- Language as a key to understanding culture
- How geospatial mapping and AI go hand-in-hand
- The fact that drones have revolutionized aerial mapping
- How mapping data on the blockchain ensures transparency and accuracy
- The importance of entrepreneurship requires taking the first step
Some other titles we considered for this episode, but ultimately rejected:
- Just Start Something
- How AI Is Changing the Game
- Ensuring Data Integrity in Geospatial Information
- Starting from Somewhere: The Key to Entrepreneurial Success
- Information About the Built and Natural Environment
Read the best-effort transcript
Read the best-effort transcript below (This technology is still not as good as they say it is…):
Michael Waitze 0:06
Okay, hi, this is Michael Waitze. And welcome back to the Asia tech Podcast. Today we are joined by Dr. Mehdi Ravanbakhsh, the CEO and founder of MapIzy and CryptoCrispy. We’ll have to find out what those things are in a second documentary. Thank you so much for coming on the show. How are you doing?
Mehdi Ravanbakhsh 0:25
Thanks, Michael. I’m glad to be here. I’m doing great in during the spring in Perth, which is almost two weeks away, and it’s really nice to be outdoors.
Michael Waitze 0:34
Oh my god, I’m in Fukuoka right now. And it is like 7000 degrees Celsius. It’s just so hot. Okay, I hope you get a good spring. Before we jump into the central part of this conversation, let’s get an introduction from you and some background from you. Are you originally Australian? Where are you from originally?
Mehdi Ravanbakhsh 0:50
Originally, I’m from Iran. And I did a PhD in Germany. So a bit of background but myself, please. I’m surveying surveying engineer by training. And after completing my bachelor and master at one of the top Iranian universities, which is Tehran University. Yeah, I was extremely lucky that I was awarded German scholarships called the the double A D, which is a scholarship sponsored by German government. And I was lucky among the top four that was awarded this scholarship. And I went to Germany for my PhD.
Michael Waitze 1:31
Do you speak German? Like did you do your PhD studies in German? Or did you do it in English?
Mehdi Ravanbakhsh 1:37
Yes, I know German a bit. Because that’s, I think one of the things that being in Germany, it’s really critical to make friends and enjoy the social events to learn German. Sure. And that was actually one of the things that my supervisor right at the beginning said that learning German is not really essential. But we appreciate if you learn German and Newell culture. So that’s, that’s really great opportunity. Yeah, we’ll learn the German a bit. And yeah, that was a bit of background. And then I’m, I was offered a job in Australia, in Melbourne Uni. So we moved to Australia and decided to stay there for the rest of my career.
Michael Waitze 2:17
Got it? Can you talk to me a little bit, just from your perspective about why it is so important to learn a language, if you want to understand the culture in the country that speaks that language?
Mehdi Ravanbakhsh 2:32
In every country, I think language is the main things to get connected to people and culture. And every concept you see in that society, as we know that the language is not just a language is there lots of thing in there, and you understand the culture, pupil society, everything in this in that society. So that was a, I guess, the great things in in Germany that there was free German courses at universities, and this was this is offered to international students. And so while doing my PhD, I spent some time learning German, interacting with people in the research organization and talking to people and it’s really, I guess, for engineering, knowing the engineering concept is not really critical, in contrast to humanities sciences, that it’s really critical if you if you’re going to know the know the big German philosopher, you have to learn German, but for engineering this, that that’s absolutely okay. You can write your thesis in English. Sure. And we still can communicate in English. That’s no problem. But I guess, knowing the culture and people and essentially enjoying your stay in in any countries, I think it’s a great idea to learn the language. Yeah, because before this, I spent a lot of time learning English. And so I knew the rule of no learning the language, somewhat. And the interesting thing is, is that the German structure is similar to Farsi, this is my mother language. Yeah, because normally German put the verb at the end of the sentence on like English, they put the verb after, yeah. So after subject, so there are stretches similar to Farsi. So in the three to six months, it’s really for for Iranians, I guess if they, they know the language, well, they can easily master the language. So it was lucky that I learned German as actually my supervisor. There was surprised in such a sort of short period of time. Having said that, in language, I guess conversation is one thing and writing is completely different. Absolutely. Yeah. Yeah. So writing in Germany is really difficult even for Germans writing the technical things, but I was lucky that I said, I was said that he introductions. Okay. And so my learning of German is limited just to conversation.
Michael Waitze 5:07
Got it? The reason why I asked this question is because I lived in Japan for 22 years, and I speak Japanese the same way you speak German, right? And I just wonder, I always wondered what it was like to just live in a society where you didn’t understand the language or things were going on around you. And you just were like, just floating past it. It’s just, I don’t know, I just think if you don’t know the language, you don’t understand the culture and you’re just missing out on something. And I think it makes your mind smaller anyway. You said you, you said you, you said you moved to Australia as well. What was what generated the move to Australia? What were you doing there at the at the university and why?
Mehdi Ravanbakhsh 5:44
Yeah, actually there. After completing my PhD, I was offered the different position at German universities in Canada, and one other position was in Melbourne University, and I check a check the internet, I said Melbourne is one of the most livable city in the world. I said, Okay. My decision is done. And yeah, so I moved out there, I moved to Melbourne, Australia based on the job offer, okay, so that was a postdoc, and particularly their position at the Melbourne Uni. And there was a research program in Australia, they call it the Cooperative Research Center. And these are million dollar project that essentially, unlike, let’s say, if I comparing Germany to Australia, in Germany, there is no funding for blue sky research, just research for research. So I guess the practicality and industry adoption is the second priorities. But in Australia, they created some research programs, that industry adoption is also critically important. Yeah. And yeah, I was lucky that my position was not purely research. But industry adoption was an important component.
Michael Waitze 6:58
So I feel like the universities in Australia, I’ve done a lot of recordings with guys and gals that work at University of New South Wales, right. And they have the same idea. It’s like, they don’t want people just doing research. They want that research to turn into something commercial. And I think that’s actually really interesting. Right? So you were researching with geospatial information. Can you talk to me a little bit about how that work intersects with artificial intelligence, which is the big buzzword right now?
Mehdi Ravanbakhsh 7:26
Yeah, sure, actually, that give you a bit of background, but geospatial information, especially information traditionally, essentially, is mapping. And unmapping gives you information about the built and natural environment. Yeah, let’s say for example, the roads, buildings, topography, and the green environment. So all this information there are mapping or geospatial geospatial is rather a new name. Because of the this geospatial is called a foundation data for every organization. So in every country, there is a mapping organization in charge of making sure that this mapping data is of highest quality maps
Michael Waitze 8:05
are just I’m staring at a map of old Nagasaki right now. Like just you know, I told him in Japan, and I’m fascinated by maps. Why are maps so interesting? And you said this kind of cool. It’s like, the built environment, which is neat, but also the natural environment, right? So it’s not just a bunch of buildings. It’s the rivers and lakes and everything else. It’s there. Why are maps? Yes, interesting.
Mehdi Ravanbakhsh 8:27
Yeah, I guess, because no maps is the basis for decision making for any activities on the planet. So if you’re going to do agriculture, if you’re going to be drawing buildings, or if you’re just doing generic activities, let’s say going after mountain, hiking and everything, mapping is the key things. And not only this is not just regular data that you see different industries, it’s a critical mission critical data, for sure. So even a small inaccuracy is not acceptable in mapping information. It has to be 100% accurate.
Michael Waitze 9:04
Can I can I share a recent experience with you? And he went off to spend a lot of time talking about it. But it was so interesting. I was driving with somebody in Thailand from Bangkok up to a place where we had not been before. And we’re using Google Maps to get that right, which in the old days didn’t exist. And like I told you, I’m fascinated by maps. So we’re driving, driving, driving. And then at some point, the map said to go to a certain place that didn’t seem like it made any sense. Yeah, but you knew the three cars in front of us were looking at the same map because they made the same turn. It was so interesting. I don’t even know how to explain why that matters. But it was just so interesting. Anyway,
Mehdi Ravanbakhsh 9:43
yeah. Yeah. Yeah. I like the region describe it, because that’s the issue in in mapping data organization that they should be some authoritative organization that they check the quality of the maps. Yes. And so otherwise, there might be some road your backdoor that in The map is not there. So there should be some, some control over the quality of the maps. And you mentioned about the intersection of mapping and AI, computer vision machine learning. So all these new technologies now, in the old days, almost 15 years ago, the creation of the map was a manual test. It’s extremely time consuming. And so essentially, let’s say for, for just capturing a building or putting a building on the, on the map, right, somebody has to digitize the coordinate of the building, or digitize roads, or digitize trees. So this process was extremely time consuming. And so what happened is that with AI, you can automate this process. So rather than doing so much counting, let’s say trees, building digitizing roads, or, for example, counting in the, in the new age of AI, for example, the concept of mapping has changed from just, you know, representing your road buildings and trees on the map through to detailed analysis of let’s say, agriculture farms, like mapping snails and slugs, for pest control. So these are new things that you can capture with imagery, because now let’s say in a bit of his background about aerial mapping, and this sort of thing, is all started from over the second. And that was time that mapping has been used data capture from above, or has been used for the reconnaissance purposes. But now, because of the drone, because of smartphones, you can put smartphones, on your will of tractors or everyday machinery in agriculture, you can capture imagery of the entire farm, then you can produce a number of different maps from, let’s say, fertilizer map through to no pest control, map, and different types of maps that you can use for various activities. So these are just some ideas. Now, for example, one of the big things that in autonomous driving now happening is that the detailed roadmap in in the past, it was just road centerline. Now, you can put a bunch of sensors on the on the weekend, when vehicle is moving, you captured this data. And this data through AI, machine learning and deep learning converted to the to the detailed information about our road network, which can be used for autonomous driving, these are just a couple of examples.
Michael Waitze 12:43
So 15 years ago, maybe 20 years ago, now I can’t even remember when this happened, right? Google started driving cars around with sensors on them to get very detailed map data, and also to do street view, as well, right? They wanted to film it and photograph it and stuff like that. How have the drones change the efficacy and the efficiency with which that can get done? So people don’t have to drive around, right? Because now, first of all, it’s better for the environment, right? Because it’s electricity, you’re not burning gas. But also you can go up as well, which you couldn’t do before, how has drones changed that process?
Mehdi Ravanbakhsh 13:18
Yeah, that was a significant changes in the aerial mapping. And in photogrammetry, which is a science of aerial mapping, essentially, with drones, you can capture at this very small size area. But not only that, but in the high level of detail. For example, if you use airplane, or satellite, you can capture these images from satellite. But the level of detail is just not enough for the different purposes. Or if you use airplane, the cost of putting airplane in the sky is so expensive withdrawn, you don’t really need the extra logistic just you, for example, as a farmer or as a hobbyist, you can use drone you can capture images. And the interesting things is that the level of detail in these images is impressive. And even for example, in agriculture, as mentioned before, we can you can see very high level of detail you can differentiate between, let’s say, wheat, or crops, or different types of trees. But in the past, we didn’t have that possibility simply because the level of detail was not was not there. So that’s one aspect. Even if direct capture direct can capture high level of detail images, they should be good quality algorithms can interrogate the data and can differentiate between different things on the imagery and AI here can help.
Michael Waitze 14:45
How, tell me how. Yeah, so
Mehdi Ravanbakhsh 14:48
essentially, you need to develop an algorithm that can identify different types of a specific object that you’re after, let’s say in the roadmap, pick your after or curves for, for example, the speed sign for different things. Essentially, what happens is that you capture a training data from images. And when you capture training data, you train the computers and algorithms to learn this object. By showing computers, for example, large number of snails and slugs, computer learn how to find those interesting creatures, on images when you fit the real images. So what happened is that algorithms can, can search and can identify those creatures, and put a bounding box around them. Now, using this information, you can create a heat map that can be used in the tractors or everyday machinery, right. And for, let’s say, for targeted spraying and for targeted, for example, fertilizer, you can use these technologies and see the evolution of this from just road and building to different applications.
Michael Waitze 16:04
Can you talk to me about map easy a little bit why you started it after all this research and all this tech stuff that you understand? And did? What led to the founding of map easy and what what did you do? Exactly?
Mehdi Ravanbakhsh 16:17
Yeah, that’s a good question. That one of the things as as you know, in academia, is that in academia, you need to do one things, but you need to be deep in that discipline, right? So you can do various things. That’s that’s not how they can name your work. So
Michael Waitze 16:33
there’s no such thing as a digital system. It’s like a generalist PhD just doesn’t work. Yes,
Mehdi Ravanbakhsh 16:38
no, that’s that’s. So what happened is that I spent 20 years just doing building detection and road mapping, when I was really tired of gene just one thing, and it was a scary to me to do the same thing for the rest of my career. And that was in 2012, I was Associate Professor at UWA School of Computer Science. And AI, the first paper of AI and computer vision came out in one of the conferences from from us. And we noticed significant jump in the accuracy of, you know, the detection identification of the object in the imagery, right. And I saw an opportunity to quit my job set up my PC and working directly with industry. And the catalyst was for this was one aspect was the rise of AI. And another aspect is that in the past, I’ve worked with large number of end user ad businesses in Australia. So I had an existing network. So I saw a good opportunity to quit, that was not really a decision that come out of nowhere. So I did a careful planning for it. I had a network, and I had a global recognition of AI and machine learning and mapping. So I saw a good opportunity to set up my PC. So what happened is that I set up my PC. But the university said that you have been great academics, we give you an office, just stay here as an adjunct professor, while doing some teaching. So you don’t need to pay for utilities. No, it support this sort of thing. I spent a couple of years at university after this, but at the same time working with with industry. But recently after the COVID, because of no high loads of teaching, I thought it takes too much time, right? So I moved my office after university, but still I have a good connection with university.
Michael Waitze 18:32
So what does my PC do though? How does it How does it work with organizations, and what kind of what real world problems is trying to address or to affect?
Mehdi Ravanbakhsh 18:42
My PC use different types of data that come from satellites are all from airplane or from multiple sensors, and platforms, this platform could be a smartphone, it could be drones, it could be an airplane, or it could be satellite. So in my PC with digest this huge volume of data from multiple platform multiple sensor, and we try to pull out the business and geospatial insight for different industries. Now, to give you some some examples of use cases that we had in the past one, one of the projects that we did in the past was for example, for roadmapping in the work in collaboration with the Syrian government. Another was for example, optimizing pest control through artificial intelligence. And essentially what happens is that we use a smartphone images captured by everyday smartphone of farmers, and we develop the algorithm that can quantify the number of snail and the slugs and producing a map for the targeted pest control. Another project that we completed with one of the research organizations trigger was for fish survey. And especially, we use the underwater stereo videos which are normally captured by research organization to monitor the health of coral reef and for species recognition. So what happens is that we use AI to identify different fishes pieces and also measure the size of the fish. So this information was critical, because diverse fish population means healthy coral reef, right. But there is no other way to measure coral reef. So if there is a healthy fish population, more diverse a fish population, that means a coral reef is healthy. And this research organization normally record videos all around Australia. And for humans, that was really impossible task to watch this videos measure fish. And another example for mining companies. And for mine closure at Monterey habitation because mining companies after the opera after the operation says they need to make sure that the mine site is back to the natural environment. Now one of the ways that they can measure it, whether the mine site is back to the environment is sending people to sample from, for example, natural environment, from water from soil back in the lab to measure whether the rehabilitation has been successful for for the government. But now, through aerial observation through satellite through drones, we do it through computer vision and machine learning. So essentially, what happens is that we quantify different types of, for example, native plants, right, we need to make sure that there are good diversity of native plants, and the shape of the topography is suitable for the environment. And we can do it more more frequently at low cost without putting people on the ground, which is risky. Yeah. And that’s that was another use case for might say, but the major, I guess, transition for us was using all this experience, and working with insurance.
Michael Waitze 22:21
Tell me how this works, because I do the largest insurer tech podcast in the world. So I spent a lot of time I spent a lot of time talking about insurance, we have listeners in 170 Something countries, right? So we spent a lot of time talking about this. I was just thinking as you were talking right, the oceans are so vast, I think we forget you give this example of fish, right. And coral reefs, I think we forget sometimes just how massive the oceans are. When we think of the natural environment cuz he’s talking about built and then the natural environment. It’s just so vast. And you’re right before mapping was done manually, some guy or some gal would have to like go to the top of the mountain and take a few pictures of it. But now we can just send drones everywhere. Right? And again, flying an airplane was absent. Absolutely. Flying a satellite is expensive, but flying a drone is ridiculously cheap. Yeah, absolutely so much more about the earth today than we did even just like 20 years ago, merely because we’ve been able to map it in a way that’s so much better.
Mehdi Ravanbakhsh 23:19
Absolutely, I guess, I guess drone images and data is good to understand the or natural and built environment or essentially our planet in greater detail. But at the same time, you see that this vision information come from various sources, whether that’s underwater, or whether that’s just for example, the station’s stationary cameras on field or whether that’s from satellite and from satellite, not only it is optical observation, which is optical data is similar to what you’ve got in your smartphone, right, which red, green and blue, but wrenching data is another aspect radar data and this is this is a new things and in radar, essentially, you can monitor the entire planet and the databases not only monitoring the surface of the Earth, but see what changed on the on the vertical dimension essentially, let’s say if there is a infrastructure somewhere, right there is a movement on the ground, you can identify this in the range of millimeters but all from satellite and this is this is really great for the safer grant movement. And we notice dif different disasters in the war just because of you know the change in the infrastructure bridges like what happened in Italy and other places. If you use this or for example, the I guess metro station in different countries they should there should be some monitoring system in place through ranging technologies like for example, laser scanning or For example, sir, or service essentially, is radar imaging. Through these technologies, you can monitor gradual movement, which the applications earthquake, the application is for big infrastructure monitoring. And so essentially what happens is that we, when we’ve got this data on a daily basis, there is a company in the US called Planet and planet captured the entire entire planet on the databases, right. And using this information, you can see where, for example, the frustration happens when there is any new changes on the surface, when there is, for example, an issue somewhere, whether that’s about natural environment, or let’s say illegal logging, illegal buildings, so anything illegals can be identified, or finding communities in Africa for helping them so because some of remote areas in Africa, they they even don’t have a map. So you can see a lot of advantages this has to humanity and to have a better community.
Michael Waitze 26:14
So talk to me a little bit about how you’re using this for insurance. We released an episode I think last month with Karolina Dreyfus who runs in Australia, who runs a company called Sync Technologies. And bio, okay, yeah, yeah. So Caroline is great, but just tell me how you look at it from your perspective.
Mehdi Ravanbakhsh 26:32
So essentially, the, the way that we operate for insurance companies is that we identify changes to buildings and to properties got it. So what happens is that, let’s say, if you as an insurance company, somebody comes to you, they asked for a code, you need to understand what’s happening to this property over time. And what’s in there, because no, let’s say, the size of the building, whether there is a risk from trees, or whether there is a bush land nearby, whether there is a water bodies, whether there is a pool solar panels, so all this information we identified through machine learning from above. And not only we identify what’s in there, which is, which is important for coating, but after the coating. Another process in insurance is for example, renewal. When renewal is due, the insurance operators need to know during this one year, what happened to property, if let’s say there is a liability in there, like for example, swimming pool. So what happens is that you need to contact them to ensure that they’ve got the sufficient protection in place and update the policy. So these are because new people are busy, they know forget to update their policy and this sort of thing. So that’s only one application for for insurance. I guess the challenge for insurance is that they’ve got the data. But this data is not up to date. And the level of detail isn’t really there. For example, they don’t know about the quality of the rooftop, whether the rooftop is tile or or metal. And know all these sorts of
Michael Waitze 28:14
things. I like it the use of technology in the insurance business. I mean, like I said, I’ve talked about this every week. So it’s just fascinating for me to see, we also spoke to a company called Trent spec, run by Derek February. I think that was his name. Let me make sure I have that right. Yeah, Derek ferry. And they do that. Yeah, Derek does similar things where you’re talking about right. And we had this concept of like, the cradle to grave of a building. You film it with drones when you’re building it. So you’re not you know exactly what you have when you get it. And you just constantly update that data. So buildings alive, right. And because the building is alive, it allows you then to write better insurance policies for and do other things as well. Particularly if it has other sensors in it. We could go deep. We did a whole episode on this, we could do more. But it’s just interesting the way you’re using it again, in this context of the built environment and the natural environment and how map easy makes that easier for people to use. I want to jump to something new, though. Because you moved from mapping into crypto tell me how this happened. And and why is this part of your idea of I don’t want to do the same thing for my whole life. So I’m going to do something completely different.
Mehdi Ravanbakhsh 29:22
Yeah, yeah, I guess it just wasn’t an interesting one. Because the magazine was founded in 2014. Right. During this years, we had a number of projects and we work with a number of organizations Australia overseas, and we had a really great success and a wide range of use cases, as I’ve mentioned, from from agriculture, forestry, fisheries, through to oil and gas mining, and build environment, insurance, real estate, urban planning, so various things and I’m really enjoying this but at some point, I thought, Man because he’s now in a good shape. And I had a conversation with my my older brother back home and he lost a couple of $1,000 in crypto. And I say how how you trade crypto and said that there is a telegram group that he gets trading signals from there. Okay. I thought, well, this should be a better way to create the signals rather than just relying on anonymous people.
Michael Waitze 30:28
On telegram? Yeah, that’s a little bit. Yeah. Yeah,
Mehdi Ravanbakhsh 30:32
it’s a bit of scary to me know, you, you’ve got this heart and money. And you put in some account and relying on just information that you can really verify its accuracy. Yeah. So. So that’s that was actually the motivation behind this. And I know if I read a bit about this, I noticed that some people actually lost all their savings for in trading, and particularly crypto. Yeah, because it’s a bit challenging not to find the to predict the market. So yeah, that was the motive behind that. And one of my team members in my PC, actually, he has some experience in crypto. And so we get together, we actually put all our thought together what to do next. And we found a really a niche market in crypto, which essentially was developing an AI assistant that can monitor the entire data ecosystem of crypto, whether that’s, for example, the on chain, the technical indicators, patterns, sentiment, no veils, financial information. So we put all this information into AI because the core technology was similar to what we developing map easy. And let AI learn everything there, whether that’s pattern, and whether which one of this data set has more significance for a particular coin. So this is the way that AI learn everything through the historical data. And with just single click will be provided signals that people can use it, especially for the short term gains. And yeah, that was the whole idea. In crypto crispy,
Michael Waitze 32:30
how do you solve for liquidity?
Mehdi Ravanbakhsh 32:33
Actually, for liquidity, as you know, this is mainly I guess, those companies that they are market maker, and they they offer trades, and
Michael Waitze 32:44
but you understand what I mean, right? In other words, if I want to, if I want to buy Toyota at a specific price in the stock market, I can buy as much as I want, right? If like, if you look at HSBC probably trades $100 million on the bid and $100 million on the offer, right? So there’s so much liquidity there, if I want to make a trading decision, simple. But in crypto, it still feels to me like it’s dominated by some really big players. And that, I wonder if you if it’s really the case, that the signals are there, the technical signals are there for sure. Right. And artificial intelligence can definitely help you get them. But is there enough liquidity on the bid or on the offer? In any particular market, to be able to actually trade at those prices and to make a difference in your day to day trading?
Mehdi Ravanbakhsh 33:29
Yeah, one thing that I’d like to make here is that crispy is a data analytics company. So essentially, yeah, we are not we are not connecting to any exchanges, because otherwise, you need two different licenses. And essentially, our focus is just on the data aspect of crypto, and digesting this huge volume of data through AI and predicting what’s happened to the market. Essentially, we do the market prediction. Now for liquidity. This is the area that of course, I’m still learning. But once you get your signals, you need to go to your exchange of choice. Sure. And there you need to execute your trade. I understand that, for example, there are trading bots that they connect to your account, and they execute trades for you automatically, which is a bit scary to some professional traders. So we don’t do that in cryptocurrency essentially in critical speed. You’re in control of your, your money and your trade. So essentially what happens is that we free up your time, you can do whatever else you want. You can create your portfolio from your exchange into cryptic recipe, and then we monitor your portfolio to the alert system. When there is a change to your portfolio. Let’s say less than a certain percentage. We send you an alert that for example with today market and this information. These assets are Under the risk under threat, so you need to do something about it. And that’s, that’s the main thing in crypto crisp. Yeah, having said that, we know that in in this space, there are different projects, like for example, algorithmic trading, robo advisors, various things, that they they target specific things. In our case, for example, we will focus, actually, our niche market is just AI signals your prediction. At the same time, we also offer different different insight, but the main things is that rather than for you to have a chart that will look at the chart, and analyze thing, finding patterns, and spend so much time, let’s say I do this job for you, right? Because no, that’s, that’s so easy. And then, interestingly, guess, Mike, you’re your expert in this. But as you know, in trading, if the accuracy of your signals is just 51%, that means that your portfolio is going to grow.
Michael Waitze 36:04
So I love this. This is what my first trading manager told me if you win, 51 was like, How often do you have to win, right? And he was like, if you just win 51% of the time you’re winning, you don’t have to win 70% of the time, or 80%, just 51? Because that means you’re winning. Yeah.
Mehdi Ravanbakhsh 36:18
So what’s the beauty of acuity crisp, is that we produce a number of reports. So every week, we produce reports that show the accuracy of our signals, something between the 65 and 70. Person. That’s, that’s really impressive. If you compare to this concept that you’re you’re sure that of course there might be the up and down, then you might have some lost rates, and might give, but overall, over time your portfolio grew. And you make profit.
Michael Waitze 36:49
This is the thing, the hardest part about trading is not understanding when to buy and to sell, per se. It’s not about missing a signal. It’s about how much pain can you take? And how disciplined Are you? Right? Exactly, yeah. Do you know what I mean? No, because when the market is going up, if you just buy Apple at $12, and it goes to 100, you feel like a genius. But if you buy it at 100, and goes to 88? Do you sell it? Do you buy more? How much pain can you take? Right? So it’s trading is actually more psychological than intellectual? And that’s the that’s the thing I think most people don’t understand. So even if, and this has nothing to do with Chris, crypto crispy at all, but even if you get the right signal, if it’s painful for you, you may not do it anyway, that’s that’s a different topic for a different time. I want to talk about
Mehdi Ravanbakhsh 37:38
that. So one thing that I like to want to add this before moving on, is that not only signals, we also create risk parameters, like take profit and stop loss. Yeah. So just to cover it. No,
Michael Waitze 37:52
it was still, this is the thing that right? It’s like if it says stop loss, people still may not do it. Like, you know what, it’s probably gonna go back up kind of thing. It’s just it’s so hard to be it’s the discipline that really matters in trading. And that’s what differentiates the great traders from just the mediocre traders. Is that just how are you on your stop losses? Right and your take profits? Anywhere? Absolutely. Yeah. What is the end? Let’s finish with this. Like, what’s the intersection of blockchain? Right? And this geo geo AI stuff that you’re working on? Because I hear a lot of conversations about this intersection of like artificial intelligence and blockchain technology? What does it look like for someone who’s participating in both?
Mehdi Ravanbakhsh 38:33
Yeah, actually, this is very this is I think, great technology. Blockchain is really critical in many of these applications. And for mapping, it’s really important because in blockchain, essentially, what happens is that gives you transparency, and gives you back the control of the asset you’ve got and the sin in mapping what happens that it’s really important, you when you’ve got the data, there should be some connection to the real world. And if you store this data in the blockchain technology, you know that this this data is not manipulative, the accuracy is intact. Let’s say recently, I saw proposal in one of the Defense Department that it’s public, there’s no competition. So just with your knees, essentially, what happened is that yeah, some, some organization, they capture, let’s say, there are some public satellite imagery, they manipulate these images and they put some artificial buildings on the image,
Michael Waitze 39:37
right? So this is where this is where it gets really interesting. I’m just thinking of this word immutable in my head once this data, physical data, right, and again, natural are built doesn’t matter. It’s on the blockchain. It’s can’t be changed. It’s immutable. So you can be charged. Yeah, that’s actually kind of Yeah, sorry. Go ahead.
Mehdi Ravanbakhsh 39:54
Yeah. Yeah, that’s that’s I think great things is surely for, I guess, many application It’s really important that when you access satellite imagery, you need to put the satellite imagery through some process to ensure that this data hasn’t been manipulated, especially if you’re using for national security application. If maybe, for example, there are some, like, similar to World War Two, the second that I guess, used to pay one invade from the Normandy, and they put, you know, some some, I guess, a plastic tank somewhere else. So essentially, what happened is that they might be some changes to images, so in a way that can deceive the other countries. So, now, using Blockchain, when you access your your images, you know that these are intact, and what you see on the image actually represent the physical world. So in that way, that sense, I guess, it’s important, that’s one aspect, which is the content of image Another aspect is the position, the position of the image, whether this this image represent, let’s say, a city, north of Australia, then you can manipulate this, and rather than representing that city represent somewhere in Indonesia, right. So shifting the accuracy. So in both aspects, I guess it’s a, this is a great tool to add, in, like in any data set in geospatial we’ve got some metadata. Metadata is also important, who is the owner of this data, right? So different information about this? Geospatial data is there. And if you know that, this data has been verified. And there is no manipulation of the data, of course, that adds more credibility to mapping in general, give people the confidence. And this is generally good for the government organization for businesses to use this data with confidence. Let’s
Michael Waitze 42:01
end with a little bit of a story about a boy, a little boy in Iran, dreaming about what his future is going to be like, getting very well educated, doing a PhD in Germany, doing postdoc work in Australia, and then becoming an entrepreneur. Just tell me like one, and here’s the context. You know, years ago, Nike came up with this phrase, just do it, like stop thinking about it, and just do it. And then there’s a company out there called Think Media. And what they do is they give people advice on how to like, start their own YouTube channels and stuff like that. And their their catchphrase is just hit record. And I think they kind of stole this from Nike, right? It’s the same thing. It’s like, just you just have to start doing it. If there was something that you’ve learned for entrepreneurship, and you could teach to an aspiring entrepreneur, what would that be?
Mehdi Ravanbakhsh 42:55
A start something, assert something. So just to start from somewhere and doing something I guess you should not. If you wait for the perfect situation, if you’ve got funding, you’ve got to know all the accreditation, that’s not going to happen. You just have to start from somewhere. And then along the way, we’ll find opportunities and everything should be alright out there.
Michael Waitze 43:18
That is the key. Okay. Yes. Thank you so much for doing this doctor. Mehdi Ravanbakhsh, the CEO and founder of MapIzy and CryptoCrispy. I think I’ve butchered your name.
Mehdi Ravanbakhsh 43:30
You got it right. I did the best. I got it right. No worries. Okay. Thanks for having me. Thanks, Mike.