ANIRBAN MAHANTI | 7Investing
ANIRBAN MAHANTI | 7Investing
Computer networking, data science, and machine learning are at the heart of enterprise software. Anirban Mahanti has serious form in the technology space. Just check out his bio below. This is a far-reaching discussion on technology and looking for opportunities from innovation.
"I'm basically focused on what the company's technology is, what problems are they solving? Are they growing really quickly? And if they're burning money, then they're burning money for some reason, right? So, if they're burning money for sales and marketing, but the sales or marketing is basically allowing them to capture a lot of Greenfield opportunity and they're locking that in. Then that's a good thing because you know, over time you can assume that that sales and marketing expense is going to be a small fraction of the overall revenue pie that they would have. And therefore, it will be profitable at scale.”
Anirban Mahanti is a lead advisor for 7investing. Before 7Investing, Anirban spent 5-plus years at The Motley Fool’s Australian subsidiary in various roles, including as the Director of Research and the founding lead advisor of the market-beating small-cap ASX stock-picking newsletter Extreme Opportunities.
“There's a wealth of knowledge out there that one can get, which is really from the annual report of some of the best companies. You don't have to read it from an up-and-coming company, but if you want to, for example, understand software as a service, then maybe pick an established software as a service company, like say salesforce.com. And that'll give you an understanding. So, I think there's a lot of opportunity to learn. And as the more you learn, the more confident you get and of course, then you can wade in the pool a little bit and then try to go into the river. Once you're comfortable going into river, then maybe you can try paddling in the ocean. So, baby steps is the way, but I think just read. Reading a lot really helps.”
Anirban is an Information Technology expert by training. Before transitioning to becoming a full-time investor in 2015, Anirban held various roles at NICTA (now known as Data61 following the organization’s merger with CSIRO), a leading Australian Information Communications Technology research center. Before NICTA, Anirban was an Assistant Professor at the University of Calgary in Canada and then at the Indian Institute of Technology Delhi.
In his research career, Anirban has invented new technologies, co-authored over 70 peer-reviewed research papers, supervised post-doctoral fellows, Ph.D. and MSc students, and consulted for the industry. Anirban’s technical expertise is at the intersection of computer networking, data science, and machine learning.
Given Anirban’s background, it should be no surprise that his favorite investing ideas are at the bleeding edge of innovation, especially in the enterprise software space. He’s a firm believer in following top-flight research, which often is years ahead in identifying where the puck will be in the future. For context, Anirban’s doctoral dissertation was on scalable video streaming systems, back in 2003, well before the first large-scale streaming systems went online. His team was also one of the first to publish a study on YouTube traffic characteristics back in 2006. And he was applying machine learning (ML) and artificial intelligence (AI) to computer networking problems (SPAM detection, network traffic identification) in 2005, about a decade before ML and AI caught the fancy of mainstream media..
Anirban received a Ph.D. and MSc in Computer Science from the University of Saskatchewan, Canada, and a BE in Computer Science and Engineering from Birla Institute of Technology, India. He lives in Sydney, Australia, with his wife and daughter, although he considers himself a global citizen with living and work experience spanning continents.
TRANSCRIPT FOLLOWS AFTER THIS BRIEF MESSAGE
Disclosure: The links provided are affiliate links. I will be paid a commission if you use this link to make a purchase. You will also usually receive a discount by using these links/coupon codes. I only recommend products and services that I use and trust myself or where I have interviewed and/or met the founders and have assured myself that they’re offering something of value.
G'day and welcome back to Shares for Beginners. I'm Phil Muscatello. Computer networking, data science and machine learning. They're all at the heart of enterprise software. I'm really happy, overjoyed in fact, today, to be able to talk to someone who has serious form in this space. Hello, Anirban.
Hi Phil, how are you?
Phil (1m 2s):
Really good. Anirban Mahanti is lead advisor for 7investing. Before joining 7investing Anirban spent over five years at The Motley Fool, most recently as their director of research. At The Motley Fool he was also the founding lead advisor of the ASX stock picking newsletter Extreme Opportunities. He has an academic record that's way too long to go into here, but you can browse at your leisure in the blog that will be attached to this post. So, you started in academia. I mean, like I said, you've got a list of qualifications as long as your arm. What was your path going from academia to investing?
Anirban (1m 41s):
That's an interesting question. You know, coming from like an Indian middle-class family, my father is a professor of computer science, my brother is a professor of computer science, so I was also a computer science, well, my brother is younger than me but, you know, we're a family of academicians. Academics was really important to us and finance was actually not important to us. We were never really taught about finance, right? So, I started looking into investing only when I was in grad school and I had some money. I already had a job lined up at the university of Calgary as an assistant professor. So, you know, I had a job lined up and I was thinking about, okay, now I'm going to actually make some money. So maybe I'll have some savings and maybe I'll be able to invest sooner.
Anirban (2m 22s):
That's when I started looking at investing. Before that, I hadn't looked into investing. And my father did invest but that's old style. He used brokers and stuff like that. Broker told him buy that stock and he bought that stock. So, the journey really was that over time as such the savings grew, I figured that I needed to learn and learn more and I ended up discovering The Motley Fool, which really changed how I thought about investing. Largely because their approach to investing is really different. They talk in terms of making it easy and easy to comprehend and, you know, remove all the jargon out and that got me really interested. And that's how I started investing.
Anirban (3m 3s):
And then overtime, I got interested to the point where I figured that, you know, if have serious money in the market, it actually makes sense to try to be a professional. And that's when I started thinking, well, you know, I've done a lot of research, spent a lot of time, maybe it's time for some career change, right? And I started applying for some jobs and I applied for some, some got rejected, for some I got interviews and eventually I landed a job as an analyst at The Motley Fool here in Australia. So, Joe Magyer hired me and he was basically looking for someone with software background because he was interested in, sort of, software and investing in that space. So, he was looking for someone who sort of, knew it from the outside without sort of having an investing, maybe, lens on it.
Anirban (3m 45s):
At least that's my interpretation at this time. You know, but once you get in, sort of, you start at the bottom of the pyramid, right? And I was, sort of, at the top of the pyramid in what I was doing in my former field. Now I was at the bottom of the pyramid in this new field. But that's okay. That's also challenging. And then you learn, you want to learn because you want to climb the ladder, so to speak, right? And what I thought was interesting for me was, at least, you know, my PhD advisor used to say that most people who get into a PhD program are smart by definition. If you complete a PhD program, it's not just about being smart, but having the perseverance to actually finish things. You know, and if you persevere in any field, you can actually learn about it. And investing is actually not that hard to learn about if you really persevere.
Anirban (4m 29s):
Of course, everything requires practice, right? You know, practice makes us perfect. So over time you'll just get better. But if you have the interest and inclination and the ability to think and analyse which you would have, if you have any sort of technical background or even non-technical background, even if you're a business person, right, you have the ability to analyse things. So, you can put your life skills to work here. And that's sort of how I got started. And it was really motivated by trying to do something different, trying to manage my own money and just more interest in learning about things. I guess there was another related angle. And the related angle was that I was really interested because I was working in technology. And most of the great companies are investments that happen to be the greatest investments of our time, are technology companies. So, there was something there in common.
Anirban (5m 9s):
And, you know, my thinking at that time was that I would be able to look at things and see things that potentially gives me an edge. You know, it more qualitative way of looking at things because I understood the technology background. So, you know, learning the investing language and then bridging that gap would be my edge. So that's how I've invested really over time.
Phil (5m 30s):
Yeah, I was going to ask that. I mean, you obviously brought some of your expertise from your areas of study into investing and there's a lot of people who are analysts who will not necessarily understand exactly what many companies are doing and is that something that you can bring?
Anirban (5m 50s):
Yeah. So, I think skills carry over, right? So, if somebody comes from a banking background, they will be able to bring that background and apply that lens to a lot of things, right? And they can learn about enterprise software as an example. So, you can learn something and then apply what you know really well and look at it through that filter. I have always looked at companies through sort of the tech filter that, you know, "Okay, what is important in technology? What's really an edge, right?" So, like, there are lots of examples I can give as, you know, I've been tweeting today about, for example, network effects, right? So, like network effects where, you know, more users bring more users and then they're self-perpetuating, right? And once you've built a big network, it's emote because it doesn't go away easily. As a tech person I would say that, well, you know, are those moats today valid?
Anirban (6m 37s):
Are they strong moats? They're probably not strong because if you realize how technology is created today, it was hard to build an app when Facebook started because there was probably the first big super app, right? There was Myspace at that time as well but it was not really a super app at that time. But the tools for building an app were not that prevalent, application developers will not that prevalent. And you know, there are not thousands of shops out there building apps like there are today. Building an app today is really, really simple. Just like going online today is really, really simple. Therefore, building something is not hard, putting something online is not hard. So, if you have a creative idea, you can actually build something creative and put it out there. And the fact that there are so many networks out there today makes it really easy to get vitality, right?
Anirban (7m 21s):
So TikTok came from nowhere and became the super app, right? So, you know, just thinking that you've got a lot of users on your platform and therefore that makes it strong, is probably a very weak moat, which, you know, in the finance world, it might be looked upon as a strong whereas I'm calling that a weak moat. Because you know, that can go with very quickly. It went away once from Myspace and it can go away again today if you don't innovate on top of that, right? And so, I think that are ways to think about that. So, I'll go back to my PhD days. So, I'm a hardcore networking computer systems person, you know? So how do you make computer systems run faster, better? How do you avoid network bottlenecks? How do you make things, you know, run with less bandwidth, less costs?
Anirban (8m 2s):
Those are sort of the things I looked at. How do you make content accessible to users? How to distribute content at scale? Those are the things that I initially worked on. I had some friends and colleagues and peers who are working on human computer interaction. They used to do, you know, studies about how do you make computers interact with humans and make it easy? And I said, "This is all fluffy stuff you guys are doing, right? You know, this is not science because this, you know, this is not, you're not doing any hardcore stuff. You're doing these experiments with, you know, 30 people in the lab." But it turns out that is actually one of the most important skills in computing, right? Then when I look at Apple as an example, I think, "Okay, you see those guys who really excel in design, they can figure out how the computer should interact with the humans."
Anirban (8m 45s):
And then that reminds me of Steve Jobs' analogy about how computing is basically a tool for making, at least in their mind, a tool for making the life of human beings easier. It's a tool for allowing us to do things. At least because they're not really from the artificial intelligence page where the computer doesn't take over your life really, right? Computer comes and helps you do your work. So, in that instant, or at least in that context, the human computer interaction is important, right? So again, you know, I was able to make this connection that yes, later on, you know, what I called fluffy, it's not really that fluffy because it's really important. And you see that design principles apply in many different ways. You know, whether it is how recommendations are surfaced, it's not really about the best recommendations, it's about how you surface good recommendations.
Anirban (9m 28s):
You don't have to have the best, but about having good ones, but surfaced in a way that interacts with people, that makes it easy for people to find. So, I think a lot of those things are important. You know, the interaction between the machine and the human. It's again, you know, you sort of appreciate these qualitative aspects to technology, which is not just hardcore science, but there's a lot of quality aspects that tell us and then of course the history of technology also teaches you a lot of things.
Phil (9m 54s):
Well, that was going to be my next question: what are the greatest changes you've seen in technology since your undergraduate days? But it seems like the human computer interaction would be right up there at the top of what you'd see as one of the great changes?
Anirban (10m 7s):
Yes, so HCI, human computer interaction as a relatively new field, right? And I was just mostly using this example. I think the biggest change at least I have witnessed is, so the things that have happened and this may be obvious is the internet has become widely accessible. The internet has become faster, highly interconnected, right? Devices have become smaller, smaller, faster, cheaper, widely deployed. And then the protocol stacks have changed in the sense that, you know, a lot of stuff now is being developed on top of, you know, the web or the HTTP, or as applications, right? And then of course, mobile changed everything with the mobile ecosystem where you could now develop stuff on top of the Android or the Google Play Store or the Apple App Store, right?
Anirban (10m 51s):
So that create another layer over which you can deploy things. So, I think those are the biggest changes. Of course, as networks became widespread, you saw widespread distribution of data centers and therefore server farms and therefore, you know, things like the Amazon web services and the Google cloud platform and things like that, right? So those things have happened. But I think the abstraction really switched from down to up. When I say down, what I mean is like you think of companies in the past, they were, you know, Microsoft, a computing company basically made PCs. If you think of, you know, the PC or PC technology, Apple was basically a Macintosh company, Cisco is a networking gear company from there.
Anirban (11m 34s):
So, we have gone to things that were made at the hardware level and lower down in the networking stack, at least. They have sort of gone up the top, you know, in terms of, you know, they're all our applications we are talking about, you know, things that are delivered as a service because you can actually deliver things as a service. You couldn't deliver things as a service in the year 2000, it was just not possible. The infrastructure was not there. Internet was not widely connected. So, I think the biggest change is the change in the internet, right? And then people would make a big deal about, you know, web one web two web three, I'd say that, you know, they're just basically just evolutions of the same internet. The big change was from the internet to web because that was just an abstraction and then the layer of abstractions on top of it.
Anirban (12m 18s):
So, I think the biggest thing for me is abstractions. Over time, the internet and the technology has just allowed abstractions to happen, right? And that's the beauty of the internet, it's designed was so good that it allowed all these abstractions to happen and we constantly have abstractions. So, I think that's the big takeaway is that abstractions have happened over time.
Phil (12m 39s):
So, having seen these changes in real time, how difficult is it now to look into the future of technology? We're getting the crystal ball out here, but we couldn't have predicted these changes, you know, 20 years ago, 30 years ago in the early days of the internet. What are the kinds of tools that you look to try and look a bit forward?
Anirban (12m 58s):
So, you know, like you cannot predict the exact path. That's nearly impossible. And the reason you can't predict exact path is there are so many variables in something happening, right? Take any technology, right? So, I'll give you one, that's pretty close to me. My PhD thesis was on streaming media. So, you could see that streaming media is going to be an important application, right? What part is going to take and how it's going to become widespread is unknown, right? And often you can't really, because there's a lot of commercial
Phil (13m 30s):
You couldn't see TikTok coming.
Anirban (13m 32s):
You couldn't see TikTok coming. But just to unpeel. So, when we were working on streaming media, I used to ask my advisor question, should we just spin out? Should we have a company? And my advisor had a nice thing, he used to say, "You know, you can always spin out the technology, but we've drove to realize that there's a commercial aspect and the commercial aspect is where's the content." Right? You can spin out and do something, but you don't have the content, right? So, you, and there's this constant loop of people with the content need to be on board. And then maybe you have something that's been out with, right? So that's just an example of there's a lot of commercial considerations. So, my point really is that you can see the path to a large extent by following top tier academic research. Because we talked to academic research almost always is looking at blue sky things.
Anirban (14m 17s):
You know, they want to solve problems that are interesting, important and relevant that haven't been solved today, right? A lot of this is theoretical, a lot of this is, you know, based on experiments and simulations, right? But they're basically paving the path for the future. And a lot of this knowledge is going to be open sourced because they're going to be published in conferences and journal and therefore, they're open source. Then the students who produce this knowledge, some of them are going to go to academia research. Some of them are going to land up in industry, right? And then they're going to take some of that technology and then make it, deploy it or deploy it and develop it at scale, right? So, you can sort see the part of things that are likely to happen, right?
Anirban (14m 57s):
How it's going to happen. It's hard to know. So, for example, an example might be that crypto is going to play a big role. Exactly how and what is unknown. Similarly, you could say that, you know, bots, for example, robots of some form are going to play a big role, but in what form we don't know, right? Similarly, you could say that, you know, you're going to have self-driving and other different general AI problems being sold in what form and when and where again, really don't know, right? But those would be some of the examples of things, right? So, I think general direction we can see, but not specifics.
Phil (15m 32s):
It's interesting that we've got to look back to history sometimes to see analogies in this. It's like the railways, everyone wanted to start investing in the railroads when they started going out. But the key developments were more what the railroads allowed an implemented to take place.
Anirban (15m 50s):
Exactly. Exactly. So, like the analogy, I guess I can draw there would be, if you think about the first company to implement a streaming system at scale, at some sort of scale was actually a Time Warner, right? They ran a trial in Florida in 1994 with a catalogue of, I believe like a thousand movies. And it was a fairly successful trial and then they closed it, right? And if you look at the history of papers published, you'd see that the first papers were also around published at the same time. And they were published in a place called IBM TJ Watson lab, this around 1990s, early 1990s, right? Now here's the thing, right? Looking at who deployed the first servers and where the forest resource papers still could not help you invest because neither did IBM become a big deal in streaming, nor did Time Warner become a big deal in streaming.
Anirban (16m 39s):
That big deal in streaming went to actually two different candidates. One was YouTube, right? Which had a completely different model to what it was trying to do, right? And then one was Netflix, which started basically as a DVD rental business, right? So, what I like to say is that, you know, you can identify the problems that are important, and then you can find which companies are getting traction. And then once you find which companies are getting traction, you can do some research into those companies and see, okay, why are they getting traction? Whatever they're doing. And maybe that tells you a little bit, some of this is storytelling in the sense that, you know, it's easy to make up a line of thought that applies to things that have been successful, right? Because there are many other Netflix lookalikes that didn't succeed, right? So why did Netflix succeed? It could be the natural question to ask.
Anirban (17m 19s):
But my point really is that by looking at who had the first papers or patents that belonged to IBM, and then who did the first trial that tells you really nothing, but it tells you general direction, at least of where technology is headed.
Phil (17m 40s):
Now, I just want to look into a couple of terms that we hear in terms of technology companies. Now this comes back to a podcast that I was listening to over the weekend, Business Breakdowns, and it was about Twitter. It's interesting, Mark Zuckerberg, I didn't realize that Zuckerberg had a sense of humour, but he described Twitter as being a clown car that fell into a goldmine and everything about the technology of Twitter seems to be just put together. And one of the points that they made in this podcast was about the tech stack that Twitter had a huge unwieldly tech stack, but then it had to be broken down and put into modules, which are put together in much simpler ways.
Phil (18m 21s):
And this is a question I want to ask a technologist. What is a tech stack and what does it mean for a technology company?
Anirban (18m 29s):
You know, the thing with the tech stack is, like it means different things for different companies. So really like for a company like Twitter, it would mean, you know, what are they doing? You know, what's their data stack, what's the networking stack what's happening at the sort of the application user interface level, what's their discovery engine looking like and things like that. So various elements that are part and parcel of how Twitter is used and how Twitter actually enables people to discover things, right? So that's the tech stack. I find that term to be a little bit abused because if you take a pure sort of academic view of what a stack really means, then for example, from a networking point of view, the stack really means at the lowest level, you have physical pipes.
Anirban (19m 18s):
On top of that, you have something that connects individual pipes. On top of that you have an internet protocol that runs on top of those things. Then on top of that, you have sort of the, you know, the transport layer that enables reliable communication or unreliable communication. Then on top of that, you have sort of applications running or application protocols, so HTTP, which is basically the protocol runs on top of it. So that's a stack in the real sense. And that's the networking stack, right? Now, individual companies would have individual stacks. So, another way to think about this is you could say that Twitter really has no stack because Twitter is an app, ultimately an application, right?
Anirban (19m 59s):
So, they're an application software that runs on servers and does a bunch of different things. But then the applications though are ultimately running on the web. So therefore, they are running on top of the web. So, as I said, most applications these days run on top of some other protocol or they're running on top of the mobile platform, right? So, the stack in that case, the real stack is owned by somebody else. And if you're thinking of a mobile platform, then the mobile platform is owned by somebody else. They own the stack because they can make changes that can ultimately affect Twitter, right? So at least in this case, they're talking about, you know, stacks or what I guess that comment, they would mean is they're basically saying that, you know, it's an ugly put together
Phil (20m 41s):
Chunk of technology
Anirban (20m 41s):
Yeah. Basically, the engineering is borrowed this and that from various pieces, that might be actually true for a lot of companies because, you know, you love to open-source technologies out there, softwares that you borrow from here or from there, put these things together. But yet, they no, they're probably talking about data stack, network stack, you know, application, logic stack and things like that. So that's sort of what they're talking about and if it becomes unwieldy, then it becomes harder to make changes, right? And therefore, to evolve the app, it becomes harder and harder because, you know, you have not developed it in a nice modular fashion so...
Phil (21m 14s):
And that's what's really important about a technology company is that it's got to evolve and it's got to take into account changes and develop into what users require.
Anirban (21m 22s):
Exactly. Yeah. The ability to change over time is very, very important and technology moves faster than anything else. So, you need to actually change really quickly.
Phil (21m 29s):
So how do you value companies that may have a lot of potentials, but no profit and sometimes even no revenue.
Anirban (21m 35s):
So, a lot of companies would not have, like a lot of companies I look at don't have profits. I would look at things like, say a cashflow from operations, for example, if they're generating actually some cash from operations. So, they might be still burning some in most of these cases, then I'm basically focused on A) what the company's technology is, what problems are they solving? Are they growing really quickly? And if they're burning money, then they're burning money for some reason, right? So, if they're burning money for sales and marketing, but the sales or marketing is basically allowing them to capture a lot of Greenfield opportunity and they're locking that in. Then that's a good thing because you know, over time you can assume that that sales and marketing expense is going to be a small fraction of the overall revenue pie that they would have.
Anirban (22m 17s):
And therefore, it will be profitable at scale. So, I'll really try to think about, can the company get to scale and if it gets to scale, how sticky is the company going to be? So, companies' revenues therefore are going to be sticky and therefore at scale, what sort of money can this company make? That's sort of the basic focus, which basically means I try to focus on companies that have large addressable markets. You're valuing the optionalities to some extent, right? So, a company, let's take Tesla as an example, when they appeared was not making any money, right? But what you're really paying for at that time is the fact that, well, if electric cars take off and if electric cars are going to be profitable to make and if they take an X percentage share of the market then, you know, they're going to make this much more money.
Anirban (22m 58s):
And then of course, then you have to think about it probabilistically. I said, what's the probability that they're going to A) get there. And then you assign a probability to it based on what you think their execution capability is. And then you can sort of think, okay, there's a value to the company at that point in time. I mean, ultimately all companies are the sum of discounted free cash flow into the future. So, you have to be able to generate cash because if you cannot generate any cash, then shareholders have nothing to take home, right? So, you know, the company might be losing money today, but then it basically just means a lot of the value's in the tail. I tend to focus a lot on optionality because if a company is able to, you know, make the optionality to work for you, then it will have a long tail, really, really long tail, right? And a good example is Apple, right?
Anirban (23m 39s):
So, for the longest time people would have said, oh, Apple is like Dell computers because it just makes a type of computer: call it a smart phone or the Mac. But its tremendous optionality in the sense that, well, it has a billion or nearly 2 billion people stuck to this device or various types of devices that they sell. Therefore, they had the option of making other devices, which they have success. Now none of them have been as successful as an iPhone, but that's an iconic, like, you know, probably the most successful device of the century. So, you don't have to repeat that, but you have the option of producing an iPad and an Apple watch and maybe something else, you know, Air pods. Each of those devices are such big successes that if they were independent companies, those companies would be like, "Whoa, you know, this is awesome."
Anirban (24m 19s):
Just because it's Apple, it's like, "Oh maybe it's not that awesome," right? And then you can wrap around. So, this is the optionality, right? You know, once you have lots of users, you have lots of different devices. Well, they love it. Therefore, they can have services. You just have the option of doing different things. So that's optionality that you have. Optionality also comes in the form of technology, right? If you have key technology that you developed, so they have an operating system that they've developed, they're the largest chip maker in the world, right? People don't realize this, that Apple is the largest chip maker and it's also the most advanced. They don't make it, they don't fab it, but they design some of the best chips, probably the best chips in the semiconductors in industry. So that's a skill, right? And that's unparalleled in some extent, because you know what they could do, for example, with the M1 Mac, others couldn't.
Anirban (25m 4s):
It blows away then, you know, the Mac is growing like crazy. Why? Because they could design these chips that give people this war experience, right? But you see, you could have seen this coming because that's not the first time they designed the chip. They have been designing chips for the iPhone for like a decade that designed their own chips for their wearables. They've designed the chips that go into the Apple Watch or the Air pods, right? So, you know, you could see that it's going to come to the Mac. If it's there in the iPad and everywhere else, it's going to come to the Mac. Once it comes, it's going to destroy the industry because they've destroyed the industry for everything else, because they're just so good at it. And then why is that important? It's important because if you want to get into other things, like wearables and new technology, the ability to control key components is really a big deal.
Anirban (25m 49s):
So, here's maybe one takeaway, some of the greatest companies of our time are going to those companies that control the key technologies that they are engaged with, right? So, if you think about Apple, it controls everything from the mobile system to the operating system, you know, mobile platform to the operating system to the underlines it, to be chips that are running on their devices, to main applications that they run on them, right? They are controlling the experience. And that allows you to do things at a much faster pace than to, you know, sublet it to other people, to do it depending on other people's roadmaps and things like that. So, insourcing of tech is really a big deal. So, if you find another company that has a similar sort of approach, and if they're really good at it, not everyone can be really good at this, right?
Anirban (26m 32s):
Because it's really difficult to be good at so many different things at the same time. But if another company is as good as this in their extra space, and that's an opportunity, right? Because probably that get into other things and be successful as much higher, right? So, let me make a sweeping statement, I'll say that, you know, potentially the greatest AI company for the next decade is not what people usually think it is but it's probably going to be Tesla, right? Because it is trying to solve the most challenging AI problem that exists today. General AI problem that existed. If it solves it, it will be able to solve many other problems that other people can't solve.
Phil (27m 11s):
What is that problem?
Anirban (27m 12s):
Well, the full self-driving problem they are developing this general full self-driving solution, right? With a very unique approach to doing it. Solving that is so hard that along the way you'll be, you know, not just you have the people, you know, still have the, some of the best talent, you'll have the datasets, you'll have developed a pipeline, you'll have developed lots of algorithms and a lot of intuition. A lot of other thinking that goes into this, you know, you're developing your own chipsets for this. Those skills are going to be useful for developing other things, right? I call it the Apple roadmap. A lot of the roadmap would look very similar is that, you know, you need to own and control critical technology. If you don't control critical technology, then you're at the mercy of other companies or other situations.
Anirban (27m 54s):
Those other situations can act against you and then you're in trouble. So that in my opinion is the biggest moat. The biggest moat you can have is controlling the key pieces of technology. Everything else is probably secondary.
Phil (28m 4s):
My head's up in the stars and up in the sky is at the moment. I'd love to even go on and talk about SpaceX, but we better bring the interview to a close. And if we can just bring it back to a basic level. And what advice would you give to someone just starting out in the market? And, I'll just put this in the context that maybe 20, 30 years ago, people would be investing in things like banks and things like oil companies and fossil fuel companies and so forth. Things have changed a lot since then. And you're very much looking into the future. So, within that context, what advice do you give someone just starting out in the market?
Anirban (28m 43s):
So, I guess the number one thing is to be open-minded about change, right? Because the one thing that's constant in the world is change, right? So, we need to be accepting of change. Now that doesn't mean that like a bank, for example, is not a good investment. At the right price, anything can be a good investment so, you know, a bank can be a good investment, potentially it has been a good investment, for example, Australians, right, have invested in the bank. So, I think that's number one. Number two is I think a lot of investors would benefit from reading the annual reports of some of the big technology companies out there today. So now I'll give an example. I love to give this, if you are interested, for example, in financial technology, then you should definitely read the 10-K, as it's called, of MasterCard and Visa.
Anirban (29m 25s):
Because once you've read the 10-K, and not really even into the intent of investing in those companies, you don't have to invest in those companies, but just by reading this, you'll understand how the payments networks and how payments actually work globally. That knowledge is immensely, immensely useful. Similarly, if you're interested in the ad market, right? Like the digital ad market, you should definitely pick up the Google annual report or Alphabet annual report and read it because that's going to tell you a lot about, you know, how the ad market works, you know, what does a traffic acquisition cost mean, for example, right? You know, what's a TAC, what's a DSP? All those things are terms that you can learn. And it's a very good way to learn, to learn from the best. You can ignore the stuff about their particular business. I think that's the way to learn.
Anirban (30m 6s):
And so, there's a wealth of knowledge out there that one can get, which is really from the annual report of some of the best companies. You don't have to read it from an up-and-coming company, but if you want to, for example, understand, you know, software as a service, then maybe pick an established software as a service company, like say salesforce.com, right? And that'll give you an understanding. So, I think there's a lot of opportunity to learn. And as the more you learn, the more confident you get and of course, then you can waddle in the pool a little bit and then try to, you know, go into the river. Once you're comfortable going into river, then maybe you can try, you know, paddling in the ocean, right? So, baby steps is the way, but I think, you know, just read. Reading a lot really helps.
Phil (30m 43s):
Anirban Mahanti, thank you so much for joining me today. It's been a real pleasure.
Anirban (30m 47s):
Pleasure's all mine. Thank you, Phil.
Shares for Beginners is for information and educational purposes only. It isn’t financial advice, and you shouldn’t buy or sell any investments based on what you’ve heard here. Any opinion or commentary is the view of the speaker only not Shares for Beginners. This podcast doesn’t replace professional advice regarding your personal financial needs, circumstances or current situation