think that's all I wanted to say um this is a nice example it's very simple and again it only takes it only implements the deposit and withdraw functions and nothing else uh and then just has an inv very testing that we was very happy to see uh when people provide uh do hackathons um now if there's still no questions I'm going to go and show you uh some project ideas and past hackathon examples let's go here all right ha those ideas and these are ideas are basically not based on building directly on top of uh the three systems just something around them like more like tooling uh so someone that doesn't have um protocol engineering um background uh it's more than design sign you can do visual momentos so this is how you can easily what's the best way to communicate uh people for to share their either their positions or information about um their s die or their sparkline positions so I'm a personal fan of ah come on Larry what's going on um there was a little nice thing that uh Larry had from a I forget what protocol cool I can't do it this um basically showing hey this is my position I am plus 50 Cent um um like in profit something like that so for something that doesn't want to code that the other one is liquidation bot uh we currently have we bu my team has built a open source liquidation bot there's also other versions of it like the Paradigm um they created like a framework called thmas as well uh so if you're interested in this also we're very interested in seeing uh implementations of liquidation Bots and how you can deal with them uh we currently have a quite uh basic liquidation system essentially if your position goes under one half of your position gets liquidated um so anything that can add Logic on top of that and make them a little bit better I know um defi saver with their Auto repayment system or CRV USD with their which essentially is the same thing with their soft liquidations that will also be a very interesting uh idea for a hackathon um user analytics I'm a big big fan of analytics so if you want to work on that only um very happy to support that uh we can look at things such as from the risk perspective but also on the growth and J perspective um how the system is performing uh comparison against competitors um a big thing for example that takes a lot of time an effort is um parameterization so deciding what's going to be the uh the Kink or the interest rate um I'm going to quickly show you in every Market apart from the so this is essentially what it looks like having a stable uh predictable rate right completely flat doesn't care what utilization uh it has but for the most part if you go for example to ethereum you do have a you basically have two curves you have the first curve you have the Kink and then you have the second curve it take a lot of time to actually uh build this and modify this and of course we're in direct competition with a and compounds and other lending marketplaces so in response to someone else changing their rates we also adjust hours so and a project built on top of that uh is also extremely interesting um yeah I put you a bunch of examples uh on us behavior um something I personally want to see build uh is a easy way to call uh web three data from uh Google Sheets that's also something that I would actually use every day um then moving on more on the let's say automation or G abstraction um let's say the simplest implementation for this would be kind of like the one click recursive leverage uh some similar to what summery or instad app gives you uh but then you can also do an unwinding uh deleveraging on the same side um creating a stop loss uh system as well um a sorry a previous hackathon project was um build a V4 univ V4 hook uh that act as a stop loss to spark line positions so so stuff like that also um very interesting I'm going to stop saying very interesting uh finally the alert and alerting system I think right now uh there's no amazing tool to just simply for user themselves to create a very easy alert based on what the borrow rate is in the in the in the market uh or a monitoring of the position outside of like a crown job um there's been companies attempting to do this uh the most sucessful one uh called Halal XYZ actually got acquired by infura so they no longer did provide the service so open source competitor would be amazing um now quite important and actually more interesting with the idea of like bundling uh different sponsor bounties together um we're particularly interested with when it comes to conduits if you're getting a conduit that links uh to another sponsor liquidity pools and so on uh that makes a lot of sense the very important thing about conduit is its liquidity at scale uh so you can have a great use case for example uh Community uh borrowing and lending uh but it does not require 100 million in liquidity into your pool uh so this is kind of the the isue think if you're doing that scale it should be conduits if you're doing that small thing uh it should be more like sparland or S uh and yeah not beautiful but here are the past hackathon projects you can check them out they bit a mix between uh conduits SD and Spark L I'm going to leave it out there I'm not going to open all the links uh and now we have about eight minutes of Q&A if you guys have any questions uh if not I'll just like Ramble On by myself yeah uh then I yeah said this super updated like meaning you're going to be by now you're going to be missing um I think nothing's substantial and yeah the only substantial thing is in Girly you won't have uh usdc and USD enabled which we do have on Main net that's I think that's the only main thing then you have all parameters but you don't mind you don't care too much about that yeah function yeah yeah everything there like it hasn't it's not breaked yet yeah your uh if you want to bridge stuff more than happy to see that but unfortunately we don't we don't have any available anywhere else apart from gly I think there's a previous hackathon winner that uh bridged the entire thing into I think polygon cgbm or polygon uh which is like I was like great job a lot left for just to show the use case on polygon uh but yeah nothing else all right perfect well thank you very much I'm going to be a booth and just hanging here until I probably midnight and also all day tomorrow so uh yeah happy second