Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Insilico Terminal Podcast Episode 23 - Pedma
108 Plays1 month ago
In this episode, Pedma shares his journey from bookkeeping and small-cap equities into fully systematic crypto trading. We dive deep into the difference between discretionary and quant approaches, why most retail trading is just momentum “seen through a potato,” and how real edge requires understanding why you’re getting paid for exposure. Pedma explains his long/short momentum models, carry strategies across perp DEXs, volatility targeting, and infrastructure for systematic execution. We also discuss alpha vs risk premium harvesting, prediction markets, perp DEX innovation, AI in trading, scaling capital, and the psychological impact of increasing size. The episode closes with practical advice: curate your network carefully, focus on useful knowledge, and understand that trading is a high-uncertainty profession that demands emotional control and long-term commitment.
00:00 Intro, Pedma's background, moving from equties to crypto and systematic trading
06:27 Small-cap volume stats that first worked, float rotation, squeeze logic and data-driven edge
09:48 Moving to crypto in 2021, momentum carryover, DAO forks and early volatility wins
13:28 Biggest mistakes, complacency after success, drawdowns and overconfidence cycles
16:31 Current strategies, long/short momentum, carry, volatility targeting and perp DEX farming
23:23 Discretionary vs quant debate, alpha vs risk-premium harvesting, edge philosophy
30:54 Execution, manual control, TWAP/slippage checks and infrastructure design
45:10 Future of crypto, prediction markets, lifestyle, psychology at size and closing advice
Recommended
Transcript
Introduction to Pedma and His Trading Journey
00:00:12
Speaker
welcome to a new episode of the in silico terminal podcast my guest today is pedma is that how you how you pronounce your name Yeah, that is perfect.
00:00:27
Speaker
And I like to ask people to introduce themselves. So in case someone doesn't know you yet, who are you? What? How do you describe yourself? Yeah, sure. So am mostly a systematic trader. At least that's what I do. um I am based in Portugal. I've been doing trading for ah for about since 2017. So almost nine years will be in December.
00:00:53
Speaker
I've started trading mostly in equities when I began, like shorting small caps, microcap equities, and then transitioned more into large caps during the 2020 and 2021 like strong cycle, you want to call it that. And then in late 2021, in September, I kind of switched more my focus to to crypto. And that's that's where I've been putting my my entire time on just doing systematic trading in crypto.
Motivations and Early Experiences in Trading
00:01:21
Speaker
What made you first get into trading? Uh, yeah. So it it's only see like I was 19 or 20 years old at the time. So it's what every kid at that age wanted, right? It's like being rich, making a lot of money and all the naivety that comes with it. So. Yeah, that that was the primary focus. Like I just wanted to, I didn't like much my job at the time. It was not like mentally stimulating. i was a bookkeeper. um And so I did a bunch of things before. I tried dropshipping, like marketing, all that type of stuff. And then I landed on trading after looking up a few grifters online and they convinced me that that was the path.
00:02:03
Speaker
So how did you decide to get into systematic trading? Because I think that's a bit unusual. Most people just like try to do it manually, especially when they learn it from ah or like get into it through these types of people.
00:02:14
Speaker
Yeah. So, because like I was um i was a bookkeeper at the time, so i I worked a lot with spreadsheets and data and accounting and stuff like that. So I always had this inclination for, okay, let's try to to get some data here. Even though at at the start, I started like normally doing discretionary, like drawing lines on charts, whatever. But over time, I started getting some um information from people ah that I respected that were very data driven, especially in the microcap equity space. And ah ah they they did very basic stuff, but stuff that made sense to me at the time. And they were tracking data on spreadsheets, tracking a bunch of different data points like market capitalization, like flow to volume, volume at time, like a bunch of different things that affect that I thought had some effect on price. and so in in ah In about 2018, I was losing money every month by then and money money wasn't very abundant. And so decided I had to do something different and started tracking a bunch of data. Like I have still i have like ah ah an external disk here on my house. We saw the data from that back then that I sometimes like to just look at.
00:03:30
Speaker
ah and And yeah, that's how I started. I started losing money. So people that I respected was doing data and data was something that made sense to me. And I began tracking everything that I could. And how did you, like, um since you don't really have a coding background, I guess, how did you get into the systematic side without that? Did you just, like, teach yourself?
00:03:48
Speaker
Yeah, exactly. So I taught myself, I have, like, a high school, I took, like, in Portugal, you can take, like, a technical route. So I i graduated high school as an electrical tech technician, and then I took a one-year course in bookkeeping. so everything was like I had to learn how to do statistics, how to do Excel. And then eventually later on, I think it was in 2019, 2020 when started to learn to code in Python. So yeah, I taught myself.
00:04:16
Speaker
And what were some of the the data and the statistics that you looked at at at the beginning? um Maybe also something where you can say like this didn't help you that much. And then you looked at something else and that then you saw some improvement in in your
Learning and Evolving Trading Strategies
00:04:29
Speaker
trading. Like what what did you look at in general?
00:04:31
Speaker
Sure. ah So ah I'll start with something that didn't help me when I it was this was 2018 and I was tracking like ah so there was this is a bit silly now, but it's it's what happened like there was like this pattern that I looked at. It was like um a lower high, a higher high and a lower high. And that was like my first memory that I see like that I was started tracking something. So I was tracking how much did price went from open to that lower high then to that higher high then to that lower high. And then I would try to um understand if it broke that lower line, ah how much down it would go. Like this is very silly now, but it was a very important exercise at the time because I started to
00:05:16
Speaker
uh, track some data and have some statistical backing, even even if naive about what what I was trying to do. And then later on, there was a lot of guys sharing a lot of information on Twitter and there was this guy, I think it his name was like all day faders or something like that. And he was tracking information about volume ah ah during the day on this small cap. So at at the beginning of the day, you had these massive moves in this large gap. So they would gap from previous day close like 50% or something, and they would get a lot of attention that day and they would trade like
00:05:51
Speaker
usually like so yeah if memory doesn't fail, miss 70% of the volume in the first hour or hour and a half, something like that. And what you could see is that if that trend of volume didn't and wasn't starting to go down, like if he traded like 70% of volume, but then in the in the second or third hour, it started trading even more volume than it was expected, like you would develop like a ratio of volume that was expected. you would understand that that day was we're setting up to be a squeeze day. So a lot of volume was bringing into the afternoon and that wasn't and that wasn't usual for the faders I was looking at. I wanted to see um stuff that popped up in the morning, went down, people lost interest and then volume died out over the the remaining of the day. And so that was a very important statistic at the time that actually ah began starting making some money then. And it made fundamental sense because those markets were very driven by retail, by retail investors even more at the time. And so volume um was everything. If you saw a lot of volume there and it was consuming all the float, that was another metric. Like how many rotation of the float did you get ah from volume? So for example, if you have floats, if a float, if people don't understand is the
00:07:05
Speaker
ah is is the the shares that are ah circulating on the market. not Not shares outstanding, but the shares that are ah circulating. um And so if if volume consumed like two, three, four times the ratio of the float, it means that it was more, how do you say, what is the word? Like it was more inclined to have squeezes later on the day. And so you could track all of these metrics and they made fundamental sense about behavior in markets. And and I started making some money in 2019.
00:07:38
Speaker
what Why ah do would you describe the first strategy or like the first data pointer to use with the lower highs and the higher highs and stuff as as silly?
00:07:47
Speaker
uh why because because it ah it has no like a lot of what i do today a lot of of what i do today it has to have some like fundamental uh effect why it should work like why would someone uh pay more than they should for this exposure or what kind of service am i offering the market like from the ideas that Robo James also shares a lot on Twitter, like those type, I always try to understand why am I being paid for this current exposure? And if you go track higher lows or higher highs or whatever, like, yeah, it's it's all like just a proxy for momentum, right? If you get to break a break out of that or a breakdown, that's just momentum. It just... yeah ah It's just being seen like in a very, and in a very crude way. You can quantify momentum in very much easier than that and much, that makes much much more sense.
00:08:44
Speaker
It just, so ah sorry, if um it's just like when we do retail ah discretionary type trading, we always type to, we always, I see that people try always to make these like patterns and stuff as some meaning, you know, and that's what I was trying to do. That's why I call it naive and silly. I was like, okay, this is a pattern I can make money. No, you are taking advantage of some underlying effect that you don't understand. And you are just using this proxy, this bad proxy for that. And that doesn't make sense.
00:09:14
Speaker
Yeah, that's kind of what I was getting at, ah especially since you look at stuff now more from a quantitative lens. I think that's ah an interesting insight because many people look at markets and um trading in general a bit as like, like kind of overfitting just but like what you want to see instead of what actually is useful to look at.
00:09:33
Speaker
Exactly. It's like, I think it was scott ah Scott Phillips that said like, you are just seeing like momentum through a potato or something like that, you know, because it's at the end of the day, most of discretionary trading is momentum. If you quantify their, if you kind of decompose their returns by, by factors, it's like, it's it's all streamed down from
Transition to Crypto and Initial Success
00:09:53
Speaker
momentum. So, so what made you get into crypto then?
00:09:58
Speaker
Because so in 2020, to give just a bit of the story, like in 2020, I was making decent returns, even though I had not ah not a lot of money back then, ah ah but I was making very decent percentage return during that time. And when the market kind of started slowing down um ah in in late 2021, if I'm not mistaken, ah like my return started tanking in in large caps. And I was still in in that phase that um ah it it was still exploratory so i saw a lot of people ah making a lot of money in crypto and i thought like these people are very dumb from the stuff that they say and maybe i can i can make it there too i since i can bring a little bit of a data backing i i knew how to code by then uh and so that that was the reason like there's so much volatility so much volume so so much dumb flow that i thought at the time and it's true at at at some point but not at the extent i thought
00:10:55
Speaker
um Maybe I can do the something there. And that's the reason. Did you use the same strategies that you used previously in Tread file? Yes. um Yes, I did. Like everything was streamed down like from momentum, even though even when I transitioned to crypto, there was a lot of dumb stuff that I was still doing very dumb stuff. But everything was like, okay, yeah I do. I've made money. This is the like, it ah In 2020 was the first time I made money trading.
00:11:25
Speaker
Like ah in 2019, in small caps, started doing well systematically, but it was not enough to cover for the fees, for the high borrow costs that you have to incur when trying to borrow shares. And so in 2020 was the first time I made like, okay, I'm making money. I could live on this.
00:11:42
Speaker
And so I applied everything that I had learned until then, what worked and I brought it to crypto. ah the The only difference is that i brought it to crypto at the top and I was doing the long, long list of them. Yeah. Story of my life.
00:11:58
Speaker
So it didn't work out that that well at the beginning. ah You know what? It actually yeah the the momentum stuff it did, but ah because I joined in September and in that curve from September to ah November, if ah if memory doesn't fail me, it was like a very strong period. I remember like having 100 positions open, which is dumb, but I remember like making a lot of my being at the gym and saying, damn, I'm i'm up 100% of my account in the last two weeks, something like that. And then I also started trading like very niche ah edges at the time that I didn't know. But for example, at the time there was ah an explore, like a lot of DAOs from home DAO back then were coming up and there was all of these um ah forks of that home DAO. And so I developed like not a system, but a method to be on Twitter and searching for keywords. And I would be one of the first on these new forks and all of the forks would like pump like act I remember the names like actor um was someone on on on so not Sonic on Phantom as well. Like and I made, I think, 800% or something on the capital allocated to that strategy then. And so I didn't do very well in 2022. But that period like kind of ah sustained me in in the years to come until 2023. Mm hmm.
Reflecting on Mistakes and Complacency
00:13:22
Speaker
Did you ever have any um times where you got like blew an account or got liquidated or like lost a huge amount of capital? or What were like some of like the biggest mistakes if you can look back and you have anything like that that you can share or that you learn from?
00:13:36
Speaker
Yeah. um Good question. I never got like I never blew up an account after I transitioned from small caps. I think I blew one or I never truly blew up. Like I never saw my ever like my broker or exchange whatever at zero because I blew it up. I never did that. But there was periods I lost a a very ir irresponsible amount of my account that that's for sure. And the periods that that death happened even recently, even in 2025, year, we can talk about that later, but all the times is that is is me letting like my foot off the gas. though my ah my Like, because I noticed that in myself is like, I start, I start making like, i start doing well above what I am expected or whatever. I start doing well and I kind of like,
00:14:31
Speaker
uh, drift off, you know, okay, I don't need to do this. I'm doing well. What do I need? Like, I'm up like whatever they did today, like I'm up more than my yearly salary salary back then. what What do I know? That's one vacation. Like that, that, that's my stuff. And that's one of my biggest mistakes. And for example, from 2023 to about 2025, I was doing mostly a long only, uh, uh, momentum system.
00:14:54
Speaker
And I never transitioned from there because in 2023 to 2025, I did really, really well. Like my count was 200% gains at the peak. Like i was doing really well. And i i always use that forward to justify, okay, we just need to go through this drawdown, which is fine. Like you have drawdowns in trading, it's normal. but Yeah. I have all this time, all this knowledge that i could apply to develop new systems and new and correlated sources of edge whatever, but I don't do it because I'm doing well. I've already done well, so I don't care.
00:15:25
Speaker
And so that would be my top one mistake if I had to to name one. I see. There's this saying, ah youre when you're, how does it go? When you feel the most safe is precisely when you're most vulnerable.
00:15:38
Speaker
Exactly. Which summarizes what she said there, I think. Yeah, and and um and if he that's all right. It's like every time it's so like every time I post a P&L that is at all time highs is the time I'm going to get smacked in the face. It's always the same type of idea.
00:15:56
Speaker
So have you have you learned to adapt to that a bit better now? Yes. I mean, yeah I think yes, at least over the past year or so, I've been like consistently ah putting it at work, improving all of my systems. Like I'm I look back one year, ah one year ago and like I'm completely to myself unrecognizable from what I was doing then of what I'm doing now. Like um And that has been ah an improvement, but it's an improvement that always comes from mistakes, right? so From losing a bunch of money. Like i need to get that wake up call once in a while to to to improve. and And that's a mistake i've I've made in the past. Yes.
00:16:37
Speaker
So I'm curious, like, how does your trading look nowadays?
Current Trading Strategies and Exploration
00:16:40
Speaker
What kind of strategies are you running? Yeah, so ah nowadays i am running I'm still running mostly ah momentum, but this time long short. So i'm but ah right now most of my book has a very short delta, um and that has helped a lot ah during the current environment. um I'm also running carry strategies.
00:17:05
Speaker
ah all of these carry strategies ah at the same time are like delta neutral and I'm running them across a bunch of perp dexes. So I was farming lighter. I was farming. I'm farming extended, variational, all of these perp dexes. I'm trying to be, um how do you say, like ah efficient in the way that you collect the points and all of that stuff. I see that I see that just as ah an extra cherry on the top. I'm doing um On the trend or momentum style, I'm doing both delta neutral and non-delta neutral type strategies. I'm applying one volatility like a volatility targeted type model and another one that is non-volatility targeted. So I have exposure to both types. Both types offer different risk profiles. Like, for example, one is more way more volatile, but yeah, can capture lot more returns while the volatility targeted one is more like higher sharp, but you kind of like don't get that convexity that you see in other, in the other type of models. And now I'm moving, I'm spending a lot of time understanding like mean reverting type strategies to add that uncorrelated property to my, to my portfolio. And at the same time, I'm spending a lot of time on these heap tree markets as well on Hyperliquid. So there's a lot of stuff to be done there. And that's a market I've spent a lot of time on.
00:18:30
Speaker
How much manual input is there in your trading since you're mostly systematic? but like Do you, for example, decide yourself, um like now is more of a bullish time, so I'm focusing more bias on the long side versus the short side or something like that?
00:18:47
Speaker
No. ah on on that side is 100% systematic. i have I make no decisions. i do have um I do have a portion of my portfolio that is allocated specifically to discretionary trading when there's these times of massive volatility like you had on during the tariffs, uh or during the summer and stuff like that i i do well ah in those times and so i i put a portion of my portfolio to discretionary trader during those times um but overall like not 80 of my year is full 100 systematic i i don't take any decisions it's all handled by my models
00:19:26
Speaker
I see. Do you have any opinions? ah I guess it has been like a month or something since the discussion came up on Twitter once again. And maybe it's also a bit interesting because most listeners of this podcast are almost all users of Silico Terminal are probably like focused on manually trading.
00:19:42
Speaker
um But you're more of a systematic trader. So do you do you have ah an opinion on that ongoing war between systematic and and manual trading? Is it is it worth it to to trade manually? Is it a waste of time? Should only quants be allowed to to trade the markets? or Do you have any opinions?
00:20:01
Speaker
Yeah, I do think that the distinction is is not so much um ah from manual to automated. I think the distinction is more on discretionary versus quant, because even though I'm still a systematic trader, I still execute all my positions manually because they have daily ah rebalances.
00:20:19
Speaker
So, for example, um every day at 8 a.m., 6 a.m., whatever time, i during the night, I have my all my models recalculate the exposures that I should have for that day. And then in the morning, I execute them manually. I do that because I like the control. I like to see what is executed into the market. And I like to follow up, OK, is my models working and stuff like that, because I'm doing a lot of upgrades of my of my infrastructure that handles all of that. um And there's nothing wrong with that. You're just yeah I don't take any decisions other than look at like what my execution algo is doing.
00:20:55
Speaker
Just that. So even though it's fully systematic, 100% make no decisions, I still get that manual execution. Now on that argument between the discretionary versus the quant, I think that both like both sides have merit. lean more on the quant side, ah to be honest with you.
00:21:14
Speaker
I do believe that most discretionary trading is gambling. Most people come ah to this place to like throw a punt, speculate on these or that asset that's or at least do very naive dumb stuff. And ah most of it is gambling and entertainment. So on on that side,
00:21:34
Speaker
on that side, I do agree with the quant, but yet there's another side that I don't agree with is that it's not everybody like yeah I've seen discretionary traders that are so good at what they do really so good that i will never probably be able to beat them, whatever I do, because they are so good at at being discretionary and taking a bunch of um a bunch of input from ah from a bunch of sources that the machine cannot have that intuition to do.
00:22:02
Speaker
And but ah ah The problem is that to get to that stage, it's really unlikely that you'll get there. And but if you go through a quant or systematic route or whatever, like you at least you quantify what you're trying to do, even though your returns won't be as large as you see some of these discretionary traders do, you'll have a ah much more um I think, in my opinion, a much more like reasonable expectation that you can do something. Now, there's a bunch of other questions like in terms of you can make your trading business operational. You have capital constraints like not everybody has a 100 or $250,000, whatever it is to live in your country to to make like a reasonable like 20%, 40% annualized returns and be able to live on that. So there's that question along the side, but in terms of edge or making some money, like making some return on capital, I do think that the systematic, like if you apply yourself and you you have a very targeted knowledge on stuff that makes sense, that there's some fundamental reason why it should work. I think you have better chances there.
00:23:09
Speaker
But again, I don't want to say, i don't like to use absolutes because there are some like superstar discretionary traders that I know that that have made eight figures, probably close to nine and I'll never, not never, but I'll probably never expect to get the the the returns that they get on an annual basis.
00:23:28
Speaker
I think it's very interesting. Yeah, go ahead. go ahead No, sorry. Sorry to interrupt you. um Yeah, I kind of wanted to make like an
Comparing Trading Styles and Maintaining Strategies
00:23:36
Speaker
analogy. I don't know if it works that well on the the quant side or whatever. But maybe you could say that the if you're going more into that route, it's a bit more like ah a normal job where you're like,
00:23:47
Speaker
have like a better expectation of like success in general and you also like learn a bit more of some other skills that you can use in other ways maybe and if you're like fully discretionary it's a bit like um if you're um how do you call it like an athlete like you're you're really trying to be really good at a sport like football or whatever And if you get to the top, if you're like one of the best people, then the reward is obviously huge. But most people that do like play sports or do anything like that, obviously never make it anywhere close to that level. And I guess in discretionary trading, it's kind of ah similar. Yeah, ah to be honest, I would be totally fine with that analogy because it really is that. is ah
00:24:28
Speaker
Like the system, if I gave, ah ah so if I gave someone like I put, I grabbed my code into the file, give them the entire strategy there just to press a button each day or runs automated, they'll probably make money if they, it it all lands on them. Like they they have to follow their, they they can't intervene. Like if they run every day, they'll make the same money as I am or percentage returns.
00:24:49
Speaker
But. if ah If I'm a discretionary trader and I'm going to explain to someone my intuition about politics, about macroeconomics and everything like that, I think that's way more complicated to do. And yeah, it becomes like kind of a sport. You have to you have to understand the call of this information, be inside the game all the time and try to make decisions.
00:25:10
Speaker
in spite of volatility and risk and everything. It's a really hard game. And to be honest, there's a subsection of and there's there's a subsection of like systematic one traders that I would include in that like ah athlete analogy because there's a difference and we can get into it if you want. Like there's a difference between risk premium like type harvesting. stuff that is very well studied that you can harvest that you can do and the quants or systematic traders that really focus on alpha and edge and those quants that are focused on alpha and edge those are those type of athlete that and knowledge that you are talking about because they are they have to be on top of their game exploring new platforms new markets all the time, trying to understand interactions between different players and developing new systems all the time to be competitive. It's like you are talking about edges that maybe ah come around like that are, how can I say, like and that are economically viable for a few months or a year or whatever. And if you are not always like refreshing, exploring new new places, like you're going to fall behind. So I think there's kind different categories of systematic or quant. And yeah, one is more like a job. you're just doing like harvesting whatever it's there that everybody knows about. And then there's the other guys that are going for alpha and new things.
00:26:35
Speaker
I think that's a very good way to put it, actually, because ah yeah what you said, like mostly the the explored edges is something that anyone more or less can do. it But the really generating new alpha yourself is is ah the real difficult and also rewarding part. I was i also wanted to ask, him do you like do any alpha generation in that way? like How does it look when you're trying to pursue new new strategies and new edges?
00:27:01
Speaker
but What do you like to look for? Yeah, sure. So in terms of alpha generation, I'm going to be honest, it's not ah right now where I'm in in this point of my career is it's like not notch much ah has not been a big focus, even though I've done some few alpha trades, especially now on on hip tree markets.
00:27:19
Speaker
And just to just to clarify, like ah an an alpha, I define as something that is like um a source of return that is novel that cannot be explained by the common risk factors or that other people have already explained so something that is novel that cannot be explained by those common risk factors so for example if if i make 100 returns in a year but those 100 returns are all explained by momentum factors i have no alpha right um so that that's how i define alpha and yet i've increasingly especially this year spent more and more time thinking about new edges and real sources of alpha. But at the same time, I try to keep the risk premium harvesting, which is what I've done well over the years, as the base for keeping the lights on, basically. So I know that these...
00:28:12
Speaker
part of my trading works pays my bills i'm going to keep it active i'm going to keep it improving and i'm spending an increasingly amount of time exploring new markets and new ideas and trying to get those sort of edges as well as part of my book but once you are scaling, like after a certain point in terms of strategies and everything, it it starts become really hard to maintain everything at the same time. Now with like, yeah now I've built a solid infrastructure where ah all of my like systematic risk pre-harvesting stuff is very, um, systematized and doesn't give me any issues on a daily basis. So that's almost taken care of and I can focus on other stuff, but you need to make sure that that base is solid ah before you try to also focus on other stuff, unless you are just strictly focusing on that because yeah, it becomes a ah messy job if if you're doing too much at the same time.
Market Events and Manual Execution
00:29:10
Speaker
I've never really done any risk-premier harvesting strategies myself, so I'm really interested in how those were impacted by October 10th and like afterwards.
00:29:21
Speaker
did Did it change since before, or is it pretty much like the same? Yeah, it really depends on what you're doing. So if you are doing basic like long short momentum style, you were it during that period of time um ah during that period of time, you were basically ah ah almost net short for most trends or momentum systems. So you might have got ADL or or whatever, but you did not lose money ah at that day. You actually would have made a decent return or be neutral at the very least.
00:29:53
Speaker
And so that day was was decently good if you were running like traditional long, short models. ah After that, it depends on the systems. Like I see some some um very rudimentary ah type systems that ah that do well, but nothing incredible after because they are not short and everything is going down. They do well. But there's other variations that I actually am working on and I have deployed that are doing really, really well. For example, my my portfolio right now, the the systematic side of my portfolio is at all time highs since I launched these these few models in and those I've been doing. But and but they need a bit of work ah there. it's It's not just taking a standard vanilla like type model and applying it here. You need a little bit of work and yeah, but ah but but it depends on the risk premium harvesting.
00:30:44
Speaker
I also find it very interesting that you said that you do most of your execution manually because I wouldn't really have expected that. I thought if you're more more of a systematic trader that you also execute systematically. How how can I imagine that? Like, do you just, ah when you see your system every day gives you this signal, you just like go on all these exchanges and press market buy on all your positions or?
00:31:05
Speaker
Um, and no. no and Okay. So, um, for the bulk of my positions, like on that I have on hyperliquid extended that all of these have like a solid API's and everything like that. I've designed like my, uh,
00:31:20
Speaker
portfolio management app or infrastructure, whatever one we call it, ah where I just press a button and it sends all my exposure. so if So every day I have a table, like a summary table of all the exposures I should have versus what I actually have on the exchange. So for example, I need to be $10,000 hype, whatever. and I'm like $5,000 long hype.
00:31:45
Speaker
My dashboard on that day says to me, look, you are $5,000 away from target. You need to buy $5,000 of hype in order to get you your um your position to what is the target position that you want to have. and then all of these position all of these adjustments are sent to like um on the backend to to a simple execution engine and the execution engine then sends all of these orders through their ah corresponding um exchanges.
00:32:14
Speaker
um It's not me going to add the exchange, just an execution engine that I've built since all these orders there. I just have to click a button. I say that is manual because it's still me clicking the button like push to to execution engine. But at the same at the same time, it's then all handled by my code that sends to the exchanges. Now, there's a cave there's ah the small variation, for example, on on on new exchanges like for variational that that I'm farming right now. They only have a read-only API, if I'm not mistaken. And so, i have to there I have to go to an exchange and execute. And I only do that because I think that the points that I'm farming there are worth my time right now.
00:32:52
Speaker
ah Otherwise, I wouldn't do it. But yeah. So it's basically kind of like a simple terminal that's like very specified to like your exact needs kind of. Exactly. Exactly. I see. And and and the targets the do the target sizes every day are determined then by my edge or my perceived edge that I have on the market, right? If my models tell me to go long, I should expect to have edge over the long term and they tell me how much longer I should be based on volatility, based on correlations and then that all all that math is handled on the backend and tells me, okay, this is what you should be long and it compares to my exchange.
00:33:32
Speaker
Do you go any more like granular into the execution itself? Like do you use like TWAPs or anything like that or just normal orders for execution? A good question. Like a few months ago when I like, how can I say like for, for, ah for normal activity, um i don't, I haven't really had the need so far and if until a few months ago to, to have done that. But as I, like I said, like I'm so starting to open my book a lot more to alpha strategies and stuff like that. I've started going into more niche markets. And that means that most of these markets are more illiquid. And so um I was executing ah very naively ah into heap three markets and I was trading equities there. And so ah blasting market orders there are into an illiquid order book. When I when i put the the size of that strategy like into the six figures, like it that was dumb. I was bleeding like 200 or a day
00:34:33
Speaker
ah just on like, sleep which just on being dumb like an ape and and sending like a market order to the order book. And so what I did there is ah this execution engine that I was telling you about. So when I press that button, instead of sending directly to the exchange, it first goes into an an execution engine that that handles all those orders. And It takes data from the order book of these exchanges. So it tells me the depth of the order book. So, ah for example, when execute a $10,000 position, okay, how many levels like on the order book do I have to cross um if I wanted to issue just a market order on BTC might be fine. on Palantir on a heap tree market from XYZ might not be a good decision. And so the execution engine then transmits it into like a north of execution check. And if it's in, if the slippage is above my perceived edge,
00:35:26
Speaker
ah it sends us either to a TWOP or to something else i'm I'm working on right now. But or or but um or I go directly on the order book and I execute it myself. That sometimes I'm better at executing than than whatever al I'm going to build.
00:35:41
Speaker
And it's much faster. Like I just take the $10,000, I go to the book and start messing around with the order book there. But yeah, that's basically it. It's just like a bunch of checks before checking which order we should send.
Future of Trading with Decentralized Exchanges and AI
00:35:55
Speaker
What's your opinion on like perp dexes in general? Are you bullish on their future versus centralized ones? um yeah Man, I have a very, i have a very large position on on Hyperliquy. So yeah, I'm i'm very ah into perp dexes. Now, am I all like ah wishy feelings like, oh this is is going to be the good great now thing and i have no ah criticism? No, of course not. For example, I'm very concerned about the regulatory regime when this like US administration ah steps off and comes another that might not be so lenient toward towards crypto. And there's all of those questions that I'm a bit iffy about, you know. But in terms of technology,
00:36:43
Speaker
if If we could have like a hyper liquid always going forward with no restraints, like I think it's the best solution to create, not the best, but at least the best we have now. Right. I don't want to have all the orders on a like on any a a hidden ah sex book, ah centralized exchange book. I want to have it all centralized and open. And I know there's privacy concerns as well there, but I prefer, like, ah I think that it merges much better with crypto, ah with the ethos of of crypto.
00:37:11
Speaker
Mm-hmm.
00:37:14
Speaker
Do you think like any other perp decks right now is interesting in particular apart from Hyperliquid? ah Yeah, I'm i'm i'm very um much looking forward to see what the extended ah team is doing.
00:37:28
Speaker
I really like their perp decks. I was one of the first users there. I think I started trading there from March, 2025, if I'm not mistaken. I think a bit earlier, but I don't want to exer exaggerate.
00:37:41
Speaker
um And so it's like it's it's I really like to to trade there. ah The liquidity right now it's for what I do. It's fine. um It has handled itself like very well.
00:37:53
Speaker
I think the team is like a ex-revolution did on execution um that they have been doing phenomenally. Like I'm very much looking forward to see what they build next.
00:38:06
Speaker
But that's the only one, to be honest. Lighter is already out and i like them. There's a few like ah things I don't, but anyways, ah all the others I'm still very, I think they they are just, ah variational is, for example, variational is already a different type. They are innovating on their, ah ah how they match orders and everything like that. They are different.
00:38:28
Speaker
ah but So I would put those three like hyperliquid, extended and variational are my like, I'm interesting to see what they do. Yeah. Yeah, i agree. I think it's good if they have like some differentiation for like their prep decks or whatever, like senators with the vault tokenization or variation with their whole new system of RFQ base orders or whatever, however it works, um rather than just being like Hyperliquid 2 or whatever. Exactly. That's a bit more interesting.
00:38:59
Speaker
they take all the that They take all the exchanges available, mostly quickly, and they just um did just match the orders that way to those exchanges. And that's that's very interesting. And they extend with their vaults also with the potential for, know, this is very speculative, but maybe that Revolut connection might be interesting, all that they are doing. So um exciting times.
00:39:25
Speaker
Do you have any any opinions on Vibe coding, which is also like a very ah popular topic right now as someone that taught himself or how to code? Do you Vibe code?
00:39:37
Speaker
Yeah. i do like I use cloud code ah Cloud Code and other LLMs to code, but ah It depends on how people define like a vibe coding. Are people just putting like the cloud whatever they use to automatically make code? That's dumb. That's very bad. that They'll probably make a very big mistake very soon, especially trading. um and Now, if you use that to ah suggest you code, you read the code, like you understand what it's doing, like then it's a very helpful guide so that you don't have to be searching like like before, like Stack Overflow or coding yourself from from scratch. I do all of my code through through cloud Code right now and everything works fine. But it's like I said, like a very strict, very strict
00:40:27
Speaker
ah controls on when to push code. I'm not just pushing like accept, accept, accept. Like i'm I'm checking the code on each iteration. I'm checking if it makes sense, if it's well written. i was never ah like there's people that are legit like developers and understand what's a code very well written. I think I was very average, always was like i learned from myself just to make my trading automated. But I do think that i have a decent like understanding of what's good code and I always advise people to understand what the code like the LLM is pushing. Yes, you can vibe code.
00:41:01
Speaker
You can, like I said, check everything, but be careful. Don't don't just push blindly because it does a lot, a lot of mistakes. that You would not believe the stuff that I'm getting that I'm getting suggested sometimes ah like defaulting to leverage. Like if if my leverage uh control uh is not um if if there is a mistake on the leverage control like an error on the code it it ah defaults to something that i don't want something like that yeah some crazy stuff like you need to be paying attention to that code
00:41:37
Speaker
Do think AI will ever be good at trading itself? Will it make markets more efficient and then squeeze out more alpha and edge and replace us humans as traders?
00:41:50
Speaker
To be honest, ah that's really like ah out of my out of my competency ah to to give a very strong answer about. ah It depends. like um If you... because there's... I think there's very... a lot of layers to that question. So if you give it like...
00:42:11
Speaker
ah But even then, why would it be useful? i was going to say like if you give him a bunch of risk premium strategies to run to run itself, OK, AI is there, but you could do it that yourself right now. If you put it on on the running on a price data or other sources of data, like how are you going to ah how is it going to go around the noisy data problem unless you give it so specific instructions on what to do that? Maybe you find some edge, even though it's very speculative to say that you can find edge from just price alone on the major assets. um And since you are giving it so such um detailed rules on what to do, was it, he what is it useful for anyways? And you can see even on these, um these tests,
00:43:00
Speaker
I don't know the name, but they're running a bunch of AI models on on trading on Hyperliquid or whatever. yeah The arena, alpha arena, whatever. And you see, it's mostly like a random walk. Like do some make money, others don't. Seems like most of ah of a coin flips. Some then invert, others go up. ah So it's really, you trading ad? Are you trading noise? its It all comes down to that philosophy, I think, that we talked about at the start. Like if you don't understand what you're doing, if you don't understand why you're getting paid for your exposure,
00:43:30
Speaker
How are you going to justify like ah ah two months or three months or ah like a big drawdown? How are you going to justify that? How do you know ah your edge is still there versus ah you never had edge to begin with and it was just a coin flip?
00:43:47
Speaker
You know, because if you if you make if you make a simple simulation of this, if if you like ah run 100 or 1000 or 100000 like simulations of coin flips, there's a significant amount of of coin flips that coin flip equity curve that will win a lot of money and very consistent and very high sharp. Does it mean that they had the edge? No, did so they just land landed on that on that lucky ah ah part of the group that made money. And so I always go back and tell people, understand what you're like, what you are harvesting there, because if not, it's going to be art to to in the future.
00:44:23
Speaker
Yeah, I think that that actually makes a lot of sense to to view it in that way. That it's a bit more like, I guess we have to to wait until AGI appears for trading
Long-term Views on Crypto and Networking Advice
00:44:32
Speaker
to be solved or whatever. But until then, we we still have time. But that leads me to another question, actually, which is um since right now we're in a pretty pretty grim phase of crypto and the market in general, some some might say that we are in a bear market, in in a very deep bear market. And also since say crypto or Bitcoin in general has all underperformed gold and everything else by a lot in the last couple of months um and weeks, do you have any any perspective on the the future of crypto markets? Is there anything that you do even care about there? Do you just like let your systems run or do you think like it's going to get better again? Do you have hope for the future? Are you saying it's over and you're going to move back to TradFi and focus on stocks again?
00:45:17
Speaker
What do you think about that? Oh man, ah I would like to try and sound smart and say I have a plan, but I don't like, I don't know what will happen in crypto, to be honest with you. Do I think it recovers? Historically it has, and it has had very strong periods, in even in 2023 to 2024 and late 2024, then there there's legitimate periods of speculation in crypto. And I think that's always going to be there. and um Yeah, if I had to bet, if I had to make a bet, i i would bet that periods of speculation in crypto will always be there. Now, the question is, are we going to see like old coin markets cycles like we've seen in the past? And to be honest, that is that is very uncertain to me because now you don't have like 200 or 400 coins like you add in whatever or Now you have not even that, I think way more. And all of them are grifts. So yeah ah to me, you're you're going to see how grifts continue to go down and hopefully like projects ah good, decent projects with the revenue, to and a good tokenomics or some, at least some...
00:46:27
Speaker
ah some utility ah to continue to perform well in going forward. But I don't think you'll see like um we continue to see like everything pumps and everything you touch makes money. I think it's going to be more selective. And it you made a second question there if I would go back to try it right now. No, like you have a lot of liquidity and volume in crypto and volatility. There's a lot of returns to be made still, even though ah despite being so down. ah So as long as there's volatility and volume, I continue to be here and I'll reassess as we go forward.
00:47:05
Speaker
I think that's a good perspective. Yeah, that makes sense. It's not really like, even though it feels kind of that um maybe ideologically or like just ah from from a psychological perspective, the sentiment is very bad. i think volumes are still pretty good, especially compared to historical periods where the market has been down quite a lot, where activity was just kind of like that.
00:47:28
Speaker
Yeah. I wasn't trading crypto back then, but I know I remember people saying about the 2008 and late 2018 and 2019 period that it was like dreadful, like there was no one trading. So those are the times that, yeah, maybe I would consider shifting some allocation elsewhere. But right now, like there's money to be made in both ways. You just have to figure out ways. Right. And it's just.
00:47:52
Speaker
where volatility is shifting. Sorry, where volume is shifting, where is volatility like even in in in like in in prediction markets, there's so much to be done there that I'm looking into every day. There's so much volume there. And with that amount of volume comes down volume.
00:48:09
Speaker
And so um there's a lot to be done. There's a lot of places to be in right now. And ah yeah, the ah up up only market is not the only market that you can make money. That's what I'm trying to say.
00:48:22
Speaker
have you found anything interesting in uh prediction markets yet
00:48:28
Speaker
um if i found anything interesting without giving away your alpha yeah i don't yeah i don't wanna uh i'll comment it on on a future on the future right yeah but just look at it we'll keep an open mind would say keep an open mind there look into into those markets and see what you can find there Nice, nice, jeez. Alpha, jeez.
00:48:54
Speaker
Yeah. um Do you have any advice? I think how I initially got interested in you is because you wrote a really big article about a bunch of lessons that you learned while over the years of trading and also explaining your story kind of like you did here. But I'm curious if you also have maybe um Maybe some advice from that article or also like some more specific things that you could like tell people if they're just starting out and want to go down a similar path that you are on.
00:49:31
Speaker
To be honest, like I've like i've i've changed my view on on that over the years, but I think that like 80%, let's assume that that you really like trading and that you have a real deep interest because if you don't, if you are here for the money, like to make a quick buck, you're not going to last long. You could you will do well doing a business, solving some issue, like creating a SaaS or whatever. You'll do far better for less stress. Understand this is a very tough game and I don't want to make light of it. so So consider, do you like trading or do you like whatever you've been sold about trading? Because it's not real. And um yeah, ah so in terms of the best advice, if somebody passes that filter is, i think, network. Like...
00:50:17
Speaker
ah having the correct people to tell you the um the specific knowledge of stuff that makes sense. I'm not saying people that will give you edge. No one, at least if unless it's your friend, no one is going to give you like real alpha, but people will give you like real good advice. And and if I may, ah if I may say some names here, um like for example, Robert James that publishes a lot about this, like Scott Phillips on on Twitter, like You have i man from the top of my head, people that actually publish like useful stuff, not just good trade because good traders has a lot. I would go to my list, my profile and see my list of people I follow. And I try to only follow people that I that i respect or at least give some interesting stuff. um
00:51:04
Speaker
Curating your network is super important. Because if you are spending years on the wrong information, it's just going to waste time. And I've wasted so many years on bad information because I was not on Twitter. I was not so looking for ah people that was actually doing this professionally. I was just on my own bubble of grifters.
00:51:23
Speaker
And that just wasted a lot of time. And so I would say that networking is the most important. See people that are doing that well and ah focus on them. Learn everything you can about them. Don't let them go and go from there and spend your like your interest, your curiosity on stuff that they've created themselves.
00:51:41
Speaker
What's the best trading advice that you've ever received?
00:51:46
Speaker
What's the best trading advice that I've ever received? um Don't try to be smart. um Do useful things. um Don't bet too much. Don't bet too little either.
00:51:59
Speaker
um And trade less.
00:52:05
Speaker
That's all all very good advice. Probably and and and nothing that the that a beginner would listen to, but I think that's all pretty good.
Psychological Challenges and Team Expansion
00:52:15
Speaker
Yeah. you have some Did you experience any psychological changes or any like lifestyle changes? as you Because you also explained from your background that you were like just working a normal job at the beginning and then you were kind of like struggling with trading. And also, I think you talked about how there were some times where you thought about giving up as well because it was just like a sink of money, basically.
00:52:39
Speaker
And um now that you've been experiencing more success and also increasing your size and your position in life in general, I guess, how has that impacted you?
00:52:54
Speaker
ah Oh man, ah to be honest with you, i live the same life as i've as I've lived like five years ago, to be honest. It's like i keep a very simple life. I don't... ah The only thing I spend that I spend on stuff more is like on my... computer on my on some vacations nothing too extravagant but I don't I haven't changed my lifestyle one thing but in terms of like psychological side that's one thing that I always make fun of and I continue to do so but because most of like trading psychology you see online is a grift but ah on the side that is not like when you are already making money and and you start increasing size to levels that you have never been before again that stuff starts to impact you ah like it's like that stuff starts to impact you. And that's something I like I had to work around with and to try to adapt. You you adapt to it, but it takes time. And so don't don't let stress like take over your entire life.
00:53:54
Speaker
um If you are feeling like it's too like emotional when when when the volatility in the market is too high maybe size down um do something just don't let it affect your decisions otherwise if you are letting that stuff external things affect your decision making in trading or in markets or in business in whatever you're doing in life like that's that's not you have to be like stoic you have to be you you have to be emotional so you have to understand that money is just a tool that you use to express your views on the market and to be ah paid for that um
00:54:27
Speaker
Oh, hopefully. And so don't like if you start thinking, oh, I could have bought a house with this with this drawdown, I could have bought this like you're you're going to struggle. yeah You really have to adapt to that. And so emotional control, psychological side starts to become an issue when you increase size. So understand that. And yeah, it takes time.
00:54:48
Speaker
Yeah, I guess it's something that can only really be fixed by, or if you can call it fixed, but like ah that you can become better at by spending time on it and also experience. I don't think there's any like shortcuts where you can just be like, hey, this is ah more money than you've ever seen in your life. Just don't feel anything.
00:55:08
Speaker
Ignore it. Exactly. is It's like that thing like, um yeah, I would imagine I'm not, but like a soccer player or UFC guy like training in the gym and going to to be standing in front of millions of people like watching them. Right.
00:55:22
Speaker
ah It's two different pieces. Like you can see on a back test or whatever you do, like, oh, it's in the piano swings by a hundred thousand dollars in three months. I can do that. But then you do that and you go through that to the uncertainty and everyone around you saying, oh, the market is bad or this is never going to recover again, like all that stuff. And you start to second guessing yourself. And when you start to get second guessing yourself and your own work, it's when you probably are close to losing the bottle.
00:55:49
Speaker
hmm. I think it's not even only about size, it's also a little bit about um just trading in general with real money. Like if you only do paper trading or backtesting or whatever, even if you like ah do a real trade and you only lose like $10 or something, of course, the depending relative to your net worth, how much it is or whatever. But I think it it always kind of like, especially at the beginning when you don't really know what you're doing and you don't really know if it's going to succeed, it kind of hurts to lose money, even even if it's like very little.
00:56:19
Speaker
For sure. It's like, and then you start comparing to your like real life stuff. Like for example, the other I was buying something and I was like worrying about a $200 discount and I'm like, man, that, that, like I'm worrying about 200, but that's what i I spend on fees, you know, on a trade. And they're like, okay, but, but you still have to keep that. You don't want to be like, ah yeah yeah, you don't want to be like splurging just because you, you spend it. That's two different things. Yeah. ah But yeah, it's it's always an emotional battle, and especially because training is a training is a business
00:56:55
Speaker
of speculating on the uncertain, right? Even if you have a model, like even if you have a very good model, whatever, you're still speculating that that model will have edge tomorrow, right?
00:57:07
Speaker
That whatever you do is still going to work tomorrow. You're speculating on the market, speculating on that. And so everything is uncertainty. Like this job is not like, oh by the end of the month, I have a paycheck on my account.
00:57:20
Speaker
It's not that it's like maybe I make money this year. Maybe not. ah We'll see. You know, maybe I make money next year. It's like it's very uncertainty. You have you are betting on the uncertainty. You have no one to respond to only yourself ah or in most cases if you're not at the firm or whatever. But Yeah. Most people cannot deal with that. it's and It's not just the money. It's like the money coupled with the uncertainty and all the emotions that can come up with it. Over the years, I've been able to, I think, control that better, but a lot of people cannot deal with the uncertainty that we do at this shop.
00:57:55
Speaker
Mm-hmm. Do you ever plan on joining a firm or like expanding your own team ah sir as your current growth or your operation? To be honest, I don't think that I have the technical capability of joining like a firm these days because it's so competitive and they look for ah kids with college degrees and stuff like that um with strong mathematics or physics. I have a high school diploma, like I said, like I i do not qualify.
00:58:25
Speaker
ah But in terms of expanding my team, yes, I want to do that in the next few years because I want to start. um Yeah, I have a bunch of plans and I would like to start focusing on and on new edges, like i was saying, and I need people to manage whatever I have ah still here and all the coding and the infrastructure that I've built. I want people to be managing that. So, yes, ah ah my plan is to expand my team in the future and and and be able to scale that ah better.
00:58:53
Speaker
Like i see so yeah in America would would it be called like a a family office or something like that. yes You're just managing your own money or your family's money. Yeah, that sounds good. That makes sense.
00:59:05
Speaker
Yeah. I think that's that's a good time to to wrap it up. um Do you have any any final words of wisdom or anything that you want to shill at the end?
00:59:17
Speaker
No shill, but if I have a final word, it's like trading is hard, man. You can do very well in in in other stuff, like do other stuff. But if you are really interested in this, prepare yourself, strap yourself to the seat. It's going to be a hard road ahead.
00:59:34
Speaker
That's a good way way to close. Thank you very much for for coming on and the goodbye, everyone. and No worries, man. Thank you for the for the invite. Of course.