Ahmed Ismail on improving liquidity in crypto trading

0:01I'm Ashton Addison from Blockfills Capital for Investment Pitch Media. And today on the Crypto Coin Show we have Ahmed Ismail, CEO and President of Fluid Liquidity. Ahmed, welcome to the show and thank you for taking the time today. Thank you, Ashton. Good to Good to see you here today. You're very welcome. Let's dive into, you know, the inner workings of

0:21cryptocurrency trading. How do we do market making and liquidity? And I think understanding this aspect of the industry is very important, especially if you're doing trading or if you have a project and you're and you want to ensure that you have liquidity because without liquidity it's very hard to trade. So, I'm excited to dive into what your team has been

0:40working on. Let's just kick it off with a high-level intro on Fluid, the solutions that your team has built, and how that relates to the blockchain industry. And then we'll dive into all the details. Excellent. So, a little bit about me. I'm Ahmed. I'm the President CEO of Fluid. I have a background in Wall Street. I was on Wall Street as a senior

1:04investment banker for 15 years. And then I left that world to be the co-founder of the first fully regulated OTC in the Middle East called Haven. You know, that company now is a unicorn. It's doing very very well even in this in these times. And then last year I branched out again to set Fluid up. And

1:33one of the reasons I did that was one of the big problems in our industry, as you correctly pointed out, Ashton, is liquidity. Liquidity is the lifeblood of crypto trading. Mhm. Not just crypto, but anything. Equities, FX. Fixed income. Commodities. And without it, you can't trade. Mhm.

2:01Crypto liquidity is super fragmented. It's a trillion-dollar industry that has the same attributes of liquidity to equities about 30 years ago. So, it's a big problem for us to solve. And while there are liquidity aggregators around, none of them are solving the problem. Because if they were, crypto wouldn't be so illiquid.

2:31So, we set it out to be our mission to solve this problem by creating an aggregator. And what is an aggregator? If you In layman's terms, if you go and buy crypto or if you go and buy a flight online, you buy it from an aggregator usually. So, from Skyscanner or Expedia, because you're going to get the best price and a verified way of buying it. And

3:01you will And what that does as well for the airlines is that it keeps competition healthy. It creates It creates liquidity. It creates price tension between Creates Brings all the marketplaces together. And that's effectively what we do as a new liquidity aggregator. We bring all We connect to all the top 10 sectors indexes at the moment, but we're going

3:27to be connecting to the two top 200 sectors indexes to bring everything in one place. Whether that be through an API, through or through a UI. And we have three types of in customers or users, if you like. We have institutional trading houses that would use our product. We have exchanges.

3:54So, they call it not the tier one exchanges like the Binances of the world or the Krakens of the world, but kind of the tier two, three, four exchanges that require a little bit more liquidity, thicken up pairs, get a get better price tension. Mhm. We created an AI-based liquidity aggregator. So, AI features very heavily in what we do. Now, why is

4:21that? And that comes down to the biggest problem within the aggregator industry, and that's latency. So, we use AI to cut through latency. Now, why is that an issue? In a very liquid market, even in crypt such as crypto, even in the most liquid of markets such as BTC, 0.5 seconds of

4:51latency Mhm. equates to around 2% of gap risk. Wow. Right? So, if you're an OTC desk or you're an exchange, and you've got a half second latency round trip, it means that you have to, in order to compensate yourself, you have to price 2% wide, which means you're screwing your customers.

5:18Mhm. You're probably not doing that intentionally, but you have to, otherwise you can you know, you can lose out. So, this is why crypto is highly illiquid. Mhm. One of the big problems is latency. So, we cut through that. We have a We invested very heavily in latency, ultra-low latency

5:45infrastructure. So, we've used the same methodologies that are used in traditional finance using things like co-location and you know, simple smart order routing. But, we overlay that with prediction engine that uses AI. Now, this has been something that's kind of we've spent a lot of time and money on.

6:11We have two partnerships with two leading academic institutions, Imperial College in London and the Blekinge Institute of Technology in Sweden. We have around 15 post-doc academic researchers working full-time for us to create hybrid AI models that predict order books on exchanges

6:42With very, very high accuracy. Now, they don't only just use the simple statistical quant methods that most high-frequency trading trading houses use. So, simply you know, you know, you know, you must have heard of you MACDs and Bollinger lines and all of that. That's kind of one feature. But, we also go into very in we go into very deep other features that make our

7:10model very, very deep and accurate to up to 30 seconds in advance ahead of time. So, we can we can accurately predict order books at the moment with up to 30 seconds. This is you know, we have the this is on live data. You'll be able to see that through our MVP. Ahead of time with up to 99.5 percent accuracy.

7:41Wow. That's up to 8 seconds. And then between 8 and 30 seconds, it goes down to 95% accuracy. Now, we don't only just look at As I said, we don't only just look at quant methods, but we also look at training our models with things like sentiment analysis. Mhm. So, we connect our models with things like Twitter

8:09hashtags Mhm. with Discord boards. And that gives a sentiment gives the algorithm a sense of sentiment of what's going on in the world. Mhm. Because as you know, cryptocurrency markets are far from linear. And the big problem with quant quant methods is that they are exactly that. They are linear. There's one input, one output. Or maybe two inputs, one output.

8:37But with AI, the reason we use AI is because it the aspects or the features that affect cryptocurrency markets are far from linear. They're very, very complex. So, at the moment, we look at As I said, sentiment analysis, which is something that you wouldn't know. Like, you wouldn't be able to guess Mhm.

9:02unless you're a trader or brain. You wouldn't think that if someone says, you know, crypto's dead. Right? Would you do that? Right? Do you still Do you think the market's going to go up or down? And so, we use all these complex features, such as even commodity index prices, the CPI index, equity market prices, all of these different features that throw unstructured data at a at our

9:29model, and our model finds patterns, Mhm. and with a high degree of confidence, we are able to eat away at that latency. Mhm. So, it's a very long-winded way of saying we use AI to give you as a trader best execution. Mhm. Meaning best possible price on any market within

9:55the shortest amount of time to execute. Fascinating, I'm at and I've you know, the AI has been in the news recently with this new AI program that just seems to be smarter than ever. And you mentioned there about the social sentiments and also finding like off-chain data on Twitter and I find that interesting because

10:23they've a lot of people have taken note that often you know, at the very top or at the very bottom of the market cycle, the crowd sentiment and the retail sentiment is wrong, you know, thinking that when it's up, it's going to keep going up forever. And when it's at the bottom, they feel like it's you know, they want to buy lower and that's usually the point where it turns around.

10:42And it's hard to judge with social sentiment sometimes, you know, in the middle of the trend when everyone agrees that we're continuing to go up, it's easy, but on those extreme end points, it's tough. I it would be great to have an AI that could that could pinpoint the top for you. True. I mean I mean, in theory, we could probably use our AI to trade.

11:06However, we don't do that because if you are trading, it's a whole different ballgame. You need to look at risk. And you need to look at risk management. You need to look at compliance. And as we all know, you can't be a broker agency you can't run a broker agency business and a proprietary trading business under one roof. As we saw with FTX.

11:31Right? Yeah. So, we took we taken that on board. We don't trade people's money. We simply provide best execution. Definitely. And how long has your team been working on this? And at what stage is it at now? Yeah. So, we've been We started early 2021. Our team is all made of ex-bankers

12:01and people who've done this for, you know, 15 20 plus years. So, our CTO built electronic trading systems for the likes of Goldman Sachs, JP Morgan, UBS. Our head of product used to be at Bloomberg and Lehman and City for 20 years. Specifically working on electronic trading, not just like at Bloomberg, but So, we have a team of dedicated electronic trading experts and

12:28crypto experts. We're simply bringing their knowledge into the fold. Our data science team is also you know, we have a professor who's a the head of data science who runs these two partnerships and has 15 post-doc researchers working for him. And that's just telling you a little bit

12:56about our team. But, we've been working on this since early last year. We have finally gotten our MVP out, which uses which showcases the order book aggregation technology that we have. Mhm. Plus, the ability to predict order books and prices. At the moment for our MVP, we're just showing it for BTC USDT, but the vision here is to really

13:25branch out and show ETH, show the top, you know, 20 currencies that are going around, but even branch out in maybe two two years time down into any tokenized market. Be that real estate, be that NFTs, be that, you know, fractional ownership of equities. Anything that has a marketplace that is tokenized and digitized, we can become that Fluid, basically can become the

13:51liquidity aggregator in that market. And that is kind of the direction of travel of what we're trying to build here. Mhm. And looking at an asset like BTC, which you would think would have the most liquidity of the cryptocurrency coins, versus a coin that's sort of in the top 100 you know, how much of a difference can Fluid make in BTC, which is already

14:18supposed to be, you know, pretty liquid, versus one of those smaller coins? Very good question. So, at the moment, our vision is to just provide, you know, liquidity aggregation technology for the top 20 tokens in the in the near term. BTC, obviously, is the you know, takes about 40% of the market, as you know.

14:45Did you know that 80 more than 85% of BTC trading happens on Binance at the moment today? Wow. This was a statistic that was released last week, showing you that the market is massively fragmented. Mhm. If you look at you know, if you go on to our website, fluid.finance, we have a little price disparity calculator, which shows the delta between the top six exchange

15:15sexes, not DEXes, but sexes, the top the difference in price on a range of assets. So, ETH, BTC, USDT on the top six exchanges. And you see that with Bitcoin the delta at the moment at any given one time is anywhere between $20 to $180. This is on a typical sideways trading day where you'd think that pricing

15:43should be really really tight. Shows you that there's a massive arbitrage opportunity there. But it also shows you that users are kind of screwed by these big exchanges. And being able to be the liquidity aggregator in those markets, even on just BTC means that what we do is that we take out the arbitrage opportunity. We make

16:11market makers across all these exchanges and all these venues honest. They will flow, but they've got to get a better price. You add competitive tension into the process. And what that will do is that we will harmonize a lot of pricing disparity between all the different venues. That's what Fluid ultimately do. Make markets more efficient. Make markets less susceptible

16:40to flash crashes. As you know, it's a big market manipulation and flash crashes, even within BTC are a thing. But with us in the room not just us, but any real liquidity aggregator in the room we take that likelihood out somewhat. You'll never get rid of it completely.

17:07But the more exchanges or the more liquidity aggregators like ourselves are there, out there the better the user experience and the more efficient markets become. Mhm. Interesting. And looking through 2021, you know, you started at the beginning of 2021 and then Bitcoin really ran up most of the year and there was a lot of people

17:35trading, you know, the volumes were way higher than they are now that it's sort of run back down through the year. Does you know, all that retail interest and people trading did did you see a significant difference in the liquidity then in you know, in the all-time high versus now when it's more of a sideways market? Or you know, obviously the it's the market is

18:01still fragmented, but does that interest in trading did that make a significant difference in the liquidity? It's funny you asked that. It's a good very good question. I'd say that the market last year you're absolutely right. The market now has become with you know, with the downfall of FTX, the market has become massively fragmented. Meaning the

18:25problem that we're trying to solve has become even greater. Meaning markets are even more inefficient now than they were a year ago. Mhm. So, for us it's a massive opportunity. Do we see that Yeah, what's happened with volumes? Of course, volumes have dropped. Retail traders have been spooked massively.

18:51Mhm. And this brings me on to another thing. Why have they not just retail but even institutional users in crypto have been spooked because of what's the cowboy nature of many of these operators within the crypto industry. And this brings me on to my second point, and that is I've been very vocal in calling for better regulatory

19:21oversight within the crypto industry. Now, even though crypto is, you know, a lot of OG crypto people are very libertarians, very libertarian, and very, you know, they love the they're almost like anarchic libertarians who want to change the system through self-regulation. Unfortunately, we saw with FTX, you know, SBF is a

19:48it doesn't stop the human element of fraud. And this is this is one thing that needs to happen better in within our industry. So I'm a very big advocate of that, and I think the venues that people trade on need to have better oversight. And also from our perspective as a liquidity aggregator, what we are looking to do, we only

20:16really work at the moment with verified compliant you know, best-in-class exchanges, of which there are not that many at the moment. We're you know, with the advent of DeFi that helps us also, you know, have work in a sort of zero-trust environment. But at the end of the day, this problem, this human problem, we are

20:46working as much as possible to mitigate through not, you know, anyone who did any diligence on FTX, saw very quickly that things weren't right there. And so we take controls, policies, procedures, SOPs very very seriously internally.

21:13And so, things will take a little bit longer to set up, but things will be done right. And I'm a very big advocate of that. Mhm. Looking forward to it. And for those next steps, Ahmed, moving into 2023 the platform is in MVP, what are the main things that your team needs to roll out to get to the next stage? Great question. We're now

21:39looking to have execution. So, the MVP aggregates the liquidity and showcases our capabilities within AI to build like you know, a kick-ass aggregator. Within that, there is a product suite that is fundamentally built on an execution layer. So, being able to actually take orders, put or put in orders into our smart order router.

22:09And have that, you know, you don't need to set up exchange accounts with 20 exchanges. You just need to set up an account with Fluid. And using our institutional accounts at all these exchanges, which we do a lot of diligence on, you'll be able to get the best possible price. Mhm. That's how that's what we're building at the moment, and hopefully that will be

22:31rolled out Q1 next year. That's great. And what is the best way for people to learn more about the liquidity side and how it all works and follow along with the updates for Fluid? Excellent. So, we have a very active Telegram channel. Which I hope you can sort of show at the end of this video. For sure. Twitter is very good. LinkedIn is very good for us. But also,

22:59fluid.finance. If you go to our website, fluid.finance, everything's there. You can just click on all our different social channels and follow us. We'll be able to give you a very good oversight on what's going on. We have a lot that happens in the space. We're very active publicly in trying to educate the world about cryptocurrency but coming from more of

23:27an institutional background or from an investment banking background, which is where the team has has has spent the lion's share of their the careers. Mhm. Amazing. Thank you, Ahmed. I will leave those links for the socials and fluid.finance in the description box below to make it easy for the viewers. Appreciate your insights into the market

23:51making and liquidity side of cryptocurrency. I think it's very important to know. All the best with Fluid moving forward and let's follow up in the near future. Thank you, Ashton.