Evan Szu - CEO at Gamma Prime

Welcome to another CEO Wisdom pod. We have Evan Su with us today. Evan is a founder and CEO at Gamma Prime, uh, not his first stint, was co-founder at a bunch of other companies, has a PhD in Stanford of all things. So, uh, Evan, welcome to the podcast. Tell us a bit more about yourself, what you're up to.

Hi Charles, good to meet you, glad to be here, and, uh, hello to your audience. Yeah, I'm, uh Grew up in the States, um, have one of these weird backgrounds where I have a PhD, but I'm also a trader, like, in commodities. Um, it's one of the really early people in machine learning, AI, uh, before people even knew it was a thing. Um, so almost 30 years in there as well.

What got you into commodities trading, and how did you stay for 30 years there?

It was really my mom's fault. I remember when I was in, uh, uh, was doing college, she said, uh, she never gave me more than like $30 for allowance a month. And then one day she was like, hey, do you want like a, you know, $10,000 to, as she called it, play in the stock market? And I was like, I've never— this must be very important because she would never offer me that kind of money for anything. And I got into it and it just tickled something in me. It's one of those things where a lot of traders will tell you the first few years are the worst. You blow up accounts, you lose money. Usually that encourages people to leave. They're like, this is not for me. I just kept going back for more abuse. And at some point you get good because, you know, you just can't help it. You make, make all the mistakes in the book. So probably about 7 or 8 years in, I started getting the hang of it. And, uh, since then it's been, it's been pretty good.

So, and yeah, your first role here that I'm reading is at Capital One. Um, yes, but it seems you were always in machine learning, like in the very early days. Why? Um, most people would think that AI was like fantasy back then. So, uh, what prompted you to, I don't know, believe in it and leverage AI to be better at your work?

I'd like to say I had some insight and like, you know, I could see it coming, but it's just an accident because my dad is the founder of the International Neural Network Society along with like a few other people, like in 1987, I think. And so when I grew up, on the dinner table, I had my dad just cramming this math and machine learning stuff down my throat, telling me how it's gonna be the future and whatnot. And I didn't know any better. So I went off to college, you know, and I was like, maybe I'll try studying it. And so I did a computer science degree at Brown University and kind of folded in some AI. And then, so it was just an accident. And honestly, for the first 10 years, no one believed in it. Now it's all the rage, everyone's an expert. But like, um, you know, back then it was a weird niche thing. It's like probably like quantum computing now, where everyone's like, I heard of it, it's kind of like supposed to be awesome, but I don't know anything about it. And then, and what has it done for me? Like, what has quantum done for anyone in reality? Well, nothing yet, you know. Um, one day it may change the world, we'll see. But, uh, yeah.

Interesting. So, but did you get gains from using, um, AI back then, or was it mostly experimental? Did that give you an edge, basically?

So I tried in the trading. I'll tell you, this is what every machine learning person does. First thing they do is they try to use it to make money on the markets. It's just so— if your audience, if you guys are wondering, can you do this, the answer is not really. The answer is like, this is the first thing every machine learning tries, like, oh, I'm going to use it to make money. Okay. Just briefly why it doesn't work, just so you understand, is it's a classic data mismatch. And what I mean is, let's say you're trying to predict the price of oil, right? There's all this stuff that goes into it. It's geopolitics and the amount of inventories in different countries, but it's also all the psychology of every trader that's in there trading these paper contracts. Essentially, it's like millions of variables, right? Well, how much data do you have to model that? Maybe a few thousand data points, and you certainly don't have any history that's like what's happening today. That's a classic data mismatch. Way too many variables, way too little data. It'll never work. Now, what it can do is it can automate some sort of like manual tasks. Right. So like it can send news alerts and constantly be watching online. So I have been using it for trading. Um, it's— I do have some more sophisticated stuff. You can do things like regime detection. Are we in a generally risk-on or risk-off regime? But even that takes judgment because risk-off in '08 is different than risk-off in COVID, right? It's a totally different kind of risk-off. And so again, how you trade it is like it ultimately comes down to the human skill. So I would say that it's, uh, it's useful because it gives you an objective kind of dashboard, but in and of itself, I don't actually feel like it's actually contributed meaningfully to my edge.

And do you feel that AIs are getting that judgment that you're talking about that would be useful to be very successful as a trader?

I'll tell you, every good hedge fund, every Joe like me that's like knows anything about it is always— I haven't seen it yet. I think it can do things. Like I said, I'll tell you where I see advantage in this. If you already know a particular trading style and you're pretty decent at it, AI can make that much more efficient, reliable, reproducible. Essentially, you are training the AI on your— I'm seeing a blank here. Okay, maybe on my end. You're training on your style, your particular corner of the market, your niche, right? And when you do that, of course the machine's going to be better than you. It can stay up 24/7. It can, like, monitor a lot more data. It can backtest more quickly. But you still need to start it somewhere where that actually works. If you just toss it out there and you say, ChatGPT, go make me money, it's not gonna work. It doesn't know where to start. It's like I could just pile variables in and just randomly start testing it. But the odds of you stumbling on something that's actually gonna create like viable edge, I, I, I'm not, maybe it's happened. I haven't seen it. Right.

Um, and how successful were you as a trader and what prompted you to, go back and study at Stanford in 2014 as a research scientist?

So I'll start with the— I'll start with the downside because I always like to do that, to be honest. I've blown out at least 2 accounts before my first 7 years. Blowing out means like I lost enough that the account was essentially dead, you know. I think it was like a $70,000 account which was growing. I thought I was a genius and And another one, I thought it was the first time I got it and it blew up. Haven't had that happen since then. In the last 15 years, I've averaged 23% return. So if you compound that, it works out to like a 16 or 17x gain. So I've taken, you know, I don't trade other people's money, it's just me and my family's money, and essentially added zeros to the accounts. So But I always— oh, I'm off again here.

Okay.

But I always like tell people it comes with a vomit bag. Like I'll have drawdowns of like 23, 25, 28% regularly, you know, and most people can't deal with that kind of volatility. Now, you know, and this last time with war, it's been even worse. Like the drawdowns have been even more violent. Because it just moves really quickly. But I'm still up for the year. So it sounds good on paper, but the reality is I don't know that most people could deal with that for their retirement accounts. Like, they would just— they just lose sleep over it. And, and so I do it because I'm an addict. I do it because I just love markets, can't get enough. And, uh, and it's, it's kind of like my favorite you know, like pastime. I think it's fascinating how these things work and, you know, using different tools like AI. Um, I'm sorry, I don't know if it might be on my end, it keeps playing, but we can hear you well.

But what was the PhD at Stanford?

Oh, something totally different. It was actually in, um, education. I know that's weird. Uh, it's actually That's another passion of mine. I really find the art of teaching is a very specific thing that most— you've probably seen this where you go to like a university or something and the professor is brilliant at their field and they're a terrible teacher because teaching is a whole different skill, you know, and I just found that fascinating. And I had a chance to go back and work with Dr. Jonathan Osborne. He designed the US National Next Generation Science Standards, and he was doing that when I first joined, and I got to work with this guy and help him build it. And I just felt like this was an opportunity I couldn't pass up. I was like, this is just— it's, it's, it's an amazing chance to like be in the trenches and, and maybe make a difference. Not teaching one classroom at a time, but building the whole standard for the whole nation, and a lot of other nations may even follow it as well. And I was like, I just can't pass it up. So I actually— it chopped out 7 years right in the middle of the peak of my career. I'd already done a startup before, and I came out and did a startup after, and another one after that. But it just— I couldn't resist. So in that middle 7 years, it just— I did it when I was like 30-some years old, so it's not like it was fresh out of school. It was just something I love doing.

So, and what was Origin Labs?

Uh, so Origin Labs, uh, was using machine learning for, uh, cancer, uh, diagnostics. So, um, my background's in science and of course the, the AI side, and, um, we, uh, took, um, breast cancer, you know, biopsies and and trained it on— usually these are read by human pathologists. So this would be like, you know, a very high-paid doctor looking at slides and trying to pick out the, like, shapes of cells and tell you whether or not you have stage 2, 3, or 4 cancer. It's like a very important decision. Like, if you screwed it up, you could send someone to a hospital unnecessarily, or you could, worse yet, miss something that needs aggressive treatment. It's a very difficult task to do. It's a perfect job for AI. Like, why should you have a human doing that? You know, the AI can scan the entire slide. So we did this, you know, it was almost 10 years ago now when image recognition was good, just getting good. And, uh, yeah, we built out a, a model. It got acquired by one of these big pharma companies and it's in use now actually. Option.

And what followed this one was another one in biotech, Arrive Bio. What— which one was that, and why was it incorporated in Singapore?

Yeah, this was, uh, was, uh, we were trying to extend this into mouse studies. So the idea is we were like, hey, we know how to read these cancer slices, maybe we can do that for, you know, let's say you do a drug and on MICE. Singapore was a natural environment because of just really talented workforce. We got some incentives there. To be frank, that company didn't do so well. You know, it closed down and we, you know, we, I won't make excuse for it. This is the reality of doing startups, which is, you know, you're lucky if you have one or two of them work out of five. But hey, look, It's, um, it's, it's the chance you take. And, uh, our timing is off. It was very tough market for biotech. Uh, we didn't get funded, so we did fold it up.

And still is nowadays. It's— biotech is not easy. Um, what about Gamma Prime? What prompted you to actually start it?

Yeah, so Gamma Prime was a chance to take AI and apply it to the other passion, which is trading in markets. And, you know, at this point, like I said, I have that 16, 17x gain. So, you know, essentially have a small family office that is now actually worth a decent chunk of change. When you add a zero to the net worth, it starts to get, you know, a little beefy. And so we encountered a bunch of these other small family offices, same thing, not like a big I got a billion dollars. Okay, if you have that, you can have your own staff doing research, right? That's not us, but we still have enough money that it really matters, right? So what we learned is there's a lot of these little kind of small, high net worth family office-ish type people that all have the same problem. Where do we go to find experts that are going to like, you know, help us find something that's more than just buying stocks, you know, because buying stocks is like— don't get me going, but I think buying stocks is like— the only thing worse than that's just keeping it in cash. Um, it's just totally— everything correlates with itself when there's a crash. It looks like it's uncorrelated today, and then when there's a crash, it crashes together. It's not genuine diversification. So, you know, They met me, these other 3 family offices. They're like, you know, commodities. And I'd be like, yeah. And you know, hard money lending. And this person over here knows European, you know, private equity. And we were like, we should just gang up together and like, just kind of make our own little like information sharing hub. And, uh, that's the Prime. And it has grown since then. We have big events now. I don't know if I can show you this here. Um, this is, uh, okay, let's see if I can share screen here. This is probably not going to work. Okay, let's see here. Screen 2, just so you can see it. So we actually do these large events. This is in, um, this is Miami. Can you see this here?

Yeah, yeah.

So this is a Gamma Prime's Miami event, um, and you can see we have these monstrous, you know, panels and and like people come out and speak, essentially sharing their information. Uh, this is a big one we did in Hong Kong. Um, these are all free, not selling anything. It's just, it's because this is how you meet people with knowledge and network, you know. Um, same thing with people on stage, not selling anything. They just go on stage because they enjoy sharing their knowledge. Some of the best VCs and private equity. Then we do stream it. Um, Yeah, about 86,000 viewers. You can see here that our next one's in Singapore at the Fullerton in October. Anyway, this is just to say, like, uh, it's— that's what it's grown into, which is, hey, this is a, this is a place for people like us who are accredited plus, you know, up to— we do have a couple of these monster billion-dollar family offices, but usually it's people that like have enough money that it matters but not enough money you can have like a a team doing it for you. And that's Gamma Prime. So, um, you know, uses AI, uses blockchain as a backend and stuff. But essentially what it is, is it's a buy-side knowledge aggregator.

So I can check at like the most successful investments, I can copy them like eToro style, but I can also do that, um, with crypto investment and Do I— am I seeing like pre-IPO, um, investment opportunities here as well? Like, it seems it's more diversified, right?

Yeah. So I would give, uh, your audience an example. I'll give you an example of where this buy-side knowledge comes in. There are a lot of places you can go and find a listing of pre-IPO secondaries. So for instance, Anthropic is going IPO, right? SpaceX went IPO. People were selling shares beforehand. That's not new. That's just a piece of tech. You're just listing, you know, the thing that's difficult about pre-IPO secondaries is can you trust listing? Does a person that say they have the shares, do they actually have the shares? Do they actually have a layer 1? Is it direct to cap table, or is it in an SPV with like 2 layers and 3 middlemen in the middle? Most of the time, if you just go and you knock on the door of these deals, you'll find they're garbage. There's something in the middle of it which is like not for real. The guy says he has shares, but he doesn't, you know, and then he gets, you know, he gets you on the phone just so he can get like your contact info and sell you other stuff because he knows you're rich, right? And like, you have all sorts of shady stuff on the on the sell side. And you have all sorts of shady stuff on the buy side too. You have people that say they have the money but they don't actually. They're just doing it so they can get the Anthropic CEO on the phone and like, you know, pick his brain about like, you know, who knows what. The entire industry is garbage. 90%, 95%. The thing that matters is not the deal, it's who's behind the deal. Right? It's actually the reliability of the human factor. And that's where the Gamma Prime edge comes in. That's why I say like we're a network of buy-side knowledge. We're not a network of deals, not like eToro where you can copy stuff. It's like the stuff that gets on the platform has been vetted by both sides. And when we meet someone that is a reliable partner, we give them more and more latitude to post deals. Right. And if they screw up once, we cut them out. So it's like the result is that you end up with a trust network and that's really the value of it. It's not the deal itself. There's plenty of places you can find hedge funds and private equity deals and, you know, secondaries, but which ones are actually worth investing in? Which ones are in demand and actually have a bunch of other family offices that have already done their homework and have already put money into it. Well, that's the one you're really interested in because it means somebody else has already vetted it 3 times and they've passed muster and they're still with the fund after a year and a half. Hmm, I'm interested. So that's GammaPrime.

And you have a product for funds as well? How does that go?

Usually we, like I said, we're, we're a buy-side aggregator. So what we'll do is we'll essentially see who's interested on the buy-side. And when we, when a fund's name starts coming up a lot, we'll go knock on their door and be like, hey, look, we like you. Other people seem to like you. Onboard onto our platform. We'll be able to like convey the value of what you're bringing because you have sort of network buy-in. And sure, we're going to charge a fee if we place more capital with you. But it's only success-based fee. Like, it doesn't cost you anything to do the platform. Everyone else, it's totally free. People can come to the events, investors, doesn't cost anything. People can speak, there's no cost to it, and the funds can join without any cost. It's only if we actively place capital with them, yes, we do take a fee for them.

And the main way to get clients, your main way of marketing, is it events nowadays? How do you get these high-net-worth people to trust you with their money.

So just for clarity, we're not actually taking custody of capital. So it's not like anyone's like giving us the money and like we're investing in it. We're just conveying information. So again, it's like do your own diligence, you know. But again, if you don't like it, we'll give you your money back. Oh, wait a minute, we didn't charge anything. So, you know, the, um, our, uh, our approach is, uh is essentially meeting people through network because this is how the entire private market industry works. It's not like stocks where you just buy on an exchange, buying and selling directly. Um, it's not like bonds or even commodities where there's like a futures market. These deals are very custom. They're very much like one person meets another, they get to know each other, they do some vetting, talk to their friend, fund knows somebody else, they pass on the information, eventually they place capital. But you know, when they place capital, they place $1 million in a pop. So it's worth it for that time to get to know people. What we want to do is make that a little bit more efficient. So we do have tech, we have a platform that allows people to convey. But we learned is that ultimately it's still, people need to meet first face to face. Which is why we do these big events. After that, they can go digital and online, then it's okay. Then it's like, okay, I already know who you are.

So, so, um, how do you make your money? Is it like a percentage of the cash they put on the platform?

Yeah, it's percentage of, uh, assets that were deployed as a result of the platform. So, uh, it, it comes from the fund side. We essentially become their marketing arm.

Are you mostly active in Asia or in the US as well? Because I saw your Hong Kong events. Is it mostly Asia or—

our biggest areas are, um, Far East for sure, um, MENA, Middle East, North Africa. So, um, we do events in Dubai, Abu Dhabi, and, um, Europe, usually smaller events, and then some in the US. It's actually We're sort of starting in the less serviced areas, you know, and the US is pretty saturated, you know, so we are doing events there because there's demand, but I would say our specialization is sort of everything but.

And why not run a podcast in which you would interview high network people or have someone on your team do it? That would put you constantly in touch with like high network people.

We do. We actually run X Spaces, um, and, uh, often our focus is really more on sort of providing information values. We'll essentially highlight someone that, uh, is in our network that has knowledge of a particular niche, and then they'll sort of talk about it. Um, and we find that's the best way to sort of build credibility, where it's just like, you know, look, we're just sharing the things that we know. You know, I only know one corner of the world, commodities trading, and even that I only know a certain set of the commodities. So if we get outside that, I can't help you, but we have a lot of people who can.

So, and crypto, is high-net-worth individuals bullish on crypto? Are they scared by it?

It's a great question. Um, I'll give you a macro answer and a micro answer, uh, because I am a trader and I do trade in crypto. Crypto has been in a bear market for sure, right? It was up at $128,000 Bitcoin and it's come down to almost less than half of that, $60,000. That's going to scare any good bull. That's not something you can take easily and be like, ah, cut in half, no problem. The crypto VC world, you know, venture capital is suffering quite a bit. I think they're having a hard time raising money. Projects have not had liquidity, so they have no exits. And when you look at that, just that by itself looks pretty bearish. That's not like a good story. However, a good trader is always looking to buy when everyone else is the most bearish. That's just a trader's instinct. We can't help ourselves. We stick our nose out and get it chopped off, and then we stick our nose out again in the same trade because ultimately you make the most money by being courageous when people are scared. And when everyone is bullish and crazy, you're the one selling, getting out. So right now everyone's talking about, oh, stock market's at all-time highs and it's just going to go to the moon and whatnot. And crypto, crypto's dead. Well, I'll tell you, as a trader, I pretty much have the opposite view. I would never buy stocks now. I wouldn't buy with your money unless I hated your guts and I wanted you to like be set up for a fall. Why? Because sure, it could go up another 5%, but the downside risk is quite a bit. Like, you know, it could go down 30%. I think we're headed into economic recession. We've had an oil spike, right? There's wars going on. The economy seems to be teetering. We haven't had job creation in over a year. We did, but they all got revised away. Again, all these little details where it's just like, I would not be buying stocks now, but crypto is a different story. I actually am inching back into crypto markets, long Bitcoin for the first time. On our internal sort of little chat that we do with our China worth people, we called a crypto bottom Well, it could probably— it's probably— it was probably about a month and a half ago, about a month ago. Uh, we were not right in the absolute sense. It did make a slightly lower low, but I'll tell you, it's behaving in the way that a trader would say it looks like it wants to bottom, like 65, 67, and like pushing upwards. And when there's like an ugly thing in the Middle East, um, Crypto, 2-year bonds, and gold all have a bit of a correlation. They all tend to drop. And what you see is gold has dropped somewhat, 2-year bonds have dropped more, and crypto's dropped less. So when you see that sort of divergence, you're like, there's strength there. There are buyers, they're coming out of the woodwork, they're sniffing. There's more than just me out there thinking, maybe I'll take a shot at this. So in that sense, I'm actually bullish.

Yeah, thanks for being on, Evan. Such a fascinating background you have and very interesting company as well. Where can, uh, people find out more about you?

Uh, come to gammaprime.com. You can find our next event in Singapore. Like I said, it's in the beautiful Fullerton, totally free. And, um, come meet me as well as our whole team, all these other kind of the other family offices that have joined together in Gamma Prime. We're all there. Um, and, uh, yeah, we just love to meet you. And if we can't find you in Singapore, just look on the page, there'll be another one. There's another one in Abu Dhabi, and there's going to be another one in the U.S., and maybe Denver. So we're all over the place, and we'd love to have you come out and meet us.

Was Evan and Charles Cormier for another