Fireside Chat I Henry Stern I The Users are Here with Privy I Pragma Denver 2024

[Music] for this one I want to bring on Henry from privy to tell you about what it means to actually make things super easy and friendly for everybody to use in the world of wallets and integrating that experience into any applications you can think of you may have already used priview without even knowing it we'll get into how and when and where with Henry directly so please welcome him on stage thanks for being here Henry hello hello there we go um there's a good chance that people have used privy already in the last few months here in this audience but uh what I want to do is before we get to where um want to kind of have you sort of talk about how preby started and uh what was the early version of this and uh and where are you now so it's fun to speak here right after Juan because the early version of this for me starts at protocol Labs where I worked basically for a long time before uh starting privy um so I I came into crypto through cryptography proper and I got pitched by a fund why don't you come and do research here instead of of trying to get a doctorate and uh and you can do fundamental research in a space that actually ships products which felt like a very good deal um and so the you know battle lapse pipeline is uh a very effective know the model Works ex it's a very effective pitch I think for prospective candidates who want to go into research but want to also like ship products um and I think the honest answer is I so I worked on on fcoin for a long time and I got very excited about um a lot of the ownership Primitives behind this and about the notion of web 3 Beyond crypto and I guess I'm making a distinction here between uh Financial infrastructure and decentralized infrastructure to balance ledgers and uh decentralized infrastructure that enables user data and user assets to be controlled by users in a deeper way but I think in working on a very sort of research uh minded part of the stack and on consensus I got a bit frustrated with how far I felt I was from end user applications and it struck me that if you look at the core development of meaningful Technologies um they are built in the place where user need meets new protocols and new technologies and so like uh I I I am I'm kind of stuck on the idea of like what if AWS had been built by JP Morgan rather than a book seller how different would the cloud look today and um we closed on the weekends between yeah a Fidelity uh their their their trading arm on crypto is closed uh outside of business hours which is very fun um but um but yeah so I think that's what got me into it which is I really believe in the space I'm extraordinarily excited by what these decentralized Primitives can make work but it will only impact mainstream users lives if we make it easier for it to reach mainstream users in the first place and in order to do that we need to abstract away a lot of this complexity for Builders so they can feel focus on building good products but products that make use of this infrastructure to begin with so there a very high level you know why we got into this um yeah what was the V1 of privy yeah so V1 was I think our core take was uh maybe having worked on consensus for a while consensus is extremely expensive uh getting notes to agree on the state of the world especially complex State uh requires resources that's why God invented tokens and um that coordination is not necessary for everything it's only necessary for uh things that must be trustless or verifiable by definition and so I think our take was uh if this stack is to work on a large scale um the Great products that will come out of it will be hybrid products that marry onchain data and offchain data very well together and the Privacy implications of putting user data on chain which we must uh if we are to build user facing applications are absolutely horrendous and so I think the question for us was how do we build a stack that makes it easier for developers to basically marry onchain and offchain systems in the right ways it started off basically as a privacy preserving data store a way to associate real world data to wallet addresses of chain and very quickly basically uh onboarding as a core problem got pulled out of us by our early customers who were basically using us to associate you know email addresses phone numbers other wallets to one another and trying to basically get an understanding of who's actually using this so I can build an experience around them um and so that that led us to the the form of the product today and kind of just jumping right in what is the product today and and I'll kind of reveal the answer what is an impeded wallet yeah so so we see basically as a as a user management and onboarding kit it's uh a tool set that enables developers to onboard any user regardless of how familiar they are with uh crypto or not um so it brings together wallet connectors uh to bring on folks who are coming in with their own consumer wallets know exactly what they want to do with the product have the level of control and the expertise they need to select which chain they want to go on tune their gas prices etc etc um and it gives developers these embeded wallets which allows them to sort of abstract away the key management system from the user enables them to bring on users in a way that's familiar to them through you know SMS email social logins and basically creates the self- custodial abstraction on top of a key signature system um that both keeps the developer away from any cryptographic materials keeps the user away um from any cryptographic materials that they know of um but gives them basically access to keys that they are the sole party to control so an embedded wallet basically is a system whereby you can log into an app with an email keys are provision on your behalf and those keys can be used to uh take onchain actions for you as you interact with the app one thing you've kind of brought up a few times is just kind of mirroring the the web two and and or rather just centralize it and decentralize aspects of of um of this Tech so in the case of the embedded wallets that developers are using what parts do they have a higher confidence in on are just not being fully are actually just temper proof or they have confidence that like you can affect that as a as a centralized service or they get to still be true to the values of the the web three system they're building on top of how do you kind of help them reason about using these wallets instead of just kind of having people get a mamas account or forcing them to install another app to kind of do all these things yeah it's a great question um I'm going to give you maybe annoyingly nuanced answer May we'll see I I I think we have to distinguish across sort of three things that are not the same even though they are often bundled together um decentralization um security and custody and there are moments in which they're aligned and having a more decentralized system makes it more robust to certain attacks uh and there are moments in which they are opposed in for example having users uh custody private Keys uh often create security risk because these users will you know put their keys in in last pass for instance let's forget or yeah or forget which is a ux issue so that's maybe the fourth one um so I think for us we've always kind of taken a uh non-religious view on this which is our main goal is user empowerment through two things one self- custody the user must be the only party in control of the keys um developers can be granted permissions to take certain actions on behalf of users when users are present and willing uh but the user is the only party who ultimately has control over whether something gets done and always has the ability to yank that something out of the system they don't like it um the second one is interoperability uh these assets and these systems must be able to work across the apps in an ecosystem otherwise you might wonder what what is it good for and in this case are we talking about the identity of that user or something else both the identity and the eoa or signer that they're using to custody the assets themselves um so I think for us we we looked at this and then we looked at ux and I think the question we we looked at early models of how do we decentralize this and they're really great companies doing work here by the way that um I think we're very excited to integrate with uh in time but folks like lit or entropy these sort decentralized NPC networks and I think we came to the point of uh user control is what we're optimizing for rather than decentralization and in order to ensure proper user control um we must split keys in such a way that no one but the user ever has access to reassembled key set um and we must give the user the means of of sort of having an escape hatch so that if things are ever not the way that they like we understand there are limitations uh to what embedded wallets can do and I'll give an example here um embedded wallets enable the developer to set the network that the user on so the awesome ux Advantage is you don't have to switch Network when you connect to an app you're already in base or you're already in arbitrum um and the notion of a chain itself is subtracted away from the user the downside is if I send mainnet eth to that app it's stuck um and so we've improved the ux but at a cost of basically user control to some extent here and I think our take was uh there are too many apps to be built and too many users at different points in the Spectrum we need to make sure that there's an escape hatch so that users if they ever find themselves in a ux hole because of the uh Pathways that the app has built can always eject and come back to you know what is I think a very good point in the design space the consumer wallet um so I think it's a different point in the trade-off space that is being made but it's a very good one um yeah happy to talk more about the the architecture of metaal said that's useful but yeah actually uh well I do want to tackle that but maybe in a few minutes um one thing that you kind of also talked about was well you kind of noticed that people are not using these readily at scale um and partially that's just frustrations around making that not easy of interaction uh given how things are right now and partially it's like maybe we we don't have an obvious thing that we can use it for day-to-day obviously cost and transaction these are all are all these factors but um the goal here is that effectively with an integrated wallet you are increasing the odds of people using this app and having them not worry about what's happening behind the scenes and that's kind of I feel like one of the factors we need to enable more consumer crypto um so the question I have is like how do you define what consumer crypto is and uh and uh I'll kind of put you on the spot by kind of having you maybe speak on behalf of some of your customers like who who are they how do they kind of perceive this space and then how do you sort of think about their success as it relates to kind of what you are able to offer in that abstraction process so I certainly can't speak on their behalf but I'll speak on my behalf as an admirer as a um as a as a spectator of your customer I I think we made an early decision that was a really good one which was to build on both ends of the extreme we spend an inordinate amount of time building very good wallet connectors so that you know in that case where you're an Android User using Firefox and you open up an app and you have have to connect your metamask but you were just last on coinbase connected to the side on another Tab and you click the button and it says connect and you switch tabs and then you switch back um every now and then the Deep link doesn't fire so you have to switch back manually and then you like click sign and then you that user must be served and that user must be served properly where the developers ripping out their hair about like basically the bug reports that they're getting on this um and we were like should we take this on and the answer for us was absolutely because um I think the most sort of forward crypto users show us the way in terms of what can only be built on crypto rails um because if we're only talking to folks who want to hide crypto altogether uh then the best way to hide crypto to not build using crypto at all and you end up with basically these like crypto marketing apps where I'll drop you an nft what is the nft good for nothing but there's crypto involved and it's really cool and it'll like you know be the project at a chief Innovation officer at a company that isn't actually that invested in understanding what do open systems give us we'll get and so I think our take early on was we need to straddle both because that'll make our product better and we need to make sure that the object we give back to developer when a user logs in is the same regardless of how the user is logging in um and so that's kind of B the montra user a user and then we follow the developers where they go um I'll give you a few examples but basically along the Spectrum you have apps um one of them that I I really like is an app called Blackbird and Blackbird is building a restaurant loyalty program um it was built by a fellow called Ben lenthal who built before that eater and resi and we get very excited when these basically industry uh insiders who really know there very popular app yeah and and really know consumers as well um are interested in decentralization and are interested in understanding what does basically ecosystem uh Network effects that this gives us uh what can they bring to a product and so the the sort of leading indicator for us is how many bz can we get who uh comes up and tries to build in crypto but doesn't end up breaking their teeth so much on the tech itself that they're like you know what it wasn't worth it we'll just build this as a centralized product um so that's one of the spectrum um in the other side of the spectrum we have a customer like hyper liquid which is a building basically a der derivatives and per trading platform on Shane um very web three native uh extremely technical crowd who is engaging with basically price action across chains um and a very different set of needs and then maybe in the middle you've got folks like Zora who are making it easier to create and build on chain sort of unlocking you know artist imaginations to actually in things that much quicker um you've got folks like Courtyard who are translating again real world assets on chain with Pokemon card trading and the like um and you've got you know folks like frch who are building sort of social applications and I think basically when we look at these our question is how well is their core customer served um does the person who intends to join the app are they able to do it in a way that feels natural to them and can they do it in fewer than you know five Taps and then the second is um what is the ratio of monthly active users to monthly transacting users for these apps and how quickly can they get to a first transaction I think that's a really interesting metric too and yeah that's super helpful I mean I've used a bunch of those apps already that you just mentioned and the fact that they're just being abstracted away in an easy environment thanks to privy is actually kind of awesome and who really puts the embedded in the embedded wallet here that because you wouldn't know otherwise um I had a question but I I think I should skip that because I want to talk about something that uh that I said we promis we would which is understanding now just not just the identity but the data Parts too so uh you said that the V1 privy was being able to have data associated with an an address so people can kind of query it and I think just a few days ago or maybe last week you announced um uh the global embedded wallets what is that what's kind of the the goal here and what will be the possibilities of this yeah I so we wanted to be very concrete I think you know uh sort of uh tooling spawn the apps that it enables and so we wanted to make sure that we opened up the aperture for uh Builders who want their users coming in for the product itself not for the crypto of it potentially uh to have an easier time and so we started with you know there are two superpowers that crypto gives you uh I would say one is self custody the user in control the user in charge it is my assets it is my keys and I decide what happens with them um and the second one is interoperability these assets as they were are composable are reusable across apps and we can sort of let a thousand flowers bloom I this dates me a little bit but um I think one of the more exciting things that I I still am airm uh I I can't I hope someone research or re uh Kindles reignites uh action around loot which was a project where you had you know six words and you could do whatever you wanted with the words and thousand people you know were building various applications on top of it um and that seems like a core superpower we tackled self custody first how do we make self- custody simpler how do we make sure that we can build in better wallets systems that scale that are highly performant and which only the user is in control and with that done so very much a work in progress but we feel like uh you the the ux is good enough enough security is good enough that um you know we went live with an fully open product a few weeks ago even though before that you had to talk to us to get access um with that done then the question is how do we make sure that users can reuse these assets across uh different apps that they connect to so this means what's an asset in this case um when you I'll use blog bird as an example um or maybe I'll use I'll use POA as an example um so POA is a customer they give you a proof of attendance sort of token that you went to an event um maybe I got my Pro app for pragma and uh maybe there's a collaboration between pragma and Blackbird that proa is not involved in at all um but basically I'm able to use uh the system with which I may be tagged into the pragma event in order to get a free drink at restaurant that I go to next or uh you know in order to unlock access to uh an investment opportunity on a uh trading app uh elsewhere and basically making sure that these assets can follow users around easily from app to app to app without them having to explicitly connect these different app accounts together is really really important so is that key Insight here that this these assessor data here is a collection of onchain and offchain things because in the case of faps I know I have the n F so even if I don't have any direct connections I can still say hey this is my address this is the nft I own and I still don't need permission but if I wanted to say this is the nft and this is my personal address associated with it to get some coupon then you need that context to be passed on to the next day yeah and I think how easy you make things ends up being a major part of whether or not it gets stun it is theoretically possible to do that today with privy there's more Bud understanding is it just onchain assets only or is it I understand um to start at onchain assets because that's where I think there is the clearest value uh for building that interoperability across apps as we look forward you know things like kyc we spent an Ord amount of time early on looking at you know what is pii that people would be interested in having carry over from app to app app and kyc data is one of them uh a lot of of these systems are regulated require kyc on behalf of their customers um having to take you know 18 photos of your passport with like a handwritten note is a major pain in the and being able to associate that data uh not just to given wallet address but to give user address uh or user object is extraordinarily important so I think it goes far beyond onchain data but we're clearly starting with onchain data because that is where uh there is the clearest and most presentent use case that's awesome and and kind of May open the can of worms here but so far we just kind of talked about the web what about mobile here yeah so we someone on the team was was talking about so so we have a uh to be concrete we have a react native SDK we have a swift SDK that we don't talk about but that exists um and we are working on a cotlin SDK and um the reality is this is also where pwas have been very interesting as an easy way of getting users back on um so mobile has uh probably three great things going for it one um it is sticky you don't need to rigate to the website it is there hopefully on your home screen hopefully doc but at any event it's here and you know it's here two um it is uh push-based rather than pull based uh apps suffer a ton from having to have the user remember to come on and see what happened it's especially true in crypto because you know settlement layers on chain take time and so my transaction's pending and I'm going to click away because I don't want to look at a screen for you know 20 seconds to two minutes and I'm going to forget to come back very often um so being able to push things to users push notifications are a major on loock and then um finally I think it's it's mostly the screen where users spend the most of their time so I I we talk a lot and this is a Circa maybe 2010's thing it was like solo Mo social local mobile and I think this is part of what I'm very excited about coming into 20124 um we're seeing a pace of activity around experimentation in consumer facing apps that we just haven't seen in crypto at least as I've been here I just waiting on so much of that abstraction on scalability that I think we're kind of at that point where you can exactly and so seeing you know we we had the the the day we launched our self serve we had an app go from uh signing up to prod in I think 40 hours with a fully built endtoend product that worked on mobile and that felt awesome exceedingly exciting to us which is you no longer need a team of 15 including three phds and a cryptographer in order to build something for an Absol approval that tip wow um well maybe just to kind of end this um as as a final question uh what are you excited about in terms of kind of the next product updates or or things that you're tackling or just kind of harder problems that you think are worth abstracting and uh what can we look forward to uh the next few months for privy so I I think our work comes in like really three parts the first is uh Global wallets and basically launching easier means of inter opt for users uh where they can decide whether they want to create sort of fresh identities as they come into apps or reuse the ones that they've sort of been collecting before um the second is basically on ramping and funding of accounts this ma MTU ratio I think remains now one of the sort of key things that's holding back crypto so things like uh being able to sponsor gas um and you know the idea that users would have to pay for their own infrastructure is insane in a world in which the cloud exists basically we would like it to go away yeah and so I think it there's a lot of runon effects in terms of you know what do unchain business models have to look like in order for it to be worthwhile for developers to sponsor gas on behalf of their users but things like gas sponsorship things like betterer on ramps is a major part of the work that we're doing how do we get a user to their first transaction and how do we fund these wallets more easily um and then basically the third is really broadening uh the funnel with uh more clients so SDK is that enable people to build an environments that their users are native to and more chains moving uh beyond the evm as well uh lot of those environments awesome well Henry thank you so much for that amazing conversation and uh can't wait to see more and better wallets everywhere they ideal on privy so F Js