EP.51 [EN] - Ryan Li: Building a User-centric and Self-sovereign Social Graph Cyberconnect
50:22 Share

Hello and welcome everyone. I'm Mabel Jiang, the host of 51%, a podcast series presented by Multicoin Capital. This show is the exploration of blockchain's rapid development across Asia with a particular focus on the perspectives, communities, and operators based in China. My goal is to bring Eastern perspective to West and Western perspective to East so we can better understand the crypto's unique market structure and how these distinct communities think and operate.

This podcast will feature a mix of English and Chinese discussions. The language you're hearing now will be the language I use for the rest of the podcast. Stay up to date with my latest episode by subscribing to this podcast. Thank you for listening.

Mabel Jang is a principal at Multiquin Capital. All opinions expressed by Mabel or other podcast guests are solely their opinion and do not represent the opinions of Multiquin Capital in any way. This podcast is for informational purposes only and should not be construed as an inducement to make an investment or relied upon as investment advice. Multiquin Capital may at times hold positions in some of the tokens or companies discussed on this show. Hi everyone, welcome back to 51%. This is your host, Mabel Jang.

Today, I'm very excited to have Ryan Lee, the co-founder of CyberConnect. He's also the CTO of the project. Welcome, Ryan. Hey, guys. How's it going? Hey. I guess before we jump into the discussion, we'd love to have you tell us a bit more about yourself and your background. How do you come into the space and your previous experience?

Yes, of course. Well, I got into crypto around 2015 when I was just in college. I did a bunch of hackathons around Bay Area and there was a couple of them that involved Coinbase API. So we did a bunch of Bitcoin stuff back in 2015. That's where I learned Bitcoin. And I didn't really pay attention to it until 2017. I saw this project called Steemit.

which was this content rewarding platform, something similar to Reddit, but rewards content creators. And I just found it fascinating. And that's what got me into the rabbit hole since 2017, been building here.

Got it. Since you were always pretty excited about the content creator economy and stuff, you didn't mention your experience with DLive. We'd love to maybe learn about that a little bit since you're also one of the co-founders of DLive.

Yes, since 2017, we got into crypto. We found this really interesting idea of helping those content creators monetize. And one of the interesting ways that I, as I said, was through this Steemit crypto incentive, where we found it's really not working in the long run, where we started to build our own. We built this own blockchain called Alino Network, which is built with Tendermen

Now, Tenement is rebranded to Cosmos. So we were probably one of the very early developers on Cosmos back in 2018. We built Lino Network as our fundamental layer one for content economy, where people can post their content on top of it and receive donations, receive subscriptions from their consumers. And we built DLive on top of Lino Network as the first product. And

and DLive was this gaming-focused live streaming platform. It's a direct competitor to Twitch and we launched in 2018-2019. We got more than a million daily active users in 2019. We onboarded PewDiePie, which was the largest YouTuber in the world. He was exclusively streaming with DLive for the entire 2019 and we had

more than 60,000 streamers with us actively streaming on DLive. And we generate more than millions of revenue every month. So that was the experience with the content, building in the content space. And DLive and DL Network both got acquired by BetTorrent in early 2020. And I stayed at BetTorrent for a while before jumping back into building with CyberConnect. That's super interesting. We'd love to maybe...

um go into detail a little bit here so if you i mean at the end you guys sold the the platform i would assume i guess like some people would want to exit at early stage and maybe you didn't so for you can you maybe talk about what you guys did right and wrong you consider when you were working on d live um and then how those experience translated um

into something that might be helpful when you're building the new project, which is CyberConnect. Yes, of course. Love to share some lessons there. I think what we did, right, with DLive was really finding the pain point for the content creators, especially for the live streamers.

they are actually one of the most undercut influencers. They get 70% of the revenue from tipping and from donations taken away by the platforms. They are just really, it's really hard for them to monetize, especially for the smaller ones after the 70% of the cut. And

And one thing that we've found, especially for me, it's so satisfying to see people making a living on DLive, getting support from a worldwide audience. I remember this girl from Venezuela. She just can't make a living. And yeah, she made a couple hundred bucks on DLive the first day she streamed here. And it's from the global audience. We don't take any cut. Yeah, that was the moments I think we kind of really made an impact

to the content creators and the things that we did wrong i would say is really we have been focusing yeah we focus too much on the product and we focus too much on building everything out ourselves we built our own blockchain our own layer one and we built our own um platforms and the applications on top we found our own influencers like pewdiepie like 60 000 more of them

I think we did everything ourselves and we didn't get the help from the community. We didn't build with the community. So that was probably the biggest lesson we've learned, especially I learned. And now with CyberCanal, we're building in public. We onboarded more developers than I can imagine within just two months to build together. So that's something I think really what changes me through seeing how Ethereum really is

become this successful and also other layer ones become this successful was really their developer community and going in public. That's super interesting. A few things that you mentioned, I definitely felt the echo. First is that the...

like trying to build too many things together at the same time. I think people who came from like 2017 and 18 all have that tendency of trying to build the front end and then protocols and all kinds of use cases, whatever it involves. I think what you said about Lino plus Eli was really a interesting approach that people like a lot these days about the protocol plus power users. However, I think there is a,

kind of like a undiscussed consensus that, you know, the protocol itself should always be kind of prioritized. And then the power user, which is, you know, could be a good addition because like it showcased the world of how you can use it, but still the emphasis should go to the protocol. Cause like, that's the most, I guess, scalable thing. And then a lot of people, like it could create a moat when a lot of people are on the network and contributing to the network. I,

I think the other thing that you mentioned about empowering the small creators, I thought that was really interesting because at the same time,

um, audience was trying to do the same thing. I think they also started 2017. Like, I think the only difference is that like they were working on audio or like streaming for the, I don't know, like podcast or some of the other people, like it's more focusing on audio rather than video like you guys do. So I guess that's why like it was like the result was different because like you don't really have to burn that much money for what it's worth.

But you guys were both at the same time, like trying to empower the long tail content creator, which I found pretty fascinating. Yeah, we'll love to then like, you know, jump into the discussion for CyberConnect. What is CyberConnect? Maybe you can talk about that first.

Yes, of course. So CyberConnect is this decentralized social graph protocol. The ultimate goal for us is really to connect everyone and really bring back the data control and the utility to the users instead of to the platform. That is the case right now. So currently there are a lot of issues with the social networks that surround us.

For the developers, it's super hard to create a new social app. It's super hard to bootstrap through the early days. And for the users, you're just locked into the platforms that you have been using for the past probably a decade or more. And you either get entirely segregated networks from all these different apps that you use, or you just have to find your friends everywhere.

And no matter what, all those data or all those friends that you digitally kind of own was never yours. Those data are always the assets for the platforms. These are just the most for the platforms. Just like Facebook can take away the handle for Metaverse.

in a day, those data were never in the hands of the users. So that's what we're trying to build. We want to give back the control for your social network data to the users to try to democratize those social graphs that you have on those platforms. Got it. That's super interesting. And then how is CyberConnect technically designed to offer something like a Web3 social graph?

Yes. Yeah, so a Web3 social graph for us, it has a couple aspects to it. It can be divided first into two categories. One is where users actually want to engage in the social connection manually. For example, if I want to follow someone, I click on the follow button wherever that's showing up or if I want to add a friend through clicking a button. So these are

We're building this self-sovereign data storage for those social connection data where users can actually own them and take them around to any other applications. It's like your friends in Decentraland can also be your friends in Sandbox, can be your friends in every other platform. The folks who are following on this NFT platform or NFT marketplace can be brought along to any other NFT marketplace.

So that's the first part where users own their social graph, where they actively maintain. And the other part is we do a bunch of indexing and recommendation based on users' existing behavior. So for example, if we both hold the same FTU, that puts us into the same FTU community. If we hold the same pullout badge, probably means we went to the same event, we've done the same things, or we have the same interests. We

analyze and aggregate all those data from different on-chain platforms or different public transactions and put them into this recommendation engine that would provide to developers and to the users for first user discoverability. You can immediately know, oh, these are the friends that I have interacted with, or these are the folks I've had fun with before. And also for developers, they immediately see these data coming into their users

after they sign up, you can immediately recommend them, oh, these are the folks you might want to follow for, let's say it's for an NFT Instagram platform. So yeah, that's a higher level technical architecture for CyberConnect, those two parts. I guess you explained

like how you are kind of offering the service and then in terms of the technical architecture, as in what different protocols or what infrastructure you leverage to offer those. Can you maybe expand on that a little more?

Oh yes, of course. So yeah, that was a high level overview. So in terms of the technical protocols and infrastructures that we utilize or we build upon right now, we mostly build on top of Ceramic for this self-sovereign data part where users own their social graph that they manually added. And for the other part, for the indexing, we have our in-house

indexer architecture. We use multiple databases, the relational databases and graph databases to build out this big query engine for the developers. So yeah, those are the stacks. So we don't build directly on chain.

because the gas cost just doesn't make sense right now. Excuse me. And also we really love the design of having this portable and self-solid data that's interacting with other platforms, other developers that's provided by Ceramic. And also Ceramic is this mutable data stream that's built on top of IPFS. So a little bit background.

Yeah, I was going to ask you, like, how does ceramic work? But you just kind of give a one-sentence summary. I think for people who want to learn more, like, the biggest differentiation how ceramic works

compared to some of the other storage solution is that they are a mutable database. And then it's definitely easier for decentralized applications to work on top because it's essentially storing the data streams, like information streams. And then CyberConnect, if I understand it correctly, it's a middleware between the decentralized application protocol builders and the

between those application builders and Ceramic. Can I say it that way? Yes, in a way. And also we... So between... More like between applicant developers and Ceramic and other layer ones. So we do index a lot of stuff from layer one and from other data sources, all trying to infer social connections. Next one. So Ceramic,

Here, you're tackling obviously very large topics, and I'm sure a lot of people have different thoughts on how different protocols or solutions should really tackle these problems of social graph or connections in general for applications in Web3. So what are some of the key features that you really care about for CyberConnect in the design space of solving this problem? Can you maybe expand that a little more?

Yeah, of course. Well, firstly, we built in a blockchain agnostic way.

I think blockchain has been said a lot, or cross-chain has been said a lot, but we have a very different approach. We don't build directly on any blockchain and we don't build our own blockchain. Most of our storage and the work are stored on IPFS, on Ceramic, and potentially other decentralized storage solutions. We don't want to build another layer one or specifically for one blockchain that introduced a lot of barriers for developers

let's say if we want to build another blockchain ourselves it requires users to bring their assets to a new blockchain and also require developers to learn something new so we don't do any of those we

built with the compatibility of any blockchain. So currently we support Ethereum and Solana. So any users could sign in with their already familiar MetaMask or Phantom or other wallets to CyberConnect. And for developers, it's extremely easy to tap into their existing stack. So let's say if they are already building on top of Ethereum or Solana for their smart contract, for their tokens, for their NFTs, they can use CyberConnect

just as easily as just using those layer ones directly.

And secondly, we are extremely focused on building something that's self-sovereign. As I said, Ceramic is this premise of providing the self-sovereign data and we are on the same, we're pretty aligned in that regards. We want to make sure that the data are always in the hands of the users. Just like, no, not your key is not your token. We want to make sure that they own their social data, social connections specifically as they

what they are already doing with their tokens. So they actually own those data. Those data were never a part of the platform. And three, we are really trying to optimize for high efficiency in terms of mutable data, because I think social networks and social connections are just a very

a very complex system and also it requires a lot of technical design to make it work in a spiky traffic in all kinds of circumstances. So we emphasize a lot on making sure all the architecture that we have are efficient.

And the last is really providing a great overview of user-centric index. So instead of doing indexing on project-based or on contract-based, which a lot of existing data platforms are doing or the graph is doing, we focus more on how do we bring those data towards the users? How do we go from user-centric model to

Instead of going to different platforms to look for what are people doing specifically to that platform, we want to make sure that the developers can immediately get the sense of where this user belongs in his cohort or where does this user want to interact with other people and who are they really connected to? What are the projects that might interest them? So these are some of the characteristics that when we're building out CyberConnect, we really want to follow.

The last point is super interesting. I'd like to go into that. So when you say, I guess like for some of our audiences, they might not be as familiar with the graph. Maybe you can explain like what is smart contract focus oriented versus like user oriented. I mean, you'd explain what user oriented is.

Yeah, I can. Yeah. Yeah. Yeah. I think that's one of the biggest, biggest concept around a user actually owning their stuff. Since

in a way in a way i i can give a very concrete example so if you go to if you want to get involved into nft platform there is basically no way for you to find that information unless you find their specific website their specific on their website specifically you find where to join their discord and everything is happening within the project space

I also maintain their specific NFT on their contract. So everything is very limited to finding those specific projects and going from those projects to where the actions are happening, like maintaining NFTs or joining the discussion on those Discord.

These are what we call a contract centric or project centric model of doing those indexing. You get all those data around those projects, like who are the biggest token holders for this project, who are the most involved persons or what DAOs are buying those tokens or whatever. Those are all surrounded with the project in the center. And for us, we want to provide more value is when

when the user can actually help we can actually help the users to discover what they might be interested in or what projects their friends are into and then they can immediately find the project instead of going the other way around that's interesting so i guess like you can also say like what graph is focusing on is like event event based data or event based um credentials

Yeah, I think, yeah, in a way, in a way. And also, yeah, I love this. Another analogy that I use, I love saying this, like when I was a kid, I go hang out with my friends and we just meet up and then we find a place to go or whatever we want to do. We find the person first.

But now it's like I have to be, I have to log into Discord before I can do anything with the folks that I'm connected to through my NFT. I have to find this specific platform. The reason behind that is actually because all our assets, all the friends that we have or the connections to those friends are actually owned and segregated by the platforms. You have to find the platforms first. So that was a big reason why we're building out this user-centric model. I know it's a little abstract, but I think this is the right way to go.

This example that you gave was very helpful. I think that helps people understand what's the difference. I think to some extent, some of the things are actually quite similar. The part about interest grab, I'm not sure if you mentioned that, but that's the idea. You try to find things that you both like and then do it.

But then there's a more emphasis of the human or the specific individual component for CyberConnect versus I think for the graph or whatever, it focused more on the event itself rather than the person first. So I think you mentioned that finding the first person first is super, super helpful.

Right. And then I'd like to go into the discussion for ceramic a little bit more. I think you are probably one of the first players in the space that tried to leverage what ceramic offers to tackle the problem that you're trying to solve. In the process of using a new infra like ceramic, I'm sure there are some challenges. So I'd be curious to hear that as well.

Yeah, well, first of all, I think I have a ton of good words to say about Ceramic. I came across Ceramic probably mid-21, so mid-last year, when I was just looking through interesting projects

data models and data solutions and ceramic is definitely the one that really stands out the way they have this self-sovereign way of describing all the data and in the mutable commit log kind of architecture where they also have a strong emphasis on user-centric model which is exactly what we are after so that's where it kind of makes a ton of sense for us to build on top of ceramic directly

So, yeah, that was the background of why are we building on Ceramic. And we are definitely, I think, one of the earlier players and also one of the braver ones who have, like, onboarded all our users onto Ceramic. And, yeah, we've experienced a lot of stuff, some hiccups with Ceramic team. We've been fixing a lot of the, some of the scaling issues with them together. But, yeah, I strongly recommend anybody building and

any kind of application that requires data storage around users, or data storage that requires some updates from the users to build on top of Ceramic. That's for sure. In terms of the challenges that we've seen, well, first of all, we do see some performance issue. During Christmas time, we were reaching about... onboarding about 100,000 users.

and at that time i mean on the way of reaching 100 000 users at that time uh we did experience service degradation so nobody was able to sign in uh it's their design of a did

which is another topic to discuss later. I think their design of authenticating user is just not up to the level of traffic that we're experiencing back then. And then there is this, they do have more complicated UX on the sign-in process where users do have to sign a specific message that says allow

or allow something to control your identity, which is kind of weird in terms of the wording. And then, yeah, I think that's pretty much something related to UX, something related to scaling and performance there. Otherwise, I think it's a great platform to build on.

So the DID part, if I recall correctly, was what the 3Box team, which is the team behind Ceramic, was working on when they first started. How does that exactly work? Yes, yes. I should have explained DID in the first place when we were talking about blockchain agnostic. Yes, under the hood, every user on Ceramic is Ceramic.

under this DID, so a decentralized identity, they are not directly interacting with the data streams using their Ethereum or Solana private key. They're using a mapping from those wallets to a decentralized identity and use that identity, that DID to act as

as an authenticator to the underlying stream. So basically in that way, you get the power of merging different accounts, merging different wallets, even across blockchains to one identity. So that's one of the power that Ceramic has. So that for us, I can easily link up my wallets on Ethereum and also my wallet on Solana into one identity.

So that's the DID part. And for the DID to work on Ceramic, you have to create those documents. You have to create a document that specifies, "Oh, this is my DID." And behind the scene, you have to tell this DID that the only way to authenticate is through this signature that's only

That's only forageable by my own private key of my Ethereum private key, of course.

So that, yeah, it's a complicated system. Sorry about the rambling everything about the idea. No, no, absolutely. I think this discussion was meant to be a bit more technical. So I'm glad that you were able to explain in a relatively succinct way. So it's good that you are... I was actually pretty impressed when I first saw you guys are using ceramic. They were just coming out

to the more public mode at that point. I was like, oh, wow, these guys are already using ceramics. So the kind of grow-together mode was almost expected. However, I think for you guys to really work, you might...

like also need to have some mitigants for some of those challenges that you were mentioning. So what's your current solution to mitigate some of these challenges? And then I guess like, what are some of the potential points of failures in these temporary solutions? Yeah, there are definitely some challenges that we've seen. Some of the solutions that we're working on, or at least some current workarounds,

I can describe how everything is going to look like in the short term and also in the long term. In the short term, we do have a hybrid storage system. Instead of relying everything on Ceramic from the first day that we tried it out, we're now writing data to both a centralized place and also to Ceramic.

In that sense, we do have a backup in case there is any service degradation from ceramic. So to be able to store those data in a centralized fashion, we also have to guarantee that those data are always available and also verifiable.

We have this design of user generating signing keys on the front end and using those signing key to sign a specific message when they are following someone and put those data

the signed data onto our back-end central server for our server to relay those data into a decentralized storage like our IPFS. So that's the current solution. It's kind of a temporary solution for us while we are waiting on some of the new improvements and upgrades from the Ceramic team. While at the same time, I think a hybrid solution is also very suitable for the long term.

Because we do see that

there are developers who want the data not only just from IPFS or from Ceramic, sometimes they want data directly on chain. They want to access the data from a contract specifically. Let's say if you are doing some kind of credit score or lending kind of DeFi protocol, you might want to know how is this guy related to a well-credited or a person with high credit score in that sense. So we have to bring this data to where the developers

want to build them. So that's why we have this hybrid solution right now with data verification. And in the long run, we have a couple technical designs on our roadmap that we're trying to push out during this year. One is doing a big standardization on our data model.

As I said, we have this data model that's kind of similar to a year to 20 token. Like I want to declare that you are my friend or I'm following you. It's somewhat similar to I'm owning a specific token and that token represents this relationship. So in a way we are trying to create a new data standard that closely represents this kind of assets or token standard.

So that's one way where one of the things that we're working on. So the data center improvements, and then we're also working on some of the compatibility with signing with Ethereum. There's this new proposal signed with Ethereum where we're working with Ceramic team on being compatible to that new sign method.

Yeah. So the other question was some of the points of failure, right? We are trying to mitigate most of them through the hybrid data solution that I described about and also through our new data model so that those models, those data are actually more modular and easier to store and easier to maintain. Because right now we're throwing a lot of this stuff into one giant blob for every person. And those data can be

more prone to loss or to other kind of service degradation that happens with ceramic or with other data solutions. So yeah, so those are some of the solutions that we're working on and we're trying to make sure that there's no potential points of failure.

That's helpful, especially on the part that like how you're working with the ceramic team to kind of create some of the mitigants or like the midterm, short term solution for that. You mentioned earlier that you have some in-house indexers working on like indexing some of the data on Chang.

First of all, how are they exactly doing those? And then are you thinking about in the long term, like open up that part as well? Because like, obviously, I guess some people who have the similar technical ability might also want to participate in indexing and then potentially earn certain rewards from the network.

Yes. So yeah, on the indexer side, how are we architecturing it right now is we do have a giant ingest. So we do have a lot of data sources that we're ingesting data constantly.

And after the ingest, we do put those data into a database that we maintain those records. And then we transform those data into a graph database where we actually have a giant representation of how our people connected to each other through this giant graph where each person or each

address is a node on the graph where the edges are the different types of connections or different types of relationships between the nodes. That's our general approach right now. In the end, we want to provide this very flexible and very high performance

query engine for the users, for the developers to find how are people connected and how are they able to discover some hidden connections between people and between projects. And in the long run, as you say, yes, we are developing an open protocol for the indexer or for the recommendation system.

in the long run there is going to be recommender processor and career so three-party into this system and the recommender will have the responsibility of coming up with new data sources interesting algorithm that kind of provide different insights into how are these data useful and then the processor will be

just running the software to make sure that the data are always available and in duplicates mode so that it doesn't happen. It will not happen like if CyberConnect goes down, it becomes a single point of failure.

And yes, the querier and the recommender and the processor will have this data exchange where there could be some rewards in the form of a token to facilitate the data transfer between them. I think this is an interesting and open question to be discussed because I think among all the protocols or projects that around

The graph is probably the first one that started this model of you have this indexer, you have curators. And then I forgot like the what, what the third one is, but basically it was, you know, along the same line. I was actually chatting with Harry from, from project galaxy about this too. Cause like, I don't think,

For what it's worth, we are an investor in the graph, but I don't think the business model for indexers has worked yet. It hasn't proven that people can actually make sustainable money from that. So I thought it was interesting to observe, continue to observe, and how it should look like. I mean, there is a chance that the indexing part could...

be better if it's centralized, but we will see. It's also possible that it's just like right now, the amount of data for people to curate and whatnot, and also index is not large enough to make sustainable income yet. So I thought that problem of whether to stay centralized versus decentralized is still up to discussion for the graph or Galaxy and probably for you guys as well.

So I thought that's an interesting point to bring up. Yes, I think that's exactly one of our concerns as well. We've been observing the revenue generated by the graphs protocol. I don't think it's there yet. Yeah, definitely the whole architecture is up for some kind of review. But the most essential thing for us right now is just be able to find those hidden features

relationships between people, between projects, finding those social syndicates and making sure that those data are valuable to developers. Makes sense. So what's your future plan and technical roadmap?

Maybe I think you can share a bit on that. Yes. Yeah, so for our future plan and technical roadmap, a couple big themes are around reliability and scalability and user experience and also this process of progressive decentralization. Some of the stuff that we're working on, on the core protocol layer, we are, as I said, we are doing a redesign

design of the data model and the data structure that represents social connection. We want to make sure that the data center is as composable and also as standardized as possible so that people can easily adopt them. So that's why we're exploring some of the

token standards to represent social connections. But also we want to make sure that the standardization not only happens on the social connection data side, but also on the whole user experience so that users actually expect the same kind of user experience when they

sign in and when they grant access to their cyber connect social connection data to grant those access to the developers. So we want to make sure that the whole experience is as standardized as possible. Another big theme is we are also working on

getting the access control model right. So it's an extension to what I described to for the user experience when they're doing onboarding. So the sign in with Ethereum proposal or other standardized sign in proposals. We also want to make sure that the users and the developers always have this right access model so that the developers can only access the data that are granted access

access to by the users through an object capability model that we're working with Ceramic. Hopefully that gets launched in Q1 or Q2. Other things on top of the core protocol layer, we do have a bunch of developer community and developer tools

happening right now. We are hosting, we're actually hosting a hackathon right now, a showing point on Gitcoin. We do have

Tens of participants engaged in this hackathon and we are doing a relaunch of our development center site with docs, tutorials, and some learning materials, some use cases. And we're working closely with a couple developer tool and developer community sites to onboard our next wave of developers.

And yeah, we're constantly just working with developers. We have on board more than more than tens of developers right now. And there are a ton of great projects out there on our pipeline as well. So yeah, this year is full of

just onboarding developers and hopefully they can build great apps and provide more utilities to cyber connect so for all the users who are on cyber can they can enjoy all those different apps without having to find their friends again or without having to reconstruct their social graph

For sure. That's definitely exciting. Like talking about the Shelling Point hackathon, I actually had an interesting idea the other day. I went to the site and I looked at who were interested, whatever. And then I was like, hmm, if all of these tasks or different types of, I guess, quests for, like, you can consider that as a quest as well for developers.

If any of these can be pushed to the developers who have certain experience, similar to these or had certain skills that can be discovered on Chang or you can just translate those credentials on Chang. Kind of like what you were saying about interest graph and whatnot and even just seeing maybe your friend

developer have done this and then you also want to discover this and then be able to complete a similar task with them. I think it's basically a function of LinkedIn email where you can get notification or push for some of the tasks that is being broadcast by different parties.

I think that will be extremely, extremely helpful because I think there are a lot of, there's like such a mismatch as of today for a lot of these opportunities happening on Gitcoin or some of the other platforms. But then like a lot of people who have certain matching credentials or matching social graph, they cannot discover this. So I thought that was like a very large opportunity for projects like you guys, like projects you can really tackle.

Yes, I think exactly as you said. We are really there to help with the costar problem of a lot of the projects and also the discoverability, helping people to find the right community, right interests that they should at least be aware of. I think this kind of theme also comes up when just like as the Etherscan launched their messaging app,

It's kind of ridiculous when you just open the app and you have nobody to talk to. And you have no idea what you're supposed to do with it. And if CyberConnect is part of it, I already have hundreds of friends and thousands of followers. I could probably easily find them if they integrate CyberConnect.

The other thing that I thought of was like for TikTok, a lot of my friends, they were using it and I was never...

into it. And the reason is I have to train it and I hated that process. I tried a few times. It just never worked for me. I don't know why, but anyway, so I thought, you know, having cyber connect, it's helping the problem that basically like, you know, when you log into a new protocol, a new application, and then you don't want to spend the time to train them. So what that application can do is actually integrate cyber connect. And then like,

like have those interest graph available already for you so that they can customize the, you know, the content for you. I thought that was an interesting idea and I was pretty amazed by that. So given that we talk about use cases, I guess like you can maybe expand on how can different users, like meaning the protocols can leverage the service that you guys offer. And then what are some of the potential use cases?

Yeah, well, first of all, a lot of the Web2 kind of social application or protocol that's like that that's like coming to Web3, like like having the content, like writing

Twitter-like blogs or having a social profile, we'll be integrating with a ton of those projects. Well, Project Galaxy has a follow button on their Galaxy ID as well that's powered by CyberConnect. And we have a bunch of those. So user profile, content curation, or personalized content based on your social profile or social connections. That's one of the use cases.

And the others are, we do have some partners in the gaming area or game industry. We are partnering with some of the games that's upcoming to launch like this Steam-like platform where you have this one place, one single friend list that you can see all your friends, what are they playing, what are they up to, what's their rank in this specific games.

and providing that holistic view of all the different Web3 games and being able to discover those games through what your friends are playing. So that's another use case in the gaming area. And the other one that I'm super bullish on is actually DAOs. I'm into a couple of DAOs now. I just found that it's fascinating to be working with folks that have no idea who they are.

And I think this is actually an issue that a lot of the DAO tools are trying to solve. It's kind of giving them the right direction when they onboard a new DAO. And we are partnering with a couple of the DAO tools right now. I can't really disclose them yet, but we are trying to contextualize those users. Let's say if you are just getting into a new DAO and you just see a bunch of zero Xs, you have

there are thousands, like 500 of them, and you have no idea who they are and what are they to do with you being here. So with CyberConnect, you can easily...

contextualize those like, oh, these are my friends, or these are the folks that have worked in another DAO before, or these are the folks I've voted in a snapshot voting before. And oh, I even know that they voted the same as I did in another proposal. So these are really interesting and really powerful attributes that we can give to different users in those DAOs. And that's where we're working on with a lot of the DAO toolings.

On the DeFi side as well, we have the project called Rossi Finance. We've been working with them for a while now. They're doing collateralized lending based on a credit score that they built. And they are using the social graph data that we provide to kind of come up with a more holistic view of how are people

Like what are the folks' credit score based on how are they connected with each other? And that's actually a very interesting use case that's brought up by the community or by the developers instead of by us. So yeah, we're just excited to see more use cases being thought up by all those developers out there.

Absolutely. I think what you're saying, referring to DAO, is actually super exciting because I think these days I've been seeing a few usually facing problems.

quote unquote uh web3 linkedin like what they're trying to do is essentially trying to show people like okay which doubts you're involved and then like what are some of the things that you've done there so and then who you worked with and then you know these people can give each other like endorsement and whatnot but like you know you can't rely on a random endorsement it has to actually um

like show certain uh proves that these people are connected um in what way so i think being able to integrate with some of the dow tools will definitely help that verification process so i'm also super bullish on that i appreciate you sharing this

All right. I think this is an awesome conversation. It helps us learn a lot about the technical architecture and the problems that you guys are trying to solve and how you differentiate from some of the other existing approach on the market. Thank you so much for coming on, Ryan. It was a great pleasure.

Thanks, Michael, for having me. Yeah, it's really fun to discuss what we're building. And I'm just very excited to share more, not only in podcasts, but share more in our app. So developers can actually play around with it. Thanks so much. Thank you.