
Interview with Kosta Ouzas (CEO) and Sebastian Raphael (CTO) of Minotaur Trading Systems on 'smart bots'—automated trading strategies that combine human-developed quantitative strategies with automation and AI/ML assistance. They position smart bots as distinct from simple grid/dumb bots by emphasizing quant-developed code, robust risk management, and emotion removal for retail investors. The Minotaur Maximizer (flagship for Bitcoin) is cited as having a 95% win rate, ~15,000 lines of code, layering ~15 indicators, and running 24/7. The product targets cryptocurrency spot markets (with optional futures), aims for plug-and-play retail deployment, supports multiple exchanges, is rolling out internationally (US-heavy user base, plus UK, EU, Australia), and claims clients keep 100% of profits while Minotaur provides code and consulting.
Chapters
Okay, so I've got CEO Costa Ozas and the CTO Sebastian Raphael from Minotaur Trading Systems on right now and looking forward to having this conversation, you guys. Costas, I'm going to start with you. I mean, the idea of smart bots, which is what you guys are launching, you know, an automated trading strategy, is getting a lot of conversation lately. I mean, obviously, you know, we're going to see only more of it as AI moves in. And I mean, I don't think you can ever fully replace the value of human insight, intuition, but you can really make that more efficient. I mean, for me, I hear smartbot, and I'm like, I want that. I want to do that, make it a little easier for me. I'll take responsibility for my choices, but sort of like, give me a better menu of opportunities here. How does it work? You?
Yeah, it's a good question. I think the idea. The idea that we've come up with smart bots is exactly what you said. It's not minimizing the human input, it's actually maximizing the human input on the strategy itself. There's a proliferation of grid bots on the marketplace that if we're going to call what we do smart bots, we could call them dumb bots. Really what they're doing is they're just a flywheel of buying low and selling high, you know, accumulating lower positions and selling them at a certain, you know, whatever that grid percentage is. And it's very mechanical and there's not much thinking that goes behind it. And eventually, you know, having tested a lot of these strategies in the past, before we built this company, what can happen if you have black swan moments or big retracements in the market is they just. They just. They just fall apart. They can't keep up. So our smart bots are actually developed by quant traders. So we've got a team of traders who are really developing the strategies from scratch. And these guys have used these strategies effectively in the past. What effectively we're doing is we're connecting them then to the bot part, which is the automated trading. So it's giving us the best of two worlds. The smart. It's coming from the quant developers themselves, coming up with the trading strategies, and then the automation part. So connecting those two dots is what makes our system so powerful.
So before I get to you, Sebastian, let me just ask you, based on what you just said, and I want to differentiate AI, pure AI trading from what you're talking about. So theoretically, you could have an AI that took, just took, looked around, you could give it parameters, took the best 150 traders, the best 10,000 traders, the last 5 million trades, whatever you want, and give me the best likely outcome. What it sounds like you're talking about is something that's a little more built on taking a group of traders, maximizing their insights, and automating your abilities. So instead of having to call those 50 people, or 20, whatever it is, you'd be able to algorithmically make better choices. Is that accurate or am I getting it wrong?
Yeah, I think so. And Sebastian can elaborate a bit more on that. But, you know, the, the developers are using elements of AI as well to test their strategies, you know, machine learning, all that kind of thing. But effectively what they're doing is they're building the strategies themselves with the assistance, if they need it, of the AI element in order to optimize that strategy. Rather than giving AI 100% of the responsibility, is we're actually taking on 90 to 95% of the responsibility and simply using any AI elements as a tool to enhance that performance. And of course, when it's trading elements of machine learning as well.
So, Sebastian, from a practical point of view, as a user, if I'm a retail investor, say, how does it work for me?
Yeah, and that's a great question as well. So traditionally, if you're a retail investor, if you're looking down the barrel of trying to understand and learn manual trading, this is something that, although it obviously has value and you can learn an edge on the market, it's traditionally something that can either cost large sums of money or take a very, very long time to learn and understand. And for a retail investor, let's be honest here, most people aren't walking around with large amounts of capital to invest into the market. So, you know, what we're trying to do essentially is bridge that gap and allow people to come in in a way that is going to remove the amount of time it takes to learn how to do this and just come in and deploy something that is as plug and play as possible. Now, I don't want to make it sound like it's as black and white as clicking a button and then printing money, because it's not. And that's unrealistic. But it's more about getting people into an environment where they can create consistency in the financial market. Now, in my opinion, the biggest problem with manual trading, it's not necessarily the time it takes to learn. It's learning how to actually manage your emotions and Then allow, you know, what you've learned to create consistency in a market, which is very, very, very difficult. So that's essentially what we focus on and what we're trying to do. And it's really the risk management side of things, in my opinion, where these algorithms shine. Because you could even have a decent strategy as a manual trader. But if you don't have the emotional capacity to execute that on a consistent basis, what you're going to find is months where you do well and then months where you give all your money back to the market. And at the end of the day that's not sustainable for long periods of time. So when we develop this, it was always with risk management, sustainability and giving people consistency in the markets. That's what we focus on and that's kind of what we do.
Yeah, you know, it's interesting. I mean, I know. Well, so I see two problems. One is trading's complicated and if you're not a pro at it, you need someone to help you. So maybe you have a friend who does it, you know, maybe you're like me and you get to interview really smart people all day long and you can pick their brains. You know, you use E trade and you have a person who helps you, but they're not gonna tell you what to do, they're gonna execut execute trades you ask them to make. I find even myself, even as something as simple as buying bitcoin, my emotions are constant. To Sebastian's point, I'm constantly at war with my emotions. It's 4am in the morning, it's a 24 hour clock and I see what's happening and I go back and I look at what happened over the last hour and over the last day and I take all the inputs and I'm like, I don't know, I think maybe, maybe I should buy now, you know, And I'm like, you know, and then I hear in my head that people are like, rob dollar cost average, dollar cost average. You know, just like buy a certain amount every week. Don't be Mr.
Impulse.
I'm like, no, I'm gonna do it right now. You know, and then I sometimes I'm like, wow, what a great call. And the other times I'm like, wow, that was a really bad idea. I just lost that. I just lost that investment. Now in the end, I'm gonna stay with it. But to the point of emotion, like, it is really hard to not let your emotions, your feeling get in the way. It's really hard to step back from It.
Yeah, it is absolutely right. And to your point, you go, Costa.
So, Sebastian, I was just going to say, pretty much what you said is what we've been able to codify and build into the code itself. So, you know, one of our strategies, the flagship on the Minotaur Maximizer, which is the best to simply just trade on bitcoin, has a 95% win rate. This thing has about 15,000 lines of code and it's also running 24 7. So everything that you just said, when should I buy, when should I sell? When is the right time? This can do all that for you. It's all built in. I think it layers in 15 different types of indicators in order to be buying and accumulating positions just at that perfect moment. And again, it's watching this 24 7. So all that stress that you were just talking about, it's all gone when you plug into our system and it's literally all done for you. But again, you've got to remember that it's not done just purely through AI, it's done through codifying the strategies that the traders have developed. And then we're using that to plug into the machine, learning to then get those results. And, sorry, Sebastian can then elaborate a bit more.
I was just more or less going to say to Rob's point, like, at the end of the day, that feeling you feel and that emotion and, you know, this could be a controversial take, but I truly believe that's something that can never be taken away. At the end of the day, we're human, right? And emotion is built into our DNA. One of the reasons I got so excited to get into this space is because I suffered from similar problems. Everyone does, even quantitative traders and quant analysts who we work with, they're not perfect. So, you know, at the end of the day, it's about building something that can remove the emotion as much as possible and get a result. This by no means, means that algorithmic trading and quant analysis is perfect because it's not right. We are trading a market that's highly volatile, has a lot of different things going on, albeit I say some manipulation. So you're never going to create something that's perfect, but that removal of the emotion is where you get the results and the consistency.
Yeah, no, and I really like that there's a human component in it. I think that's ultimately such a great use of the technology to take. To take great human insight and build an AI model around that to maximize it and then make it efficient. Can you Trade, what's the range of what you can trade in and the option tools? I mean, can you buy crypto, buy tradfi stocks, futures, options? What can you and can't you do on it?
So what you'd be looking at with the product and the service offering we have is simply cryptocurrency markets, only for a couple of reasons. First of all, when we first went into development, we considered commodities, stocks, currencies, these sorts of things, but they didn't have the volatility or the movement that we wanted. Now another thing I kind of want to touch on, another reason we only use cryptocurrency markets is we wanted to create something that could kind of remove this air of high leverage returns that a lot of people are looking for. So, you know, in stock options, futures trading, in forex trading, so many people are over leveraging themselves using like 100 to 200x leverage, essentially borrowing against the collateral and assets they have in their account and then trying to get large returns. Now, although this can work in some scenarios, if you actually look at the historical data, you'll 99% of the time give all your money back to the market. What we actually focus on is spot markets, meaning essentially trading without any leverage whatsoever. We do offer the ability to trade cryptocurrency futures, and that is part of our service offering. But we wanted to create something that was viable in spot markets. Now the reason, once again, I know I'm harping on about emotion a lot, but of course already we know traders have issues with emotion. And when you have the ability to go and leverage and borrow against the assets, you have to try and get an increased return. This creates a huge myriad of problems. So for us, it was developing something on spot cryptocurrency markets because at the end of the day, you can't borrow and you're stuck to simply trading the assets you have in your account, which in my opinion is a very, very good thing. But people don't like this, right? Because inherently, once again, as humans, what we want to do is get rich as fast as possible, make money as fast as possible. That's not how trading works. You speak to anyone who has an inkling of how the industry works and how a chart moves and how price action moves and volume, it doesn't work like that. And if people can kind of gear themselves to just understanding, hey, if I can just create consistency over a long period of time, you're going to get the same result. But there's no making money and getting rich overnight. And that's not what we're selling and not what we're trying to do, either.
Yeah, no, that's a really good point. I want to ask you a couple, like, just a couple specific questions about how someone would use this and access it. But first, you know, we started this out, you know, with the conversation around the bots, you know, and Costa, you made the point about, you know, these sort of dumb bots. So I don't know the answer to this. So I'm going to ask you guys this. I saw this last night, which is funny because I didn't know it was unrelated to this conversation. But Visa and somebody else said basically they're tracking. Excuse me, Visa and somebody else are tracking, you know, USDT Circle, USDC, USDP and PayPal's coin. And they said most. They're all pegged to the dollar, obviously, that they see a lot of inorganic activity, and basically a lot of the stablecoin transactions that they see processed are not coming from human activity. Is that, Costa, what you mean by dumb bots? I mean, they're just, in other words, executing their trades. They're just executing based on a thing. You know, they're not saying it's fake. They're just saying smart contracts, instead of humans, are driving this stablecoin arbitrage and liquidity.
Yeah, so what I was referring to, more so when I was referring to, let's say the dumbot or the basic strategy, is that there's just. There's just no. There's no underlying strategy. There's no code required. It's just simply based on, you know, a percent between the buy and the low. You know, buy low and sell high, and then that's just a flywheel that's happening over and over again. So, you know, that doesn't require any kind of, you know, quantitative analysis, any kind of code. So, you know, moving from that to a smart bot smartbot is something that does require that quantitative analysis. It does require a script to run it. And that's a script that's developed by a very smart human being, and that's why we're calling it smart.
So, Sebastian, sort of similar question to you, just because I know people see this and hear it, and I honestly didn't fully understand, like, why would Visa care that smart contracts are driving the stablecoin market space? I mean, is that a bad thing? Is that a neutral thing?
I think it's. I think, like, it's a really good question. Personally, I feel neutral on it. And to answer your question, essentially, what you're more or less talking about is arbitrage, which is essentially, you know, pre programmed algorithms, you could say, that look to actually, you know, buy at different prices on different exchanges. Now this has been going on since the inception of cryptocurrency. There was arbitrage as early as 2009, 2010, during Bitcoin's inception. So what we're talking about is essentially something that has been happening for a very, very long amount of time. And I can't really speak too much on where Visa would stand with that. But you know, it's definitely, it's definitely an interesting point. And look, at the end of the day, yes, the crypto market and transactions and smart contracts are driven by algorithmic technology and AI and bots and these sorts of things. But I think it comes back to the point of the way I've always looked at this market, to me has been a financial market driven by the future and driven by technology. And I think that's where we're going. And I don't think that at the end of the day, 10, 20 years down the line, this market is going to be only human interactions trading. I think where we're headed is algorithms, and I think it's going to continue that way. And I think it makes up a large quantity of the market's transaction. So you're definitely right there where Visa stands on that. I can't answer it, but you know, this is the way the market is going and this is where we're headed.
Yeah, I think that's the relevant point you're making and thank you. That helps me understand it. I think that intuitively makes sense to me. That's where we're going. That's where this technology is going. You can fight it or you can figure out how to wrap your arms around it and make it more useful. And I like where we started Coastal with the idea that this isn't saying we're replacing human intelligence. I say this with a lot of AI. You know, I don't care if kids go to school and learn math if they can, or foreign language, if there's a chip that can tell them what they need to know. I'd rather they take that information and do something really intelligent with it to, you know, actually maximize their ability now to do things with that knowledge versus whether they did the hard work to earn it, just like I did when I was a kid. You know, I feel like that's where we're at. Like, I don't want to have to work harder. I Want this to help me so I can actually use my brain and do more with it. It seems like that's kind of what you're talking about with the smart buy.
Yeah, so that's exactly right. So you know the tool allows you to take your analysis to the next level. Right. So you can, you can only get either so far so fast, and using that tool is either going to help you get further or just get there quicker. It doesn't replace the intelligence of a human being who spent years or even decades in a field, learning that field, studying that field, and then also someone who has the experience of doing live work in that field. Plugging into that AI is just allowing them to enhance everything that already exists. You can't get someone who has no experience or no idea what they're doing to just ask the AI to build them a strategy. It's just not going to work. We need to be engineering the prompt in a very sophisticated way. We need to know exactly what's the end goal of our strategy. And the AI can simply help us to achieve it, or sometimes it just helps us plug in the holes or find the blind spots. And we can do that through the initial development stages, or we can do that through the machine learning stages. So there's a lot that it can do to help us. It's like moving from a screwdriver to an electric drill. It's just allowing us to be more efficient. But you still need to have the builder behind it.
Yeah, you still got to put the thing up to the wall and put in the.
Right, that's right.
So, last question before I let you guys go. Is this available? I know this is. You're launching it, you're announcing the smartbots who can get them. What do people do? Like, if you're a viewer or listener to this and you're like, well, wait, I'm interested. What, what do they do and can they get it anywhere?
That's a really good question. So I guess we've been in sort of the beta stages for a while. You know, we have launched to the public, but what we're going to be doing is really rolling this out now, internationally. A majority of our customer base is from the usa, but we do also cover a lot of the European market, the UK market as well, some of the Australian market. And basically, you know, people are interested to get involved. They can simply, you know, book a call, they can Google the company or perhaps there's going to be a link somewhere in this video. But book a call and get a demo we do complimentary demo sessions. So basically you can book a call, you can get in touch with one of our specialists. They're going to basically talk you through the strategies, how they work, and how they're customizable as well, depending on the needs and the preferences that the client has. That's another thing that we have. There are settings that do allow us to customize the strategies based on your risk profiles or simply sometimes just based on what assets the clients want to use. But we are going to be available internationally. We plug into most exchanges. One of the most important things we want to learn is that clients, we're really a software development and consulting company. So anyone who works with us, if they do make profits, they do get to keep 100% of the profits. The gains are all theirs. But all we're doing with the people that join our system is providing them with the underlying code and the strategies and then provide them with the consulting that they may then require in order to deploy those strategies.
Really interesting stuff. I want to see where this all goes. It is the future. So as far as I'm concerned, you guys are part of what. It's going to take a whole lot of new twists and turns as we go down this path, but it's an exciting ride. So I hope people take advantage of it and test it out. And you know, and you know, it makes them not quick rich, but long term. Sebastian, smart rich.
We do too.
Smart rich. He's smart rich. Not quite rich. Great talking to you guys.
Awesome, Rob, really appreciate your time. Thank you so much.










