E116: CoFounder of Sonic Labs: Why He Gave Away Billions in Crypto (and now completely regrets it)

28:49Everyone's north star was immutability. You wanted to deploy something that never needed to be touched again. And up until I kind of took a step back from the industry around 2020, 2021, that was always my ethos. Like not a single contract I had deployed during that time had any kind of upgradability. So the contracts were of such a nature that I could not modify them even if I wanted to. And that's different because back then everyone was building software for everyone else to build off of or use as well. Nowadays, we're building crypto companies. And there is a distinction there. Like the stuff I'm going to be launching going forward definitely will be under sort of my company banner with equity for me, with tokens for me, because they're not meant to be immutable primitives.

29:50They're meant to be products of a company. Is that better or is that worse? Or it is just what it is? There's value in both. Look, also the stuff we're doing now, it was practically illegal to launch a crypto company back then. You had no path to successfully do it. Basically, if you did do it, you would have gotten hunted. I mean, I had, you know, double-digit emails from the SEC about stuff I had to provide them and turn over and things. And I mean, I'm a South African living in South Africa, hasn't been to the US, like didn't sell a token, didn't provide a service. Like, guys, how am I in any way your jurisdiction?

30:40But still, you know, that doesn't stop all of these different agencies around the world from hunting you because of that stuff. And, you know, let's face it. I mean, if it wasn't for the more positive administration here currently, I wouldn't be risking launching this new stuff. The only reason I am risking it now is because there is finally a path forward. We're going to talk about that, the SEC and what you're doing now. Obviously, you made people a ton of money and created billions of values in just a few weeks. You became a god. Some called you Jesus. That was, people shouldn't do that. But there are many problems that come with this.

31:16What's the ugly side of online celebrity? Um, everything's expectation management, right? Um, I think that's what I struggled with most towards the end because you, the more, I don't want to use the word famous, but for practical reasons, let's use that, that the more niche famous, right? Non-traditional famous, the more niche famous you get, the less your audience knows about you. so you know early days everyone that interacted with my stuff knew my process so they understood that I did a lot of my work on chain because back then we didn't have the tooling we have now so a lot of things you had to do on chain and most of my apps were very composite so they involved you know so many different apps interacting at the same time that you could almost only practically do it on-chain.

32:12Like you couldn't do it in a replicated environment because now you try and testnet A has those two apps, but not those two. Testnet B has those two, but not those two. And so, because everyone was kind of siloed and choosing their own stuff. So, so that made it really hard. But the original participants, they all knew how I worked. The biggest wake-up call for me was with Eminence. So I had launched a theoretical framework of how gaming NFTs could work because I didn't like what I was seeing that was happening with NFTs. The tech is very powerful. How it was being used was garbage. It still is, mostly.

32:53Sorry, Bengu.

32:58But part of my process is doing things on-chain. So me and the developer I was working with, Anton, we have an iterative process. I deploy my contracts. He does his front ends. He catches bugs. I redeploy. We keep kind of back and forth thing like this. And as is natural with a process, you know, we work for 10, 12 hours a day, and then we have our normal lives. At that point, when I stopped deploying, people had assumed that that last one I had deployed, oh, cool, he's done. He's launching this thing now. and they started sending money into that a lot more than they should have 16 million dollars needless to say was exploited makali eight was returned and the other eight is still sitting in the contract whoever acted if you're watching this by the way i think it's enough time you can return the rest of those people now um they've never used it which is weird um but that was a big wake-up call to me that you know now there was a very large active audience that were not aware of how I had previously done stuff.

33:59And so everything just became about expectation management. But that in itself was exhausting because now I had to spend so much time with disclaimers and preface and blah, blah, blah. But I also think it's just, there's also just a learning process that you as an individual have to do. Because when I started, my goal was 100 % happy users. So if one person had said something negative, I would private message them. I'd get them on Telegram. I'd talk to them about their problem. I'd try and solve it. I'd try and see where they're coming from. Maybe I can do something better, et cetera. Fast forward a little bit, and all of a sudden, that's completely impossible because there's a demographic that you will never appease.

34:43They exist to hate you. So then my 100 % went to 51%. I was like, as long as I keep 51 % happy, I'm, you know, like standard proof of work, consensus-based. 51, I'm good. and that again worked for a while. I could appease, et cetera, but then as that fame got bigger again, started realizing 51 % is impossible. It's not going to happen because now all of a sudden there are people who are financially incentivized to hate you. So now I'm on the Pareto principle. I try and keep 20 % happy that provide the most value to the stuff I'm doing and probably as I continue along, that number will keep decreasing, but that's something that I needed to internalize and make peace with.

35:24How did you do that? Depression, hardship, like self-destructive tendencies sometimes. There's no easy way that you're going to go through that process. Like you have to rewire your brain. And unfortunately, thanks to kind of, you know, survival of the fittest, the only way we rewire our brain is through hardship. Did you do any therapy or you just did everything on your own? No, I should have probably, if I'm being honest, because I went through a really bad cycle after Eminence, but luckily I got through it. You mentioned these expectations, right? I also read about this project associated with called Solidly.

35:59Yeah, well, not associated. That was my code. Yeah. So people aped in$2.3 billion in like days or weeks. And then when you announced that you were leaving because there was a bug in the code, the total value locked by like 90%. Well, no. So I wasn't leaving because there's a bug and there wasn't bugs in the code. the bug people are referring to was a literal design feature. So, I mean, that code, Solidly Code, and VE33 specifically, that went on to be the code used by almost every single prominent Dex on every single new chain. It was Metis, it was Velodrome, Aerodrome, Ramses, Faro, Shadow, like all of those guys is that code.

36:46The difference, so I just want to address the bug statement first. The difference is, again, at that time, I couldn't launch companies. So that code was 100 % immutable. Couldn't be changed. There was no proxy upgradable. There was no multi-sig. There was nothing. It was exactly as it was. But now one of the things is the whole flywheel works on you vote emissions onto a pool and then you get the fees for that pool. Now, something that someone could have always done, and I wrote this in my Medium articles, and you can go back to them if you want to double check me on those, is I had written about the theory that someone could create a bricked pool.

37:30So they can create a pool that no one else can participate in, that only they can farm, because they just create two tokens that only they can mint. They can then add that pool. Then they can buy a lot of the token, so significant enough that they can impact the emissions. Then they have to lock that for four years, and then they could send emissions that way. Now, that is a way that they could game the system. I had believed no one would do that, because you'd be shooting yourself in the foot. Because the second you do that, other participants are going to withdraw out of the ecosystem. And you are going to be left with, sure, you can now vote on this own pool, but you're going to devalue anyone else's participation, which means everyone else is going to withdraw out of the system.

38:15Unironically, and I believe that's against human nature, you know, who would spend millions of dollars to acquire something that they then devalue by the very action that they are doing. And unironically, two different people did that day one within hours of launch. The one guy spent something like$13 million to do that, which became worthless overnight, exactly as the original theory. Why would they do that? I had believed that all participants in a system are rational actors. That assumption was false. That was also a good lesson for me because nowadays I don't build systems that assume rational actors.

38:57but so yeah so so the bug because then a lot of the forks for whatever reason whenever forks fork solidly the the first thing they do is talk extensively about how shit solidly was even though it's the entire premise of their company and model but anyway the the bug everyone is referring to is that decentralization so the fix that all of these teams did is they decide which pools get to be voted on. And they control that via their multi-sig or their team controls or whatever. So they removed the decentralized I had put in there and they had added a centralized component. So just to comment on that.

39:37Now, the other thing, this is also expectation management, because, and you can go check these on all of my Medium articles about Solidly, I always stress that this isn't a mutable primitive. The second it launches, I cannot change or update it in any way. So, you know, my stepping away from the ecosystem was seen as a reaction to that, but the second I launched it, there was nothing I could do about it anymore. And I had made it clear that the second it launches, I am no longer involved in it. The timing was probably a little bit annoying because like a few days after that, the SEC She ran up a lot of stuff, so we were like, okay, it's not worth the effort anymore.

40:20Let's just take a step back. But that was not the sequence of events. Yeah, you say go read my Medium articles, but most people don't read, right? They're like, Andre's new project, I'm just going to A-pin. Correct, like I said, that's part of that expectation management process I learned. And I mean, look, against my own statement, like I had known that sort of do your own research is basically a meme at that point even though I still again sort of believed people would do that but I mean it's just like your little NFA there on the side right like it's a meme exactly yeah yeah absolutely so so I mean if anything whenever you see NFA they're gonna be like this is financial advice thanks bro just means I don't have to pay you for it that's the only thing yeah so So, I mean, another thing I had to learn was that our audience had drastically started to change over time.

41:22You know, when I first got involved in DeFi, the participants were all highly technical. By the time I had left, it was probably 1 % technical and 99 % of people that just heard about something on Reddit or 4chan and then just somehow sent their life savings to it, which is still actions that I will never understand. But I also needed to recognize that the space had changed so drastically that it's an entirely different demographic now. How does it feel to have that much influence over people's money? Like, did you manage to dissociate with it? Or because it could feel like a lot of pressure. I've never really thought about that.

42:04To me, I'd say I'm the inverse. I don't actually think I have that influence. Like, I'm, even with Solidly, I thought, like, my, like, 50K or whatever I put in there would be the only money. With Yearn, I thought the Phantom money was going to be the only money. Even with the new exchange I'm launching now, I'm assuming it's just going to be my cash. But you say I put 50K in Solidly, but then you have$2.3 billion coming like that. you must feel something like stress or maybe oh that's awesome or anxiety yeah yeah no stress and anxiety are there but the stress and anxiety isn't because of the influence of the money it's because of the responsibility of the money um that's just purely like because when you know it's let's say 10k and something happens i can work with those people to try and or recoup that or figure something out.

43:05You know, if it's 2 billion, there's nothing in this world I can do to try and help those people. That's done.

43:14So that definitely exponentially increases the worry. But for example, I know it's a meme about test and prod and all of that stuff and that I didn't do audits. All of my projects had multiple audits from multiple teams. You can talk with the audit companies. Like I'm on first name basis with most of those guys. Huge amounts of test cases, this huge amount of tasting. I mean, the only stuff I had that was ever exploited was imminence that was literally actively being developed where people put all of their money in there. So I always took every reasonable precaution to protect people's money, but it did get easier over time.

43:54That's also something that your mind kind of gets used to because I remember with Yearn, when that went up to, I think, like 700 million, which I was not expecting at all and I couldn't sleep. I would be awake almost every five minutes checking to make sure everything's okay, double checking things. But by the time it got to solidly, I have to admit, I wasn't that stressed because I had known that there was no opportunity for something to go wrong there.

44:24It gets easier over time. You get used to the quantum. You get used to all the tolerance to stress. arises with time, right? Obviously, I mean, with experiences. I read that you didn't make any money on these things. That is true. Well, I mean, I gave away 100 % of the tokens, so there's no way to make money. Probably lost a little bit of money because I actually bought quite a bit of salt, but still have that, so whatever. How did you feel? I'm not making any money. I'm actually losing money. And I have all this stress and anxiety, right? It's not, I'm building something amazing that people buy or invest in and they're generating a lot of money for people.

45:04It's more at least the stress and anxiety and responsibility should be kind of worth it. Look, again, each project, definitely I had a different experience because like with Yearn, more so the participants knew that by me giving away 100%, I was giving up responsibility. So I didn't have that much. But again, it was a technical audience. They understood what was going on. By the time we got to Solidly, it was such a non-technical audience that their expectations of me were a lot higher. That, oh, but I launched this. I'm supposed to stick with it. Blah, blah, blah. Where am I? Blah, blah, blah.

45:51So again, changed a lot over time, dependent on the audience you were working with. but yeah I mean that still sucks and that's why I'm not going to do it again you know because you by the time it got too solidly I had none of the incentives or upside but I had all of the responsibility so it's just like this this is stupid which yeah like I said going forward that's definitely not going to be the case I'm going to make sure that incentives are aligned you mentioned before that one day you received emails or letter from the SEC yeah frequently frequently what did the first letter say it's always just a request for information under the guise of you know don't share this or it becomes an official investigation and as long as it you don't publicize or speak to anyone about this then then you know this is just an inquiry it starts friendly enough it quickly wraps up though because like the the more it's very much like my experience with tax agencies as well, where they're kind of just like, oh, we know everything, but we're going to not tell you, and you need to now give us everything, and then we're going to compare.

47:08And if it's wrong, you're going to end up in jail. Very similar experience. As it progressed, because there was probably, like I say, 10 plus sessions of this. And the amount of detail they request is comprehensive to the point where it was taking up originally days, then weeks, then months of my life to just collect the information they're asking for. And often it was information that I didn't have, so I had to go get alternative sources to provide it. It was a ton of effort. And you kind of start under this guise of, okay, cool, if I answer these, then we're good. But you realize that's just not the case.

47:55And they just try and look for different angles to drill down. Because when it started, it was all around the Yearn token Wi-Fi. And what investors did I have? What seed was there? Who did I sell it to? How much did I have? And a lot of that stuff is also proving a negative, which is really hard. Because now I have to prove that I didn't have any tokens. And it's like, how do I actually prove that? So basically they're asking you questions until you make a mistake. Correct. But I mean, like, again, I luckily had Brown Rutnik and Gabe from Metal X helping me as attorneys. And they'd been through that journey a few times.

48:33So they were great in guiding me on how to do that stuff. But we also saw, you know, the pattern evolve. So like when they started realizing, look, the token isn't an angle of attack, they started looking at the product as an angle of attack. because then they started, you know, sort of making the claim that the Yearn Vaults themselves are securities because a user is giving it money with the expectation of return. So then they started framing that as a securitized product. So, you know, it just kept scope creeping. And it really only stopped when, you know, the next election started coming up.

49:08Then it went quiet. Because you were based before, you said, you're based in South Africa, you're South African. What's the worst thing that could have happened? What can they do concretely? I asked them and they said that extradition is on the table if I don't answer. I decided not to taste that theory.

49:29And I mean, like, I'm a white guy in South Africa, South Africa is not going to stand up for me for one second. They will throw me to America for a banana.

49:42You ended up leaving crypto? I ended up leaving Twitter, which a lot of people assume was leaving crypto. I guess I did kind of leave crypto as well. I actually tried going back to my TradFi job. I also think once you've been in crypto, you can't get out. So I was still doing my part-time stuff. I was still doing some stuff with Phantom and Sonic, and I was still helping some of the Yearn guys, etc. And I could never stop thinking about the stuff, the problems I want to solve, the AMM, the lending market, the options, the derivatives, so much stuff that still needs to be solved in this industry that isn't solved.

50:24Yeah, that probably lasted maybe two, three months. And then I went back from part-time crypto to full-time crypto, but still lifted Twitter and all of that stuff off because that became so much noise and toxicity that the R &R just wasn't there. So there's no point to take that risk. I have no more faith in humans. I don't trust anyone anymore. I don't think anyone behaves in anyone's best interest unless there is a transactional incentive. Yeah. Yeah. That's what you said. Why? I still stand by. Well, like I said, Solidly was the perfect example where I thought that people were rational actors.

51:05And over the last few years, I just realized that no one is rational. Like everyone only behaves in their self-interest in the given moment. So one of my core beliefs up until probably like 2022 still was that all humans are inherently good and sometimes they do evil actions in bad situations.

51:31Don't think that at all anymore. I think all humans are inherently evil. That's how I like to think. And only act good if there is self-interest. Wow. That's a big change. That's a complete 180. How do you live with that thought? Oh, you just get used to it, right? It just becomes your new core principle. And my life has been significantly easier since I've adapted that philosophy. Even in your personal life? Everything. Is it the same? Everything. Everything, everything, everything. Like, when I believed... Because, for example, like, I... Again, not money motivated, but having been encrypted this long, I had opportunities to make a bit of cash.

52:12But my living expenses were completely insignificant in South Africa. So 99 % of that money I gave away to friends or families or when I met someone that had a cool idea, I was like, oh, here you go. Go do your startup or whatever. In a very nonchalant way. So never paper, never contractual agreements, never you're going to quit pro quo me, whatever. and like clockwork, 100 % of those things I did, I ever turned up into them hating me because I stopped giving them more or if they made it successful, it was all them and they didn't need my help and they refused to give anything back or sometimes just straight up scams.

52:54They just take the money and they never communicate again but 100 % hit rate. 100 %? 100 % hit rate on people who would behave that way. So now I just don't do that. Wow. And life's a lot easier. Do you still angel invest? No, I don't do anything anymore. That's so sad. That's so sad.

53:17So it's not only in crypto that people are bad. I was scammed more in TradFi than in crypto, unironically. Like in TradFi, I'd lost almost everything. In crypto, I'd never lost a cent, which is a little bit of an irony in there. You're back today and we're here in New York. Yeah. What made you come back?

53:47That's actually a really good question. When did I activate my stuff again? It was somewhere around when we were planning the launch of Sonic,

54:01which I do think is currently the best blockchain tech out there. And it was just an opportunity to amplify the stuff again, so I activated the Twitter. I mean, I'm assuming we're using that as the metric of comeback, is basically switching the Twitter on again. And it was really just to, you know, amplify some of the news and share that stuff. And then in the back end, you know, I'd always kind of still been working on some of my new primitives. There's a bunch of stuff I never launched. Insurance stuff, AMM stuff, derivative stuff. and then with the new administration there's now an opportunity to launch that again so which that's just coincidence and luck

54:45so yeah just increased participation from there but the original one was literally just activating the Twitter again to retweet Sonic stuff What is Sonic if you had to explain it to your mom? Provided your mom is not as technical as you because you said before my mom wanted me to do computer science, so maybe she's a computer science geek. No, no, no, she's not technical. I mean, the easiest way is just it's currently the fastest blockchain, but that assumes you know about blockchain, right? So I think the easiest way to describe it is analogies people are used to. So current blockchains are, you know, a 56K dial-up modem, and Sonic is ADSL.

55:27I think that's what it comes down to. It's just higher throughput, more responsiveness. You're saying, I believe, Sonic is the best blockchain out there. Currently. Why? Fastest consensus, lowest ETF, fastest read-writes. There's obviously a myriad of reasons. From the technical side, even when we launched the original Phantom, the consensus was one of the fastest out there. And the consensus we've almost not changed at all. There's been a few optimizations, but nothing significant. So when we released the original Phantom, this was, what, 2018, the consensus on raw transfer transactions could already do between 50 ,000 and 60 ,000 transactions per second.

56:15But it was hard capped at 200 transactions per second. That's the absolute max it could do. And why is that? That's entirely because of the Ethereum virtual machine. So since then, we basically spent that time rewriting that virtual machine into what we have now, where you still have solidity and Viper compatibility, but underneath the hood, it's something completely new. And that now allows us to get upwards of 10, 20, 30, theoretically 400 ,000 transactions per second. But we're also at a point where we have a lot more capacity than we have demand. So it doesn't really matter at this point. So the things we're focused on is time to finality.

56:50And that's responsiveness. That's when, again, if you look old internet, you click the button to download an image, you would go make a coffee and go make a pizza or whatever. And when you come back, it's still kind of buffering and downloading. That's how you can think of finality. It's how quickly does that image get done? Because even I could take Ethereum today and get them to a million transactions per second without a problem, but block times are going to be like 10 minutes. So you're going to wait 10 minutes until it's actually finalized. Our time to finality with Sonic is sub 200 milliseconds.

57:27So you click, it's done. It's faster than human responsiveness can recognize. And then other small things that we did, right? It's like almost all RPCs used by blockchains out there are asynchronous. So how that's designed is if I click on that button, it sends my request to that server. And then in the front end, you build some polling system that keep asking, is it done? Is it done? Is it done? Is it done? Now, normally you have to do that because if it takes, you know, a minute, you can't do it synchronous because your request is going to time out on the HTTP protocol before it actually returns the response.

58:04Ours is synchronous. So you click there, you get the response. By the time it gives that response to you, it's finalized. So stuff like that makes the user experience better. Then there's stuff that we've started to, I mean, technically, it is the fastest blockchain, but that to me matters less. So I'm still designing this blockchain from a DAP developer perspective because that's what I am. So I need a blockchain that can do certain things that make my apps better. So one of the first things, for example, is fee monetization. So normal blockchains, 100 % of the fees go to the validators. So every time you do a transaction, you know there's a small fee connected to it where it's a cent or whatever, 100 % goes to the validators for securing the network.

58:50In Sonic, 90 % goes to the DAP that generates that fee. So if you use my app and you generate$100 of fees, $90 is coming to my pocket. So that's already important because it allows me to subsidize user actions. It allows me to decrease it if I want. It allows me a revenue stream if I need because not all apps have a revenue stream. So they need some way that they can actually make money. So that's number one. Then number two is fee subsidies, which we're releasing next, which allows me to subsidize another user's transactions that is interacting with my app. So that means I can start having users that don't have a wallet, that don't have gas.

59:25Because to me, that is such a big barrier to entry for 99 % of dApps. If your first touchpoint with a user is go download this wallet and then go buy this token on an exchange, you've lost 99.9 % of your users. But if you're like, hey, just come use it. You don't need anything. all of a sudden you have a lot more attention. And once you have them hooked, you can still try and convince them to do that stuff. So that's fee subsidies. Then the last one, not the last one, but another one that we have in there, which is all about that sort of end user experience is native account abstraction with economic abstraction, which means the user will never need a wallet.

59:59They'll use their Google off, email password, fingerprint, face, whatever it is, to access the DAP and to interact with it. And they'll never need to know about Sonic or RIS token. And these are things I need to build dApps that are better for a user experience. So all of these things I've mentioned make our blockchain slower because every single transaction now has this extra little bit of overhead. So if we were complete performance maxis, we would never add this. But the goal isn't to be a performance maxi. The goal is to have enough capacity that we can maintain the demand. And, you know, it's always that sort of catch up because, like, eventually the demand will exceed it and then fees will increase.

1:00:36and then we have to hopefully be ready for the next capacity unlock, which we have, and that's what we're working on for a roadmap. But at the same time, we need to improve the UX and UI from the DApp developer side, because otherwise we're never going to have enough users to reach the demand that will fulfill that capacity. And I think that part's almost more important currently than the speed. I mean, the speed is great, but I mean, there's enough fast blockchains out there. And users also don't really care. I mean, L2s are showing us that it might as well be some dude's PC on AWS. Ultimately, they don't actually care.

1:01:10There just needs to be some security considerations. Absolutely. I had a brand done from Phantom on recently. We talked a lot about UX, right? It's everything. It's everything. It's everything. And also what we're seeing, you know, I mean, the way you interacted with these apps in original DeFi was basically directly via smart contracts. you know, CLI tools or Etherscan or whatever. Nowadays, it's all about that user channel. And then we can see it, right? I mean, Solana is going through that whole thing where Radium is now launching a Pump.fun and Pump.fun is launching a Radium because they all know how important keeping that user acquisition pipeline is.

1:01:51Absolutely. So it's changing. What will you tell people who are confused between Sonic and Phantom? What's the difference? Phantom's dead. We migrated everything over to Sonic. I mean, Phantom will keep running in theory, but that's like Ethereum Classic versus Ethereum. Us as a team, I forked, we're focused on Sonic, we'll be maintaining there. Phantom is obviously a decentralized blockchain, so it will exist for as long as there are validators or anyone out there willing to validate for it. Then we'll keep validators on there for pretty much as long as we exist, but the focus and attention isn't on there.

1:02:26Phantom was a, at the time of its release, was the fastest blockchain available. But it was released when this space looked completely differently. So it is no longer viable as a product for the current market demand. You said in the beginning, we're in New York. I don't like to be here. That's true. Where will you go? Cape Town. I don't even need to think twice about that. No, Cape Town, South Africa is home. best nature, best weather, best food. This place can't compete with it. When do you think that's going to be? As soon as I'm done with the exchange. That's probably a long time, but when done, then Cape Town.

1:03:14That's for sure. Thank you so much, André, for doing this. That was nice. Brilliant conversation. Don't forget Jupiter. Oh, sorry. Apologies, Mantle. I put my cup in front of you. Thank you.