DeFi, Ethereum and Cryptocurrency with Kendrick Tan

Matt: 0:00

Hi, and welcome to the router, the official podcast of the UQ computing society, where we explore the human side of tech. I'm your host, Matt and we are back after exams with a brand new episode. Today we're having a chat to Kendrick Tan, also known as@kendrick on slack. He has worked in the blockchain space for many years and currently works at DFX in decentralized finance, where he is the chief technical officer. He's here to give us a 20 minute rundown on blockchain, Ethereum, decentralized finance, and more as well as addressing some of the concerns regarding the past present and future of cryptocurrencies. All right. Welcome Kendrick to the show. How are you today?

Kendrick: 0:47

I'm good. How are you?

Matt: 0:49

Yeah. I'm all right. Um, before I begin, it would be good if you could introduce yourself to anyone who doesn't know about you, who you are, what's your background, where you work that kind of thing.

Kendrick: 1:00

Yeah, so I'm Kendrick. So I graduated from QT and 2019 with a computer science degree. Uh, the reason I got a computer science degree because my mom wanted me to get a degree and I was like, probably quickest way to get Out of uni. And so I got one. So after graduating, I've been working full time in the cryptocurrency space. Um, I originally wanted to get kind of FAANG job, but after dipping my toes in this ecosystem, I decided there'll be such a pity if I didn't kind of g rab t he opportunity, to go in such an emerging field. So since 2 019 I've been working with, u h, various companies, u h, one of them being, u h, the Ethereum foundation. So I contracted with the Ethereum foundation for about half a year. And we did a lot of work on z ero knowledge proofs, which is like basically m oon math o f people l ike to call it, but I was the one reading the papers and applying the math rather than coming u p with the m ath. S o that a very big difference. U m, in that it was very fun. Like I l earned a lot about the processes of how, like these cryptographers came u p, came up to these conclusions and how to actually use them and how, like, it's really interesting how y ou can like use these properties in, u h, in, in, in if y ou're specifically, anyway, I was i n l ove with a couple of t he layer two scaling solutions a nd f iguring, w hich i s like, u h, products or applications o r protocols t hat a re dedicated to create help make your scales like higher, u h, h igh T PS or higher transactions, the second. Uh, but eventually, um, during 2020, um, I found out about DeFi and the decentralized finance and after dipping my toes into that, I decided I wanted to go full time into that. And so I worked for a company called UMA. Unfortunately the time zones didn't work out. And so I left and now I'm currently working at DFX, which I'm also the CTO.

Matt: 3:03

Oh, very nice. So, uh, you kind of started working, um, in Ethereum or generally, and then moved into, uh, decentralized finance.

Kendrick: 3:13

Yeah. Yes. So it was mostly, I was interested in the cryptography space, but then suddenly this wave of innovation came from, like on the finance side, I didn't know anything about finance back then. I was still at like a geek, uh, like more like a technical guy, but just the amount of like smart people in the space kind of convinced me that maybe I should check it out. And the minute I did, I was like there's no turning back. Yeah.

Matt: 3:42

That sounds super fun. Um, I guess to begin, um, because personally, I don't know too much about, uh, this area of work. Uh, could you begin by trying to give like a, I dunno, explain like I'm five kind of introduction to what theorem is, how it differs from, you know, Bitcoin or other cryptocurrencies, um, and maybe also a bit of an intro into what decentralized finance is. Yeah.

Kendrick: 4:08

Yeah. So Bitcoin at its very core is just a ledger, which all the participants in a network will agree upon the state of the ledger. Um, and in Bitcoin's case, there'll be how much Bitcoin everyone owns. So for example, uh, if this, like, if it is like 10 participants eventually, and let's say, I'm sending you 10 Bitcoins, eventually everyone, like you will broadcast a signal telling everyone in the network saying, Hey, I sent you 10 Bitcoins, eventually everyone's ledger. What kind of agreed upon that? I've sent, you 10 Bitcoins, and so they'll deduct 10 Bitcoins from my account and add the coins to your account. So that's like a very at its highest level, there's like a more there's like more complicated stuff, such as peer to peer networking, routing and consensus theory. But, um, let's just ignore that focus on like the, the core idea, which is like a decentralized ledger, which will form eventual consistency or like a distributed database, if you don't want to put it that way. Okay. So that's the current state A to state B. I want to kind of emphasize that state transition because it's very important in Bitcoin's case it's very, uh, very application specific. It just tells you how many Bitcoins is in transferred from A to B or like the, the, the delta, or like the state of how many Bitcoins you own. Ethereum on the other hand can be seen as like a hack of Bitcoin. So instead of having a very rigid system, a very rigid state transition system state A to state B. You any programmer can create an arbitrary logic to kind of indicate how the state transitions should be from state to state. So if I want to like say deploy a new token on Ethereum, and I'll basically create a new token and deploy it onto the ledger, like there's a good database, where it stores, like the byte code of your application. And let's say state A doesn't have the tokens. Once you deploy the contract, State B will state that, Hey, in this, in this, uh, storage slot, I like to call it, there is now a new token with this bytecode and now I can execute arbitrary code against as, or execute like some kind of a code on Ethereum against that deployed by the state to change more states. So it's more like a, a generalized state transition machine. And also keep in mind that Ethereum, runs in virtual machine, just like the JVM, but they call it the EVM, it's your emotion machine, which is also occurring complete. So now you have like generalized s, so that's like a complete state machine. Um, and Bitcoin's like more like, uh, application specific ones. So, uh, in that sense I would, that's what I like to call it like a hackers Bitcoin. It's a lot more exciting, uh, because you can build any, anything you can dream of. Well, right now, right now it's mainly finance. Um, because if Ethereum and Bitcoin are speculative assets. So, uh, like the main applications built on. And right now I financially related, which brings us into DeFi. Um, so DeFi is this movement from kind of like the secret world the wall street and like the centrally controlled, like Mon like sovereign money into this, like cyber punkish, uh, like democratized, uh, money platform, if you will, which is which essentially what DeFi is. I remember three years ago when I started out in this space, there wasn't a stable point, um, when I started up, but now there's like so many, so many financial infrastructure, it's impossible to keep up in the space.

Matt: 8:02

I mean, so there's so many like different words that kind of get thrown around when it comes to like cryptocurrency and this sort of thing. So it's good to hear about, um, a few of those things. Are there any other kind of more specialized things that you work in, maybe give a brief intro into, you mentioned zero-knowledge proofs before. What is that like? Like what, what is it like working in that space? Um, I it's, it's a cryptographic thing, right. So I guess is like a lot of math involved, that kind of thing. Yeah.

Kendrick: 8:29

Oh, keep in mind. I was the one who was reading the papers

Matt: 8:31

Yeah. Yeah. That's all good.

Kendrick: 8:33

Yeah. Yeah. But essentially what I did my work in the zero knowledge proof space is, how do we, it was like a voting system. It's like, how do I, how do I vote for someone without revealing my vote? And in fact, if you're interested in it, I think, uh, EY also released something to do with like transacting on the blockchain with zero knowledge proofs. So how do I create a receipt to accompany that I'm dealing, uh, that I have like some kind of relationship to kind of prove that I sent payment without actually revealing the payment, because keep in mind, these payments are publicly, uh, on a public ledger. And if your competitor knows, like who you're paying and how much you're paying for what amount of like, uh, supply and it's like kind of bad news, because then they can kind of use this information to their advantage. Um, I forgot what it's called, but if you look up EY zero knowledge proofs ethereum it should come up.

Matt: 9:25

Yeah, yeah. I will. I'll link that in, in the description of the podcast. Um, yeah, I'll find it. Yeah.

Kendrick: 9:32

So essentially it's like, how do I prove I have something without necessarily revealing that particular something, um, is the whole idea of like so much proves there's also like other rabbit holes that you want to dive deep into and it's consensus theory of which I'm not, I'm not I'm, I haven't touched, touched at all, but there's also something called MEV or minor extracted value. Um, when I, when I mentioned like the state transitions, the all their state transition can be exploited. It's almost like an order on an exchange like high-frequency exchange. If you were to suddenly jump the queue, you were like profit and loss. And it's the same with like mine extract value because mine is kind of dictate the order of transactions that occur. And how do you like game the system? So what order of the state transition costs will your maximum value is quite a, quite a new field. It's only come around in the last four months or five months, but it's slowly gaining momentum.

: 10:33

Huh. Sounds interesting. Um, yeah, I'm new to all this stuff. Um, I guess another question that people ask, I think, um, when they hear about, uh, Ethereum cryptocurrencies and this kind of thing, um, at least from what I've heard, it's, it's kind of, some people think of it as something that you tried kind of treat it like a stock or like a, you know, a currency you're like just buying into it. Um, and you just hope to see the price rise when you buy, you know, try to make some money off of it. Um, but I guess some people like, like me, I'm wondering, like there are some of these benefits to, you know, decentralization, um, and I guess, you know, getting rid of that, no mysterious wall street hole that the money seems to go into. Um, but like what, what other, like, does it solve any other kind of problems? Like what, what other things can you apply this theory of, um, idea to, in terms of like technology,

Kendrick: 11:33

That's definitely the main narrative. So Bitcoin that's like a decentralized store value. That's like not... To answer your question, sorry. That's one of the main, that's one of the main use case just recently, because Bitcoin has been a transfer of value, like a, like a store value that can electronically transfer, like Bitcoin is money as a data structure, which that concept is, in my opinion, incredibly powerful. I guess we don't really, we don't really experience it here, like in Australia because it's a lot of that we have a stable political system or currency is relatively strong. There isn't much need to kind of call, um, outside of like sovereign money. But if you live in say Zimbabwe or Venezuela or Turkey, or recently Turkey where there's like a dictatorship happening and, um, there's like, it's political instability, that's a real, real need to have access these kinds of instruments because you want to be able to leave the country with your wealth intact. Like it will be, I mean, if you can escape the country, but start from zero, that wouldn't be ideal. Right. Um, yeah, but to answer your question, that's like the main use cases and I can perceive certain other use cases such as voting, uh, on blockchain. But that kind of brings up a lot of, um, other, um, potential issues with that. Yeah.

Matt: 13:04

Yeah. Like I guess, um, confidentiality and, you know, making sure that your votes are secret and I've heard about some, uh, approaches to this kind of, um, you know, making elections better, but, um, it seems like the best way forward is to keep developing this technology, um, seeing if we can get around those issues and it's better than at least it sounds like there's more potential in these blockchain approaches than the current system of digital voting, which is just some dodgy windows XP machine in a, in a booth that you just type your votes into and hope that it, you know, this is correct. And

Kendrick: 13:46

Yeah, because with voting, like how do you guarantee that the hardware you're running is going to execute what it says on the screen? We don't even know that, like it's a lot, a lot of issues about like from the hardware level and and if the software is perfect, how you, you can't guarantee stuff in hardware that was simply just too many potential, uh, backdoors. Mm.

Matt: 14:11

Um, I'm gonna pivot over to something a little bit different. Uh, so you're talking about, I guess the use of Bitcoin and crypto currency in general, like it's, it's kind of going up and I've noticed, like, there's been news stories about how, like, you know, in El Salvador and countries like that with us starting to, you know, accept Bitcoin as like legal tender. Um, so Bitcoin is becoming more and more popular and people are more and more interested in, uh, possessing Bitcoin mining Bitcoin, obtaining Bitcoin. Um, and that's kind of led to this discussion about the environmental impacts of cryptocurrencies, because obviously currencies in general have their own environmental impacts, you know, generated by all this production of, you know, minting and things like that. But it's a bit unique and it's a bit different because now it's kind of intersecting with the tech world a little bit and where, you know, there's been this rise of people purchasing GPU's to mine, Bitcoin and, you know, different methods of verifying ledgers and things like that has led to the increase of like computational use of power. Um, so I guess there's like a few, few things been brought up about that. Um, I guess, could you give a bit of background from your like cryptocurrency perspective about what, what, what are the issues about it? How, what is the issues of Bitcoin or cryptocurrency and in terms of that, how is crypto trying to tackle these issues?

Kendrick: 15:46

Um, well, first of all, I completely agree with your statement that it's consuming too much electricity. Um, however, I do, I would like to point out that a lot of the, like, like the criticisms surrounding the electricity consumption by the media, it's a little bit misguided in some sense, um, cause like Bitcoin and cryptocurrency mining in general have this very, a unique property, like their power. They consume power in a very geographically independent manner. So we're used to consuming power. That's kind of close to us. So for example, Brisbane, we used to pack consuming power. That's kind of generated in Queensland. It'd be absurd to think that we will consume power generate in some way or Singapore or somewhere in New Zealand or somewhere in Iceland, China, but that's exactly what a Bitcoin story it's actually consuming electricity in, in areas that are kind of, uh, in areas that have an overabundance of power. So for example, like the main, uh, mining published in China and Sichuan and Inner Mongolia far away from the city center as well, um, in Iceland where like the city, like actually cryptocurrency, mining consumes more electricity than I think certain parts of Iceland. The point being that like these electricity consumption is actually being consumed and also keep in mind that energy, energy is really hard. Like electricity in particular is very hard and very expensive to store. So the minute you produce it, it kind of needs to be consumed immediately. Otherwise it would be wasted and electricity consumption has to like kind of go with like the peaks and troughs of our consumption. So for example, in the afternoon might consume more electricity, but like at night as Bitcoin's like constant. Uh, so I guess what I'm trying to articulate is that a lot of the times like these energy consumption by cryptocurrency and Bitcoin mining in particular, Bitcoin mining particular would otherwise be electricity, that's kind of wasted. So you kind of monetize and, uh, and, and kind of like a resource that would otherwise go to waste. Um, however, I do agree that there's contributing to a lot of greenhouse gases. Um, and, uh, there are some ways kind of, there are plans in Ethereum, not for the coin, the coin kind of staying on this whole proof of work thing where you use like computational resources to kind of validate the block state. Whereas if you're in this kind of moving on to this, uh, this, this system of calling it, using the system called proof of stake where you capital to verify, um, the validate like the block state, rather than using some sort of, uh, some are they're using a computational power, uh, which would in theory, significantly reduce, uh, the energy consumption.

Matt: 18:55

So I guess there's, there's, there's more approaches being developed now to try to get around this issue, I guess. That's good. Um, do, do you think that as a result of this, um, people might move away from Bitcoin or do you think that Bitcoin is kind of going to have the same value, same role that it has now,

Kendrick: 19:15

Uh, it has past the litmus test, um, too many times, uh, it's, it's, it's going like a lot of times, like these ledgers on like these cryptocurrencies are kind of like, instead of, I kind of like a social consensus amongst people, I like to think that, um, like there's now a social consensus amongst all the hedge funds and all like the big finance finance companies that Bitcoin is here to stay and they're kind of looking more deeply into it. And so I do not think that, and Bitcoin has a dominant, Bitcoin is the dominant asset that's looked at as a stored value kind of hedge against the recent, uh, US supposedly US inflation. So I do not think that people particularly, uh, I don't really think anyone would really move away actually, because of these energy consumption issues.

Matt: 20:16

It's good to keep in mind, you know, um, the impacts of that kind of thing, and

Kendrick: 20:21

Yeah, yeah, totally agree that it can be, can be better, but yeah, you know, if you, I mean, nationally, and also if you think that, uh, Bitcoin as a settlement layer is worth$0 then naturally, or it really depends on how much you value Bitcoin as a settlement layer, as you mentioned, currencies creation and minting of currency do generate kind of like green and greenhouse and contribute global warming. So if you think that Bitcoin as the settlement layer is$0 and naturally all the energy, like no matter how much, how middle energy consumes it still

Matt: 20:58

Is waste. Yeah. Yeah. All right. Um, I didn't really have any other questions to ask. One thing I wanted to ask is if, if some listen up someone listening would like to find out more about these kinds of technologies, uh, where do you think they should go? Like, do you have any kind of resources, things to read things to watch? I don't know, um, recommendations,

Kendrick: 21:23

Uh, personally I found that creating a Twitter account and following a key set of thought leaders is probably the best way to get forward because the information just there's just too much information out there. It's too sparse. No, one's, it's no real, uh, there's no centralized repository where you can kind of go through and find out what to do. Like it's just too much happening every day. Like every day, that's like a new concept, opening protocol being released. There's like a new paper that you should read. And so I would highly recommend creating a Twitter account and following a few key. He key influencers, if you will. Um, that's like how I stay updated, otherwise it's just impossible.

Matt: 22:08

Yeah. So it sounds a lot like ML has been for the past few years where it's just a constant stream of information and the only way to keep up is Twitter or like arXiv and just it's crazy.

Kendrick: 22:21

Yeah. Yeah. Definitely some similarities then. Yeah. Yeah.

Matt: 22:26

All right. Um, that's it for me? Uh, thanks so much for joining us on this, uh, episode. Um, I will make sure to link everything that's been mentioned in the, in the description and, uh, yeah. Uh, thanks for coming along.

Kendrick: 22:42

Well, thank you. Thanks for having me. Thanks.

Matt: 22:44

All right. That's all we have time for today. Thanks for joining us as usual. Our next episode will be out in a fortnight, but until then come join us on our slack community at slack.uqcs.org.