Dylan Zhang: Crypto Native AI Models Behind the Future of AI Agents
Episode publishedPodcast: The Delphi PodcastThe Delphi Podcast
Transcript preview
Preview only
You're now plugged in to the Delphi podcast. Hello, everyone. Thank you for joining and listening on the Delphi podcast. Today, I'm joined by Dylan Jung from pond AI. And we'll start off with just learning a bit about him and and love to understand, yeah, how you got into crypto and and particularly crypto AI and and why the intersection really excites you. Sure.
So, nice to talk to you, Yan, today. So, I'm Dylan. I'm one of the cofounders of, of Pond. So how did I get into crypto is, it's kinda long story. So I heard, about Bitcoin back in 2017. So by the time, there was a there was a Harvard PhD, you know, was trading tokens and, tried to, you know, get me over to trade tokens with them. And so at the time, I I didn't really believe in the the the story of Bitcoin because when I looked it up, all the all the all the news that I I saw, they are really about, you know, scam of of of Bitcoins, and then it's nothing really positive at the time. But later
it. And before kind of going into the the specifics, I think it'd be useful to, you know, set the stage a bit and and just provide a brief elevator pitch of sorts on kind of what the world of DeFi looks like with Pond. Yeah.
So, apparently, DeFi, I think everything was DeFi now is kind of fixed. You know, no matter it's going to be the fixed structure or it's going to be, you know, the mechanism itself, it doesn't have any dynamic parts. But what machine learning and deep learning can bring to DeFi is definitely a more dynamic, structure and a mechanism that allows people, especially developers and project owners to adjust, some of the, you know, fixed parts of, within the DeFi to be more and more dynamic. And, also, I heard from, one of the founders of a very successful DeFi protocol. So what do you feel about AI
Yeah. No. That's very helpful. So so diving a bit in, you know, you you guys are building, graph known up neural networks, GNNs instead of LLMs, large language models. So I think it'd be useful to kind of describe the difference between the two and and with a bit of a focus on why, the former works well for for blockchain? Yeah.
So AI model, the the time when you wanna develop a AI model, you need to find the best possible model that's specifically designed for this type of data. So in web 2, the majority of the data actually is language data. So, you know, the the the model or the foundation model has to be a language model. But in web 3, the most the crypto native or the most of web 3 native data is 100% on chain data. So on chain data, first of all, is not a type of language data. It's, actually a graph data. So what is a graph what what what is graph data? Actually, it is as long as it is a data structure, it can
And and just thinking about kind of the the journey that you guys are taking from building a single GNN model to building a decentralized AI model layer, can you kinda touch on on the the rationale there and and what what that's gonna look like over time?
Yeah. So, so initially, we started developing a large GN, but we gradually realized, actually, because ecosystem is is kinda early that, comparing with some developing a large model, it will make more sense for us to to develop, different smaller models and smaller use cases driven models. So this is something I would be working on. So just because, like I said, in the in the very beginning, all the use cases, no matter using social, in DeFi, MEV, and, and all that, they are actually all in the same graph network, which means that, we can develop a bunch of models. And these models will be, yo
it. Got it. No. That that
Keep exploring with a free account.
A free account preserves the short preview. Upgrade to Pro for the full transcript and episode chat.