Building Programmable Lending Markets in DeFi | Edd Shadowy - Silo Finance
Ethereum DenverΒ·Thu, Oct 9, 2025, 12:00 AM
A technical workshop on building custom lending markets using Silo v2 that can plug into other applications in DeFi. π Get Ready for ETHDenver 2025! π We're already hard at work preparing for next year's biggest Web3 event! Keep your eyes peeled for more info on ETHDenver 2025βitβs going to be epic! π
Transcript
[Music] he everyone my name is z and I'm from Silo Finance I want to talk to you about um programmable markets and how they help us with product Market F uh F Well product pmf with pmf um and how they can help you to find it as well so let's start from the beginning um Silo was created as a necessity to um make the landing Market more secure in uh you know during the all times defi summer 2020 2021 um we had a lot of demand for risk assets and long um you know those those risky tokens that were just relas and you know that demand was met with forks of compound or a um and obviously there's they're not designed for uh risk assets and what actually happened a lot of hacks and the simple asset one single asset introduced to Shared pool design like compounds just spoils the whole bunch and uh everyone loses money so that was the reason why Silo was created and um we actually launched the the first version in 2022 and we got to a pretty good uh tvl uh we got some nice adoption but really quickly we kind of reaching the ceiling um glass ceiling and we couldn't go through it and as you can see our chart of tvl is full of pain and the reason for that is silo V1 was designed for one use case in mind and obviously you know you have your assumptions and then you launch the products and your users will tell you that you know it's good but then there's a bunch of other things that you should do differently there's there are other features um that you should uh you know introduced the project but we couldn't do it because the whole thing was essentially mutable so there's nothing really we could and any Improvement to the um to the protocol you know those small things that a lot of times turn around to be big hits were very expensive uh to introduce because just a single line of code to a code base that is like two three four thousand lines of code can introduce a critical bu which you know many times it happened already so that begs the question how can you have immutable contract immutable product in Defi and still improve and introduce new features to find product Market fit and the answer is Hooks and that's exactly what we did um we designed the hooks for new version of the protocol in a way that um they should be this the simplest possible way to introduce small incremental improvements um to the protocol and this was our Nerf Star right we wanted those small things that you can do quickly um to be isolated in a single contract so you can focus only on your stuff on your features and then the rest of the protocol can handle you know the kind of infrastructure things so in design they're very similar to uniswap G4 and um you know everyone knows about those uh by now so when we actually developed that um the The Silo hooks feature and we start started doing benchmarks right we wanted to see how much simpler it really is and the first example from the top in lending uh in two asset Landing PS a lot of times you have assets that are good for collateral but are not good for Deb they're not secur this way so you have to forbid um borrowing one of the assets so essentially you create one way lending um pool and if you wanted to do this in you know the C base the core code base we had to remove like maybe 120 lines of code which introduced like a lot of changes a lot of security assumptions change and now even though the code does make like become simpler it's you have to go through the whole process of like testing uh peer reviews formal verification fuzzing audits multiple Audits and it just becomes a huge huge process with hooks you can do exactly the same you can CH exactly the same features with 18 lines of code which you can write during the evening and then on Tuesday send to your bodies for reviews and by the end of the week you have lunched this on production so this is a absolutely huge Improvement in terms of like those small incremental improvements for your uh for your product and this allows you to actually find your product Market fit much better right and this is not something you know that we just uh invented uh web to has been doing this for many many years now is just in defi the problem is when you have actually mutable uh contracts something that the code that cannot change so on top of the silo Hooks and our you know Vision to um to do incremental improvements we also introduce the deployer fees so essentially what's what now the developer can do um they can write the hook with the custom feature um or maybe multiple Hooks and connect to a single Silo Market and deploy The Silo for themselves um spin up their own UI and essentially launch a brand new uh Landing protocol with their own their own features and differentiators and it will take you much less time than it took us to build this whole thing from scratch so we launched V2 um about a month ago um we were able to get to pretty decent tvl by now and uh we have a lot of markets already with like deep liquidity uh High aprs and um I'm bragging about this for a reason so imagine this case right there is like new hot real world asset um that is launching and those assets a lot of times don't really have the liquidity right they don't have really big secondary markets um because they most like most of the time they come well not most of them they always come from traditional Finance so what you could do is you could come up with this idea I'm going to build a lending protocol just for AWS right I'm going to focus on this so the first problem you're going to hit is liquidations how do you handle liquidations with the asset that is not liquid you need Liquidators that actually have underlying balance right you need someone that is willing to hold the real world asset in their own wallet and then repay the collateral or rather debt um in the silo so for example if you have rwl paed with usdc then someone must be comfortable holding this rwa because well that's how liquidations work so it's very probable that the depositor of usdc already accepts the risk I mean he has to right because you deposit money so you're sharing the risk with this asset and that means why don't you just give them the underlying asset instead of liquidations right so you can default rather than liquidate so when the position actually becomes insolvent the usdc depositor gets the RW assets with a discount because there's always a liquidation fee right so that's one thing you can do that's the Hoke you um you develop and then immediately you can just spin up as many silos for as many rwi assets as you want and immediately create an UI and you have your own lending protocol with big nice differentiating factor and then another problem that you're gonna hit very quickly is like you know you can rch 10 markets but um if they're all empty then it's not very attractive so 0% APR doesn't really bring that many lenders so what you can do you can throw another hook on top of it which is going to deposit idle liquidity from your Silo from your lending protocol right to our car Landing right so as you guys see um hopefully because maybe it's too small but there's like SDC Market with 11% APR so what you can do you can pair your AWA with um with usdc and this underlying usdc immediately goes to this Silo and earns 11% which means if you start your own product you don't start with bunch of zeros you start with all those inheritance aprs you can tap on this liquidity immediately and you have plenty of those right you have plenty of asset with very deep liquidity so you can jump start your own protocol so if you merge all of this together the the hooks the liquidation that the permissionless deploy um deployment um the ID liquidity management you can literally start your own lending protocol within weeks and bootstrap all the liquidity using the underlying liquidity in Silo for your own lending so this is the stuff that the hooks can do for you and on top of that remember when you deploy The Silo Market you can choose on your own for yourself how much the fee for given Market you want you're going to take you have the rench to choose from so it's not given to you and also this will basically bring you Revenue right the same as as it brings Revenue to um to other deployers to the Dow Etc so you have a way to launch your own lending protocol within couple of weeks and differentiate very very well actually in the market and you have those examples already in un swab before hooks right there's so many of them there's so many projects that literally with like 150 lines of code they're like a project on their own they have um they have user base they have liquidity and they work uh very very well so this is the biggest takeaway um from from the hooks and um from this from North a Northstar idea that that we got with the very small amount of effort now you not only can just literally Fork Silo without forking Silo right you can Fork Silo without having to deal with all the deployments all the challenges and like all the security assumptions that you're supposed to know this is all of this is gone but you get all the product stuff product site on your side and you can simply just launch your own project um project very very quickly so after that the search for project um product Market fit is also open for you because you can just stack more hooks to your own markets right you can just keep experimenting you can keep innovating and you don't have to worry about all this underly l in protocol infrastructure that you need to do in order to get to this point so we created the quick start repository for you um there's like a playground where you can just go you have uh all the heavy lifting done for you um you have The Silo deployed as a fork for you you have um um the dummy markets also configured for you so all you have to do is just focus on the differenti differentiating factor that you actually want to implement and um yeah and then just don't forget to set deployer fee so you can bre uh bring sweet Revenue when you actually um launch your project and um thank you guys for for listening I hope uh I'm looking forward to see what you're going to build thank you
Automatic transcript β names and jargon may be misspelled.