DAPPCON 2024: Intents And The Future Of DeFi - Tigran Bolshoi
DappCon 2024 · May 2024 · 20:50
Tigran Bolshoi examines intents, a design pattern where users state a desired trade or outcome and let solvers compete to execute it, rather than specifying the exact transaction path themselves. The talk considers what widespread intents adoption could mean for DeFi's user experience and for how liquidity gets routed across protocols.
Speakers
Topics
From DappCon 2024, hosted by Dappcon Berlin.
Transcript
hi everyone my name is tan some of you know me as a kinetics man today I would like to talk about the internet new area of Internet defi intents where we are now and what we are building in Defi and how the defi affected uh on a like whole system of anti system of internet and what we are in kinetics do here so I will start from the beginning because organizers they give me 20 minutes I need to spend it right let's take a look um on the early stage over internet how it was difficult to create a website I even don't say
about to like build or develop some software it was possible only for big companies and and uh let's let's take a look of the evolution of Internet so everything starts from aranet and uh TCP IP aset we start to building manufacturing for computering for mobile sorry for that for mobile one of the biggest push which was happened uh in our Cent is email and DNS it's uh when we start to like translate
move data instantly and uh worldwide web so I know a lot of people says that crypto is a speculation but if you take a look on the history the boom of.com it was also speculation that's how we can attract attention of u k uh Partners like U developers venture capitalist users for Mass adoption after this uh during the 40 years the internet was extremely um developed so we got like a lot of things which was not comfortable and it's not possible to understand right now our life is out this is like a search engine mechanism mobile internet
data web to infrastructure clouds everything like this and if you take a look uh where we are right now around us is a I robotics aot Quantum computering but it's still not um enough as we can see uh during this Evolution uh most of the companies like OG companies they they get control of the data and the Technologies but if we talk about the Technologies like AI um social system banking system system robotics it's could be really dangerous that a few companies just control it in
their hands so that's why we need to have a decentralized technology and distributed networks for example how how the internet looks right now for us for example if I wanted to do any searches in the internet I need to open a hungus of website I need to analyze all the info by myself to get where is a fake where is a sponsor it which one is true and I always need to play with a wording so the same is like we with a AI I need to put exactly the name like of my my request and get get the result I'm not sure that this is a
perfect system or ideal system and uh if we will if we would like to talk about how the system probably will look in future and what's the like ideal model of this so we in kinetics believe that it will be modular and intent based system where is the top level it's applications applications can be anyone so we can go out of the web free is also we can used in a uh is can be synchronized with the traditional economics so we can use like a taxi service real estate uh some e-commerce infrastructure whatever the next level is it's a delegation and we call it intents so it's uh the meaning of this
is when I can ask somebody like a professional Market maker or professional player to help solve me my task with the guarantee uh after this going level of AI agents because AI agents they cannot work right now independently so we need to monitor focus them and we need to learn them the the layer of infrastructure where is the AIS can be run is a blockchains so it's some kind of layer where we can execute everything and under the hood it's um like government's level with a consensus of verification let's take it deeply so
what is this why we need the intense in this for example I have a request uh and uh I don't want to really go deeply and this how all the complex things working for example for me if I want to receive like a passive income I want to send this I want to send my intent to the aone where is like professional players they can start a competition in a aone and provide me with a quote like quote of the bit as soon as soon as we match the order uh before the user sign it we need to be be sure that my intent is secure C and solers they provide the
collateral the next level the solers they can use the help of AI agent so they can help them to like analyze a lot of data use a search mechanism to develop the code execute the code and uh make it more efficient after this we can go to the level of execution uh in my opinion right now the blockchain is some kind of part of speculations because uh we're not enough get from this technology and as soon as we get enough it it probably will be on a state of uh in a level of state so for example like when the governments they can start to creating
their own digital currency like Euro dollars in or whatever and inside these blockchains the company they will do their own Solutions which can be built on all the like type of uh business what we have right now is like similar like worldwide web but with more functionality and the level of ZK is uh we need it because all the data in the blockchain it's uh public and transparent so we need to provide the Privacy for this and make it more confidential but the ZKA stuff is not only for uh for the confidentials is for example like if you take a look what's
happened right now with the sent Labs uh platform they create amazing solution like sp1 when I can for example use my basic calculator I do a lot of calculation stuff uh during the whole day next day my partner can ask me like what what is the result I say I get result XY Z and he said okay I want to check it so but I can provide for him algorithm how we can prove how I can prove all the my calculations and we don't need to do it again so that's a that's the future and if you take a look right now how the Dy uh looks for the like General users it's a lot of information that
they need to keep in mind it's like oracles gas bridging swaps RPC NS whatever so basically you need to have a good knowledge of kungu in a defy to not like lost your money just in one click and what we are do here in the kinetics so we are creating modular framework for chain obstruction applications where users can delegate their task so if we take on a common defi problems uh we can manage them in a different groups for example users they need to use the experimental like experimental UI and take all the responsibility so they need
to have a good knowledge of kungu in a defi to set all the settings like a slipage tolerance price impact uh the developers they they need to create an infrastructure for this so they need to take the responsible for all the mechanism like oracles price rate uh notes uh they need to use a bridge system with a entrust to like R tokens and the same problem with the networks networks they always need to attract like experienced users and uh like put all the responsibility to the users about all the sections we uh I think this is not a really good solution
so for example like let's like let's imagine how I will move my assets for example like in some raking protocol if I have like salana or Bitcoin let's say so I need to switch wallets find the bridge I need to bridge my asset to another Network I also need to keep a guas pay for the gas set all the settings after going to do a stake I receive my virtual token sometimes I need to wrap it I put it in a rest staking and finally I get my lert token and for example if I use like a puffer system and I want to double my points I need to also create a pair in a like pool of curvy and I exact I need to know exactly how much to put like What
proportion or my tokens can be just moved in in in one token to another and the solution what we would like to offer it's a like a new generation of application which we can build it's a chain abstraction so chain abstraction we call it because we don't need to uh care like the users they don't need to care anymore about how to uh like breach and uh how how to communicate with all the different networks one click because we can create any combination of intense gas L is because the resolvers they also can take
care about the gas so basically I want to express my intent for example I say I want to pay in a and receive a b and resolvers can they you do like under the hood everything just in one combined transaction across the network and the current system is also broken for the Developers for example when the developers uh would like uh to like launch some store in a web to I can use a lot of solution out of the box for example I can use a web web uh WordPress engine to develop my uh website just in a few weeks after I will integrate the payment system delivery system and the
all I can start to receive advertising and sell my goods but in a defi it's not look like this in defi we need to care about all the different things it's like how to connect different wallet how to create a contracts with our how to do a bridging at multi multichain infrastructure so this is the biggest problem for for the developers if I open the pool in one network and I want to attract liquidity I I locked inside this network so for example it's like a new L2 Network on bitcoin and they have no not not not much users right now it will be really big problem for me and I need to explain to all the users how to do the bridging system like I previously I said it
and what we are do we creating uh interface interface for the developers and for the users where you can uh delegate and uh create a intent based model which can be integrated in your any application uh easily just using API so you can concentrate more on your business side and you can concentrate of the logic uh like what you are developing so you can attract for example like it's a new al2 work on a Bitcoin and you can you need to attract the new users so you can easily integrate it in your application just in one click the users they can put LT token LST token or whatever and they don't need to like care because
everything what will do resolver they will do with the fix it rate and without like all these Kung Fu things like a slipage so the layer of this is on top is the user when the user communicate with the applications like a wallet or any any protocols and a settlement layer it's resolvers when the resolvers um like under the hood on the back end create goes for this and algorith how to do it execution level is a networks where we can launch it and validation level on on the back end where we can prove all of this information so how how the Z ZK like clients work in our solution so like in
a in a simple work we take all the historical data from the Bitcoin and we deploy it we shink it to small like client and deploy it on agnosis network so right now using just a smart open Smart contract open source we can easily read all the state from the Bitcoin and uh for us it's super cheap update in the batch mode for example like we can take 250 60 blocks and upload there like 10,000 of transactions just in the one chain in the cost of of one onchain transaction is like less than one cent all these transaction which resolvers execute they secured by the collateral so this mean for example if user want to like swap let's say in simple mode 100
usdc so the collateral 140 usdc so it's like a 4% 40 percentage more and this is the in ins situation program for Liquidator so if the resolver fail intent they can start a new competition and because this is a onchain transactions can be done just one time so they will start to do it as soon as possible and that's how we can demotivate resolvers to make a fail of transaction and uh create an in Wise program for for Liquidators right now in the kinetics we planning to scale in L2 networks we are focusing on L2 networks for Bitcoin and ethereum probably in the next two months we we will add most of the popular
networks and also we open a Gateway for the projects so if you have your project and you're looking for like a tvl or you need to uh attract new users from the different chains so you can easily use our API and uh uh create us our infrastructure like any kind of combination of intense which resolver will will will Sol solve like a few days ago we already launch in the main net right now it's limit um with a 500 so will be 1,000 and more and uh for example we we already done like integration with osis pay so you can easily move your Bitcoin native
or any other asset from any chain directly to EUR e just in one click and you don't need to store a guas and you can delegate this to the resolver who will like launch it uh and do all the actions on the back end soon we will open uh public resolving so all the resolvers they can join um and we don't require like any uh stake for this so it will be open in public as the result uh we can see here execution time is few blocks because there is no intermediates so we don't need to pay uh for validators in the process of execution so if the like user uh ask
resolver to make an intent and resolver complet it in time we don't need to pay any intermediates and we don't need to wait for them the same situation with the gas cost so usually when you do combination of dexes and bridges uh your guess will cost around like 6 700s so but we get as a result like 65k gas units just for one on chain transaction so two on chain transaction is around like 130k and the same situation with the rates because we don't have a intermediates like a pools and we don't have a formula which uh is a inside the ecosystem so resolvers they can provide higher uh amount than one uh because they can do arbitr strategy and they can
also do a rebalancing strategy uh we're planning to use a token for decentralized system so it's like similar to P um when you spend your electricity uh to update the state of the ZK light clients and also we we will use it for decentralized matching so that the all the like match of the of the orders from the resolvers will be State on the notes of our stakeholders um we already live if you will need to move your ass instantly with a without any fee and with a cheap gas so or you will need to use a API uh
contact me I will happy to answer on all your [Applause] questions what how we going to solve the liquidity fragmentation problem for example I want to get for around 10 Millions on more Network and your solver could have like one solver have a five 100 K and another 1 million and that's all because how it could be done in more way to really fulfill the order not to wait until the
bridges could work and so on to deliver liquidity because currently what we found in gearbox that the lack of liquidity on the many L2 networks create a huge barrier and even Liquidators for example who want to liquidate such big positions are in trouble and it could be solved I believe with intent um if if you talk about the biggest amount like uh millions and 100 Millions uh we have solution but it's not in the current version so the the current version is more like adoption of intents and our focus is uh to create a Gateway possibility to use a project with a
framework uh and not build all the multichain infrastructure but uh we will we're thinking about this problem for a long time and we have also solution uh this solution will be in the next version so it it will be like a uh chain obstruction uh multi-chain pool with a like stable coins so when you can um like it will be not a trading pair so it will be like a liquidity which you can uh use for the Bing mechanism and you can solve a high amount so you don't need to split like order every time when you have like 1 Millions between all the reservers but you can use this liquidity
in the B and Ling protocol thank you [Applause]