Stani Kulechov I The Road to Product Market Fit I Avara I Pragma Singapore 2024

[Music] hey cik socks and sandals interesting that's a uh sandal sock Market fit here you know you've made it when uh you can be on stage with socks and sandal talking about product Market fit yeah it's it's actually our latest fashion it's it's really really nice comfortable and uh really fits well to to this we well and these are avara socks from what it looks like yeah this is L lens socks yeah it's really cool um very comfortable I can I can send you a pair do you have any do you have any socks here can we get some do we have there some can I get some for for the audience too I just want a few yeah can I can I get who wants some socks who want some socks yeah all right let's do this so I'm going to give some I'll I'll hold the mic there there we go all right all right who who wants one all right oh one more we who wants the green one now there we go [Music] more all right this is go sock two more more more here we go more here all right there come and get some we uh we'll make sure everybody gets one and you run out of it successfully all right that's a good way to start we should make this the standard yeah definitely let's get one of those t-shirt cannons and uh that's the countdown so those who didn't get any sucks don't be disappointed we have few pairs so you know come up the talk and and maybe uh you will be lucky all right well the discussion we want to have today is about understanding what product Market fit means in crypto and you've been working on a for a long time and now you've kind of seen a few evolution of the space but also your own products and I wanted to kind of start off by asking you how kind of do you think about product Market in Defi and just sort of walk us through some of the early decisions you had to make and kind of was this even like the consideration like did you said hey we can only do this thing because you want to reach product Market fit or was it mostly reacting to what people were doing and then just going deeper in those uh areas yeah I think um the reactiveness is is quite um uh challenging especially because I think um um crypto and web particularly has a a very like a uh thick social layer which means that you know we we like to see what other people are doing and kind of like like also um maybe like take the best of of what what else are doing and integrate into our own product and um we we like to like talk about narratives and um build upon those narratives as well and and sometimes that actually uh might be interesting uh but in some cases it can be really uh big distraction uh as well so one of the kind of challenges I see in the space is that when you have you start building something you have a great vision and you want to unlock that vision and build something very useful um and interesting uh for the users there's a lot of things happening and and maybe a ecosystem of other um participants in the market that are building something similar and creating narratives and very easily kind of like uh that takes the the the driver seed and it's kind of hard because eventually like what's the most important for um reaching product Market fit is to build something that the users want um and and some of these users might be part of the social layer um and and talking about your product or or the space in general the category uh you're building um but necessarily aren the users of your product and that's a really important thing to do is to to recognize in pinpoint who are the actual users um what they think about your product how they're using it um what's missing there and how to make it better and then over time to understand how big of a group of people this is and whether it could become actually a product or or a business so that's a really kind of like a hard thing to navigate it's very easy to be reactive and you know you have a competitor that integrates a feature you everybody has points now everyone has points you you're following the meta The Meta is even like a kind of like the bigger challenge because everyone is trying to search for the same uh thing and I do think there's good thing about meta and it's it creates like a familiarity so if people are using web3 products and they can familiarize with an existing um kind of like a idea and and you can write that idea maybe that's uh interesting but it in most cases it doesn't fit to every uh use case uh essentially so that's one aspect the second aspect is that many of us here are building because they believe in decentralization and they believe in building resilient infrastructure that can be better let's say better Finance more efficient uh better uh ways to own content uh building social networks or predi prediction markets or whatever that type of a use case might be there and in in some cases like um the decentral decentralization aspect might be something that makes it more difficult to build something that the users want um and you have to kind of like come from still an angle of like you know how you build the best thing for the users so in some ways um in there's a lot of distractions that can actually derail you building what the users want or taking your time away from talking to the users in the in the first place so I think those are the things that are quite uh challenging in the space and and one last thing is that actually knowing who your users are because in a example like when we started um it was really hard to find um the the people who are using a and and eat land in early days because um we didn't really uh we didn't collect data just an address there's just an address and you know you can just speculate who that you know address might be obviously today it's a little bit different because you actually have labeling and and whatnot so so you can actually recognize for example the liquidity docs um and that helps to connect with those people and ask like what do you like about a like what's particularly interesting about that um lens is a little bit easier because people can create content they identify themselves so uh it's it makes much more easier uh the job but that's the kind of like a challenges that we've we've seen in the past that has been uh really um uh challenging one thing that uh many have noticed is a transition from a to avara and you've expanded the the set of products you've been working on obviously people got the the go Soxs and uh you have go you have lens OB of course and also soon family wallet um walk us through kind of the decision to sort of expand that and and is kind of the goal here to vertically integrate all these products or or something else um give us sort of a overview of like what the avara vision is the original um Vision obviously was just to build something interesting cool um in and I was still studying in University um I was J 16 um and I'm lead Bloomer so I I dropped out of um um High School um to actually program and that was during the web 2 oh boom when Ruby and rails was interesting thing and nobody cares now about it um and then went back to actual University and and end of my studies I started to read a lot about ethereum and and things you can build there and previously I built uh social applications and financial applications and for me it felt really um interesting to experiment something on on ethereum that is financial and defi didn't really exist at that point so I wanted to actually um create a a way to actually earn interest um collateralize uh assets and and borrow against them on on the blockchain because it's transparent it's auditable um and you can guarantee a execution with with smart contracts um so it started as a pretty much like a um side project that then U became interesting for a lot of users and I was talking to people back then and they were asking like hey can you improve this feature can you um can you do these other things and and started to actually improve the the whole product and at some point in 2017 I realized that um this could actually be a a a good product and and started to focus on full-time on that and that vision for me was just using this technology um to democratize finance so starting with obviously uh lending and borrowing and and then expanding and and just making available for people uh equal access to to financial uh opportunities we're still living in a world where most of the people don't have access to Dollar um speaking even of a dollar yield so if you can have actually borderless um composable Finance uh we could actually have more interesting opportunities and the financial infrastru will be better and it could work the same way as as the internet is working today um just purely protocols that are used by by everyone who is accessing the the the internet and much more later what we realize is that here we actually are building a completely new and Next Generation internet uh which means that it's not only about Finance uh it has Financial rails and financial value and that can be actually moved around but it's going to be a more it's about um ownership economy it's about um a economy where users are participating in the infrastructure that they build and contribute to um and that's how we formed avara and avara is about building uh the people powered internet so we believe that in the future um lot of this infrastructure will be decentralized that even was centralized in the past like social networks or like finance and and Banking and being that decentralized we still think that there's going to be a lot of businesses built but in a more fair um and open way so we're going to see a lot of this infrastructure being accessible to everyone creating more opportunities especially more opportunities to innovate for um young Entre rurs like like everyone else here um but at the same time um we will see like actually businesses build being built and there's just more power and more sa for the users what they can actually do so with avara we started to expand to building lens as a soci resilient social network um where we are trying to bring ownership not just into Financial assets but also ownership into your social identity the N networks you create um as a social graph and also uh creating content and being able to own that um as well and with family wallet what we believe is that we can expand the idea of what if we actually have uh a more of a end to-end approach and actually Focus how we bring more users into the space by building a a wallet that has a really nice design and experience and and make it it super easy for everyone to contribute and every infrastructure that we built is actually open so even if we have our own interfaces anyone can actually plug them in and integrate and build and innovate uh on top of them and I think that's the big value proposition of of this new um Next Generation internet is that everything is open you can build algorithms you can build aggregators um and it just makes a much much better um way to collaborate whether it's Finance or or something else and that's basically avara we have still a long way to go um because building um this decentralized internet will will take probably like the next couple decades it's a long road but we're we're making progress um well speaking of uh kind of future things that we get to see um can you tell us more about a V4 and uh what's part of this uh new version a V4 is is it's going to be more optimized version um of the a protocol so we've been building a for I would say like past seven years I'm focusing on that um that opportunity and um I I think the AV V3 is is really very optimized uh well functioning uh piece of public good now and I would say that at the moment a protocol probably has um roughly like 70% of the market share 6570 and if you look at the code base including all the forks that's like 75% of the market share so it's a really good type of infrastructure as a public good and V4 is just taking and optimizing um the a protocol and its use cases to cover also um things like real world assets um flexibility adding ability to add liquidation strategies different borrowing strategies as well um and also this concept of liquidity layers so now a a one entry point Market uh which actually makes it very efficient and and scalable and that's why that has been like one of the drivers why AA has been uh growing uh especially more lately so we have 17 billion worth of net deposits at the moment where almost half came uh during this year um but we want to actually enable ability to create your own liquidity layers that can be seated from the main Market as well so we see that uh the kind of future is that there's different kinds of risk profiles um and and different kind of use cases like rwas essentially so to facilitate those use cases uh and to avoid liquidity segregation uh we want to create a version where actually you can see the liquidity from one particular uh pool and there's a concept of um liquidity premiums which means that we can repic the risk uh and reward within the same pool without fragmenting the liquidity this means that if you're bringing into a the most lanest Cal and borrowing the most lanest um asset like let's say eat against usdc or go um you're paying the risk-free rate and if you are um providing more risk into the protocol you're paying more premium which means that the protocol earns more and the the um the the suppliers are going to earn more um there's a lot of features cartic there like there's way more I want to dig into but not enough time but maybe maybe uh I'll ask a similar version of it for for lens so you recently announced that you're switching to the ZK stack um tell us about why that decision and sort of what does that unlock or kind of what does that actually now mean you have to do more of to make it be fully ready for prime time the main thing I was what what I was searching for when I was looking all these different stocks that are available and and there's a lot of good technology out there there's a lot of choices and and probably this is one of the challenges for everyone there that are building something so so for me what what was really important is to choose a stack that is technically represents the capability that I'm trying to re receive um I didn't care too much about what is the kind of like a latest boss and and where people are launching meme coins um that is IR one because that's part of that kind of like a chasing meta um um narrative for me was to find something that is very suitable for me uh and know for our vision what we wanted to achieve so one big factor for for the lens team is that we actually want to ensure that users don't need to pay for uh uh blockchain fees and transaction costs so I think like previously even like if you think about like um text messages people used to pay per message to receive and send which to receive and send messages which is quite uh crazy if you think about it today because now you know you you don't really you pay packages and you don't really pay per per message for us like we want to provide a uh infrastructure for lens where no one actually from the users are paying any transaction cost so so the blockchain needs to be free to use um and and then you can have some sort of like a freem model where uh everyone can use it for freely but then there is value added things that can be added that actually drives um some of that uh revenue and the way to actually do that is to use a validium so validium uh means that you're using ethereum as a way of submitting State transitions with uh zero knowledge proes so you package the information and S send it to ethereum but then for the data storage you're actually using for the State Storage you're using a data availability layer um so what we actually chose is a uh ZK sync setup um so we're using the Z casing era as a basis and then using aail a data availability solution um to actually lower these transaction costs and what we're actually doing here is that we're looking at the blockchain Tre level so it's a classical problem in our space where um you have scalability um decentralization and and security and you can't have all those three things uh at the same time so you have to compromise somewhere I think it's about applying the right amount of security uh to achieve the scalability um and same thing applying right amount of decentralization to achieve scale so suggest you you're saying you never want to compromise on security to achieve scalability is that is that what you mentioned I I think it's like you you will never compromise security if you add apply the right amount of security I see so if you think about um if you have like a house and you need to secure the house uh it's kind of like choosing the right side of Si size of a lock you know or like security measures there um so like if if you have a lock that you know is sufficient to keep people out you know you have enough security um if you build you know a wall around the house um you have more security if you have if you build five walls around the house you have even more security if you put alligators between all these walls uh you have even more security right um then you know if if if you have a you know helicopter you know s on the roof exactly so it's all about how much security you need for that use case that enables like ownership right and that's what we're looking here and currently the way ethereum for example works it's more of a light switch so you you basically get the the the security that ethereum provides with its staking Capital um with l2s you can inherit that security and then the next stage is that well you have also like reaking where you can kind of like decide how much of security you need and validum is also part of the idea where hey let's use the amount of security we actually need for social and consumer transactions where we enable ownership so people can own for example their uh identities they can own a feed that they create and collaborate with other um uh people uh online or they can create social graph graphs and connect and basically create this ownership where um that that ownership doesn't exist in a in one social network and a in a platform that they users have more power and they own it so that's the kind of like a reason how we went um about it and and when we actually turn the mindset of uh users not needing to pay fees it's much easier to own onboard new users because you're using the blockchain the same way as you're using internet today when you browse between different web pages you're not really paying AWS or the developers other costs you know the developers are observing these costs so the same way as you're paying AWA bill you'll pay a very tiny bill of transacting on lens for example electricity uses for your gas like exactly and if you are like aspiring developer and you want to build on lens Network obviously at that case we approach these teams or you might approach us and we give credit so you can actually go and build and and try to reach that product Market fit um and if it happens obviously you don't need us anymore um so we want to like even jump start and and and give credits to to to build on lens Network Amazing well kind of maybe my my last question here will be U kind of bringing it back to product Market fit there's a lot of people who are starting companies especially in defi um so whether even if it's not defi like what kind of general advice you want to give to Founders now especially in 2024 for how they should think about product Market fit and actually getting some early indication of things are working and kind of use that to hone in on what they really need to build and um kind of what would you say to to Founders here yeah biggest probably learning is to find the 10 um 10 most important users uh for your product and and really um have a good discourse with those users understanding what they're doing and also talking to them I think it boils down a lot to the users and one of the reasons like people actually love to use like really love a um is actually the the application so we did this amazing work on the the protocol side but then when you go to the application they have a lot of things uh there for example we disabled a a feature um and as I mentioned we don't collect that much data so we disabled one feature um for a couple of weeks and we get a lot of like complaints that like why did you do that because like like you know this is why I like to use Ave like this is why a is so amazing because you have this feature and you realize okay so you know you know let's fix it and put it back in on the application Level so um the protocol itself or like the smart contract piece you know can't be just one one like a um solution you have to think all the way to the experience the application and even the way that the users own board so the more data you have and understand the whole user Journey um how they get on chain you know I mean it will be good to even know like how they start their day you know maybe that's helpful too but how they get on chain you know what do they come from coinbase do they come from like onr um you know what wallets you're using what kind of hard the steps and and just thinking how do you eliminate some of those um um steps around do you think this applies to the MVP as well or L for hey you kind of have an idea you've built it and then all you need to do now is listen to the users and see like if 10 more people like that same thing and you can of then go directly into that Forest but is there any distinction or Nuance that you have for what does it mean to ship the earliest version yeah I do think that you should ship as early as possible um and one of I think one of the challenges also is that like you want to ship something that works as well I I think there's there's one school that thinks that you know ship something that is half usable or like you just get it early there um so so do it if you think that there's like just like a very low like Market window to to to create something but very usable but like if the product really breaks and it doesn't work like people will go somewhere else uh eventually um so there's nothing wrong to actually wait and and build something really well I'm pretty sure like if you know it takes you one year um to build something I'm pretty sure D5 will be still here you know so it's it's kind of it's been true for seven years I think year eight will be just fine exactly because people have this you know as a as a builder you probably will have a lot of anxieties and one of them is probably this how you get it faster you know you see competition coming you see people talking about a lot of this the same category you're building so all of this things are really distractions like if you just focus on the users um you can remove a lot of that anxiety and and focus on how to just make a better product and if you ship something something early very simple uh make sure it's good enough that even if it's simple people can use it and and they can create a relationship with your product amazing um stanie thank you so much for this amazing conversation hopefully you had some great insights from this too and uh for those of you got the socks enjoy them and uh please give a big round of applause for Stan thank you so much [Music]