AI, Advertising, Entrepreneurship and oneAPI - How SiteMana Innovates Solutions Using oneAPI

Tony [00:00:04] Welcome to Code Together, a podcast for developers by developers, where we discuss technology and trends in industry. I'm your host Tony Mongkolsmai.  

Tony [00:00:18] Entrepreneurship is one of the most critical parts of the technology industry. One of the topics we've been talking about this year is the journey from developer or technologist to company founder. Today we're joined by Peter Ma, who has founded multiple companies and will talk to us about his company Sitemana and how oneAPI is enabling his company to move quickly in the dynamic AI and advertising space.  

Peter [00:00:38] Thank you for having me, Tony. So I guess a little bit about myself. My name is Peter Ma. I'm a co-founder of Sitemana. You know, I'm part of the Intel Software Innovator since like 2014, seems like almost ten years now. And outside of that, you know, I've been doing a lot of different type of projects in the AI community.  

Tony [00:01:01] Yes. So how did you come to the idea of Sitemana? And what exactly does it do for your customers?  

Peter [00:01:09] So we basically we basically do AI prediction on the user's buying intent and we kind of use 200 different type of parameters ranging from, you know, like online behavior, like how much visitors they have, how long they visited was the duration between each visit, the stuff they click on, as well as offline, you know, information such as like their their credit score, their, you know, their their ethnicity, gender and then demographic. So to kind of compile a prediction score on how likely they are, you know, to making a purchase. Once that happens, we can kind of know what their email is. And this way, you know, the online shop owners can kind of retarget them on the specific, like on the specific thing that they'd like to purchase.  

Tony [00:01:59] So that's a pretty cool idea. I know that there's a lot made of things like Facebook being able to do targeted ads, and from what you describe, it's kind of like saying you can. Look at the way a person's interacting with the website and then actually decide whether or not they're worth sending some type of information to.  

Peter [00:02:20] Yeah. I mean, this technology isn't really new. It's just being used by, you know, Amazon, Facebook, Google for, like, a long time, right? That's the sort of like the power of gods, you know, like the Greeks. You know, they have the power of gods. And but the thing is that the regular shop owners don't really have that power in terms of in terms of like competitions. So we're kind of trying to help the smaller guys to compete in a larger scale.  

Tony [00:02:48] So as you do that, how much of the effort for someone to integrate Sitemana with their website, how much effort is there in terms of like, do you have to integrate with their website directly? Is it something they can do separately as a service? What's the amount of effort needed if they wanted to integrate something like that?  

Peter [00:03:06] It's like one line of code and in the our customer can literally do that within 15 minutes if they like on our setup calls, usually, you know, within 30 minutes includes explanations of what this with us as well as, you know, the things that they have to do to be a privacy compliant and others.  

Tony [00:03:26] Wow, that's pretty cool. Actually, I was actually thinking from my experience with web design and I was thinking, you're going to have to add in all of these hooks and stuff like that.  

Peter [00:03:33] Yeah, we do the hard work and they kind of this is essentially a more powerful tool. Like there's analytics tools and there's see our old tools. We sort of consider ourselves a CRO Tool. As more of a conversion rate optimization, right? You, you're, you're basically like the amount of traffic that come in the how many you can convert, you know how many of them you can convert into a sale and we are basically the filter of like what's a high purchasing intent customer. That's kind of what we're trying to do.  

Tony [00:04:08] Yeah, and you mentioned optimization and obviously you're part of the oneAPI for Startups program. What type of optimizations does using oneAPI give you as you roll out your product to your customers?  

Peter [00:04:21] So I mean been doing AI since like 2012. Yeah, there's a lot of different methods on it. There's, there's just, I would say in the problem with the AI industry, there's just too many libraries and too many things. And this really help us to hammer down like all we need into one single library, righ, which is oneAPI.  This is from training down to deploying in, there's like down to pretty much streamlining this. That gives us the advantage of it's just like, hey, like I just need to I need, I need to get this up and running in production within like 48 hours, then this can be done. And if you were to do the traditional style, it's like this is almost impossible because it's just piping, just piping from TensorFlow into TensorFlow Lite. That took a while. Right because that that that library and every time every time TensorFlow, I mean don't get me wrong we still use TensorFlow for most of our training purposes. But the thing is every time it updates, you have to update the library and then think of as backward compatibility issue. And then the on the inferencing issue and the whether the inferencing is like is compatible. You know, it's those are I think the major overhaul for for like AI developer and the one if you are going to solve all of those problems, it becomes a very optimizing for us as a developer. And then also we can optimize our traffic of our customers.  

Tony [00:05:48] So what type of benefits does you do your customers typically see? So now, instead of somebody just coming in to a website like I would go to, to a website and like I would click through different things, you know, and if I wanted to buy something, I would go and buy it. But you're saying you're giving people kind of a CRO, this Amazon like or Facebook like the ability to look at your customers and determine what they really want to buy or how likely they are to buy. What kind of uplift does this give your customers?  

Peter [00:06:16] So so that it's more we are more of a personalization. I would say there is talk that's like there is a lot other tools are specific to analytics that is based off like figuring out which what product people likes to buy, what product people want to buy and things like that. But from our own, we kind of don't want to compete in that market or cause it to be right in the beginning where we're trying to, you know, we were trying to do that, but it was just not a it's just it's a it's a very red ocean. But what we want to do is personalize down to a grain of who wants to buy what, right? It's like, hey, Tony wants to buy this gaming GPU. Like, that's that's that, you know, and then you basically supposed to target that person for that. There's different tools for I mean, modern day is, you know, there's different, in adtech, there's different tools for retargeting, like we specifically focus on email targeting as retargeting. There was the first line is basically the people with a higher intent generally focus on the email retargeting because then, you know, because they are more likely to make a purchase and you want to because you want, you only want to email relevant content. The this is avoids is avoids you you know, sending unnecessary emails and as well as you know you don't want it you don't want people to see this as a spam. This is why the purchasing intent is super important. And second is basically the ads retargeting because modern day when you use Google Pixel and Facebook pixel, they literally just retarget practically everything. It's like if you visit a site, they start retargeting. And so when when I was talking to like, like the ad industry, the current problem for a lot of bigger advertisers is no longer trying to focus down to was nitty gritty of, of a person's purchasing intent but rather like the true problem they're trying to solve was like you know the the last scene right who gets to put out ads before a person's eye, before the person make the purchase? That's what they're fighting for. And if you look at that, this is really not sustainable in our perspective. Like we want our customers to be like only, you know, it's like only focusing on those of those who have intent to make a purchase. Um, I think that's in the longer term, that's a much more sustainable way. And then obviously there's other ways of retargeting. Retargeting such as like, you know, you can mail retarget, um, as was a billboard or a target like, you know, based on that, based on the person's personalization, this, this is, this is quite important. So, so you see that remember, remember when you see the thing is like where you when you visit a gas station and when you, you know, pumping gas and you see a little TV screen of people are fighting for the screen space right now. This is this is you start understanding why these things are happening.  

Tony [00:08:50] That's funny. Every time I go to the Shell gas station by my house, they keep telling me about how ice cream was something only for rich people back in the the dark ages. Now I should buy a FroPop or something like that, a frozen pop. So, yeah, now that definitely makes sense. How did you come up with this idea? I know that you've had a couple different startups, and I'd like to get into kind of your experience as an entrepreneur, but particularly for Sitemana. How did you decide this was an area that you wanted to to work in?  

Peter [00:09:24] So the we this was actually not even our original plan. The so, you know, my last startup wasn't exposed. We kind of raised a seed round and then that was like in virtual fitness space. But you know, as, you know, like the pandemic gives pandemic takes, right? It's like is like we see the traffic goes up during pandemic. And then we were really happy with the growth. And, you know, we raised a bit of money and then all of a sudden, you know, people could start going back to, you know, regular gyms and stuff like that. You could just see the rate drops. There is a way to lie to yourself, like, hey, you know what this thing is like? It is in over long term. You know, that, you know, the virtual fitness will always have a space. But in in terms of running a company, that's not enough to to to see how, you know, this is just not enough to, um, to see the steady growth. And as a startup, you either grow or you die. There's like there's there's really no middle ground. Oh, it's not a mom and pop shop business where you can just, like, hanging, right? This is where we kind of, we kind of wanted to work on, like, next idea and then, and then the guy who was a hand. So my friend who I like, like we met over like ten years ago, back in LA, he was working on the ESP startup. Um, so he was basically like, so we in the beginning it was so he was able to able to retrieve the technology, which we can kind of see what the email of the person are through this. It's the same, the same technology that Google and Facebook uses, you know, over, over aS technology. We do want start stuff on top, right? And then what we were going to do was the well, we were really going to do was a AI based testimonial given. So if you're like a grandma or you go to a like website and you know, this website would be like, hey, there's, you know, Grandma will give you the testimony. This one is great if you're like Asian, you know, male. And then it's like, you know, you know, thirties and Asian male in the thirties, Hey, this wine is great. So essentially the idea was that you are more likely to purchase from yourself, right? Because if you look at someone that looked like you, you know, where you basically from there you can get their social network. And so you can, you know, you can scan the photos and see which and then you use that to a four for photo comparison between which testimonial like like which person will give a testimonial look mostly like you. Right. And if you can't get that, you can still get Asian ethnicity and then you with that demographic and then you match started with the with the person will get testimonial and if you can't get through that you basically give a general. And then people was really receptive the idea that we think we signed up to like almost 50 like the prepaid, they paid us three month prepaid plan without a single product, without a single line of code being written. And we're like, okay, this is great. This is going to be next startups. Everybody likes us, everybody wants it. So we end up building it. But as we build, the testimonials start coming in and we're start seeing those like, okay, our original premise was like the testimonials at least be giving as good as these influencers. We didn't expect a professional quality, but we expected like at least influencers, the influencer just use your phone, you know, it's just like every day, like, like it. But the thing this testimonial come out is like they're just not very genuine. Then they're like, just like in the video, quality is just like, really bottom. Like this can convert anybody. This can't even convert me. It doesn't even look convincing. And, and then we started with like, I was like, What if we get you the testimonials and then we start realizing, okay, this is what the entire ad ad industry and then do, right? The actors actually is more convincing than the real buyers because this is the entire industry built on that. And in order to find you really and then the influencer industry is best to build off of that backbone as well as because essentially they have the they have the charisma, they have the they have the convincing act that this is a good product and then and then you to take on that industry, it was like this is by a single regular person is not that simple because just a normal person is not very convincing. There is this is just that's why they're normal person. They're consumer, they're not influencer. And then what if they do that? That's entire different startup that you are basically solving part of a challenge and we're like, okay, let's just scrape this. This is not like this is not something that, you know, this is not something that we want to do because this is the entire different tangent of the idea. Then we have to discuss what else we wanted and our customer was like, Hey, weren't you using emails to get those information. We're like, Yeah, like, Hey, can you just give us the emails instead? I'm like, Oh, that's much more easier. And then we just kind of we swapped out like, like 80% of the code. And then this is how this idea came to be.  

Tony [00:14:01] Okay. So you're always kind of searching for that, that next idea that your customer needs.  

Peter [00:14:07] Yeah. Yeah. So you see, the one thing with a startup, right, is people think, Oh, I have this great idea, blah, blah, blah. It's not really about your idea. It's about what people want to pay for. And then if you look at the history, when Google and Facebook came out and people like, oh, you know, you don't need revenue to look at Facebook, I'm like, dude, Facebook, blew up entire Harvard server within a day. And Google did the same thing when they launched back in Stanford. So it's like you we are more like we are we demand more of the Google and Facebook and ChatGPT and even Microsoft Windows. So like, you have to see whatever you build as an entrepreneur or whatever you build is like how much of how much of the market is really pulling you rather than just like, Hey, I have this great idea. Like, let's see who should use this? And with the first startup, none of it, none of us really. Our idea is all based on what the customer wants and we're just a solution provider solution for, you know, we solve customers problems.  

Tony [00:15:06] That's a good take. It's really interesting because I talk about this... I've been trying to talk about this with the oneAPI for startups program people this year and, and trying to help developers understand how to bring their ideas into the startup space potentially. And it is interesting because most engineers generally like to take their idea, which they think is a great idea and it could be a great idea and then make a product out of it and then create a startup out of it, which is sounds like kind of what you guys were doing, originally you had your idea. But the hard part for many engineers is then saying, My idea that I thought was a great idea apparently is not a great idea. Do I still want to be a startup? And what you've kind of described is that you enjoy coming up with ideas, but at least you're willing to morph to fit what the business needs. And I think that that's a real challenge for a lot of engineers.  

Peter [00:15:55] Yeah, you have to take your own ego out of this because otherwise, because of that, the customer is the one that pays. And they don't pay, then you don't. You go hungry. That's if you have that. If you have that right up front, then you take your entire ego. It's like you no longer get attached to any of these ideas per se. You're just a solution provider to. You're here to solve someone else's problem. Right? Because it's just like. And the more problem, the more people's problem you can solve, the better your idea is. Just basically put it that way. The artist in the beginning of the career cannot really like focus on what the artist themself trying to express. The artists in the beginning have to figure out what, what, what, what the audience want. Once, once you build a once you can build a reputation as an artist, then you can express yourself because you're because there is a there's an intrinsic. I guess it's like after a couple of fails felt about I mainly knew what people did not want through hackathons because every hackathon I like every hackathon, I won 150 hackathon, every one of them people are like, Oh my God, this is a great idea. But then again, nobody's willing to pay for it, right? And then you can be have a very crappy idea, but yet people are willing to pay for it. And this is what difference between a business and like you're in one of these side projects, right? That's kind of a thing. That's a that's one of the thing that we need I think as an entrepreneur, as a developer, they need to keep that in mind and all the tools that, you know, like AI, mobile, blockchain, all these tools are just amount of things in your toolset because you know, inside the toolset that this allows you to, this pretty much allows you to it's like you have a hammer and then you have, you have a saw, right and you've got your chop some wood, you build a house, right? The house is essentially what people want, right? But the hammer is a tool like the hammer. And a hammer could be the tool for the hammer, could be the AI. And the saw is like the ChatGPT. I mean, these are the tools that you possess to build a build. Something that people want is like a house. But like, I want this like, you know, I want to build this weird cave. I'm like, that doesn't like nobody wants to live there, you know, you want you want to build the house where people people want to live. It's like two stories, you know, especially with a giant glass window we can see outside. You know, that's that's kind of the, you know, as an engineer, as a like as an entrepreneur, that that that's the way I see this.  

Tony [00:18:20] So you mentioned ChatGPT, which is obviously still in the zeitgeist. It's interesting because things are moving quickly and sometimes things come into general mind share of technologists. But ChatGPT has obviously captured the mind share of everybody just given kind of how accessible it is. How does that influence what products you build? Is it something that you guys are looking at saying, Hey, maybe I can integrate this even for better customer experiences? It's something that's just an interesting technology to you. How do you perceive kind of like something like a ChatGPT and how large language models are being used?  

Peter [00:18:55] So. So there's two things I want to talk about this topic. First is the there is definitely demand in our market. Like all like almost all of our customer wants this when we explain to them. But the issue we have is this whether this will actually retain because this is like the newest candy that you'll find in town, like everyone wants jump, jump on it. But whether this thing will actually generate revenue for our customers, we don't know. There's still it's still toward the other, the other technology is the stack is still more important like what their demand is because there are stuff that immediately drive the revenue would build this feature out and then we know this you know revenue will go up and you get paid or you give yourself a bonus. You can start hiring people. I mean, that's just you can expand the team, You can you can, you know, you can you can you can do a lot of things with money. But on the other side, it was it is an interesting technology. That if it basically ever proven to work for for for our customer on certain things because it can definitely help like it can definitely help the transition into a more optimized world, I would say, because right now the thing is but the thing of the second part I want to talk about this activity is like it is a new is like it is a newest toy and just like everything have waves, the job of a being an entrepreneur and being a technologist is is not to basically drop you driven by the waves. It's like one I like. I don't remember the hackathon days because I was a rapid prototyper, right? Because I was the one that really interesting to every new technology that comes out. And even today I still am on. But in the hackathon world is, oh, you know, like, like the blockchain is all this build everything in blockchain, everything, you know, in this world that would change with the blockchain. And then and then next year and then you see the crypto rise, crypto fall, and next time you see the AI, oh, everything with machine learning, everything will use this. And then you can see none of the startup actually like build money of like, like okay, can you name five AI companies that that that you personally use and pay you know very little.  And Stable Diffusion comes out of you like oh that would change everything and eventually it's like everybody put all these resources in and very few people except for MidJourney  and a couple other company was able to, you know, make money off the Stable Diffusion. Most companies faded and the ChatGPT is like a newer version of that. So one of my friend was like dude just within past day he got pitched by 75 different ChatGPT based apps and then was like, how many of these will really survive? It's like the like I as a developer, you kind of play as a tool, but as entrepreneur you had to feed people. You have to basically you have to think about like, Hey, I've staff to feed like, like I have to think, you know, and the thing doesn't survive. All of us will be out of all these. It's not really like us ourselves. I think this is the other part and we discuss on next is like people don't really understand like startup founder have to go through. It was like, you like ourselves. Yeah, like I you know I can be, you know, whatever but the things I, you have to cover your investor you have to cover your, your, your your staff and then when they when, when they're out of a job, you basically it's kind of like in the way it take a toll on you to, you know, because end of the day they kind of give you their sort of give you their trust, give you they're like, if they believed in you and then and you let your believers down in a way. And so so that's kind of the thing. It's like like a lot of this is why the people say, why can't why, why, why is nobody willing to fund me? Because you don't have that and you don't have this sort of like mentality. So you have to you have to, you know, eventually because to do startups, you know, for developers especially, is that you have to start thinking about like how to scale this as a company, not really just for yourself. And I had to learn that hard way through being a consultant and into like, you know, actually raising money and and funding the startup and dealing with all the problems you're like and never have to deal with a consultant as a consultant is like, I mean, like as you know, it's like you build the technology and if you don't, if you don't like the client, you don't like the project. They just like, I don't want to do this. I'm sorry, This is just not my expertise or this is too boring for me or even just like you just hire someone like, you know, offshore or someone less skilled that's much, much cheaper than me. Like, why are even bothering me with this? But as a startup, it's like everything you build, especially in the beginning, it's like it's you have to hit. You have like every little mistake and cost you because it can cost you morale, can cost you like you can get in depression is nobody wants the stuff that you build and it's like, Oh man, I missed it again because the easier to just let customer demand what they want and just give them what they want.  

Tony [00:23:29] Yeah, I think keeping it sounds like you're saying just keeping focus and not being distracted by the waves that come in and out of technology, which I think is very true because like you said, there's there's all kinds of startups that seems like nowadays that are being built on basically the same technology and just saying, I can use somebody else's technology to solve a problem in the space, so I deserve to be a startup and get funded. But the reality is you're not necessarily innovating anything that has a ton of value. You're just wrapping something that somebody else did and you don't even know if that market exists yet. And so for you, what you kind of just described it as I understand it, is knowing what customers you're trying to hit, what they care about, and even though they may ask you how do I get something cool that I heard about into my product, you need to stay focused on what actually can make them money. Because if they're not making money, you're not making money.  

Peter [00:24:24] Yup, exactly. And then all all these new R&D stuff really can come later. It's a very different way. A mid-size company, you have extra resources like, you know, why don't you work on this and stuff like that? And then and then and the other part is like people, I think in startup founders are like, like most people don't realize it's like, hey, like, like, like as a staff that like I talked out about this just yesterday with another friend is like, the staff is always, Hey, I deserve to get paid more and stuff like that. It's like as a startup founder, you're literally want to be like, Dude, if you can do my job, I'll pay you more than my pay myself. I'll literally pay you more than I pay myself. So you kind of free up my time so I can actually go explore ChatGPT and stuff like that, which is what I enjoy. But the thing is that that doesn't feed you, that doesn't feed me, that doesn't like, you know, we have to feed everyone first, right? And then and then when you take investor money, yes, you're feeding yourself on the investor money. But that runway is very short. That's just like you're looking at a limited time frame as like I have a year and a half to do to really get this done. And then it's like everything is like time is really ticking, right? I would say is being a consultant or like being a developer and a consultant. It's like doing a startup without accountability, right? And that which is really fun doing things without because I was doing that like this is majority of my career. That's kind of what I did. But as soon as you run a startup, there is accountability, especially if you're a founder and the CEO and you know, so that could just like the two or three founders together is like there is accountability, is accountability to your staff, to each other and to your investors and to your customers. So this is like all this thing you have to manage. I would just say like try to manage the least amount that ideally is just a technology and customer it if you can get that done, this is like, great, That's just like really, really great that you don't have to deal with all of this, all the other stuff, right? Because that's just you have too much to do. Not not too little. And the limit on play with the new technologies. But but, but the question is, you've got to be honest with yourself, right? Because the magician cannot...when a magician believe his own magic it's done.  

Tony [00:26:34] And you've been doing this entrepreneurship, being an entrepreneur for quite a long time. Like you said, you put you were part of the Intel Innovator program. And I think Intel Ignite  program probably too. And now you're in oneAPI for startups. You've been doing, you've done multiple startups, but you also previously worked, you know, at a company as a consultant. Why is being an entrepreneur so interesting, exciting for you that you want to do that versus what I will say most people in the world do, which is not do startups? there's plenty of people that work at Google. Microsoft, let's name are big companies Amazon, that are happy to just work at a big company, but it takes a different type of person to be an entrepreneur. And most of the time the people I know, they're entrepreneurs and they don't go back. So what drives you to be an entrepreneur and not just a technologist?  

Peter [00:27:25] Oh, well, it's that you... There are some people like to be part of the system. I just. A majority of people like to be part of this. They like the security. The to be entrepreneurs, first of all, is like the risk. The risk. I would say like the risk tolerance. Your risk tolerance is very different than everybody else in the company, because as a consultant, I would say like as an employee, I think if you value security, like you want to get a paycheck to paycheck, like like, you know, paycheck arrives on time, you have extra money, you're on your mortgage, you know, a couple of kids and things like that. That's great. Don't get me wrong. That's just what majority of our society is sort of built upon. And eventually we do need those type of you do you need to hire this type of people because they their they values stability and then then the company eventually have the values, stability and growth like, you know, growth margin time over time. Right. And that's kind of but this not like who I am in a way, because once you start doing entrepreneurship, it's like due to you see the excitement in and you enjoy you enjoy in the companies of other like founders, right. Rather than you know as well as investors as well is you look up to a certain type of people. It's like, you know, you look up to like Steve Jobs and, you know, Musk and it's like these are the people who you want to be. And then there's like, there's no way you can get there if you're like, you know, in in a corporate job, like, no matter where you work, unless, you know, unless you work for D.O.D. and and it's even that you're not you're not the Secretary of Defense. You are you are the you're, you know, just a contractor at D.O.D., which is nothing wrong. But the thing is, this is this is a different tangent of humanity, because is this is what literally you need that to really build a foundation of the humanity. We can all be entrepreneurs because the thing is like, like, like being an entrepreneur and unstable and being a family man unstable, that's that's almost irresponsible, right? Because you that, that basically that you're you're basically dealing with you know, you're you have to see your kids. You know, your kids might go hungry and things like that. Right. And that's not like that's just not something that I would recommend if you're like, wow, this is something you value. I just that's not something I'd recommend or anybody would recommend you to do, because this is this is because, you know, we need people to run the society and that doesn't mean you can't be great. Nikola Tesla was never good entrepreneur. He was a really great scientist and one of the best workers. But the thing is, you, you know, it just that, well, he works for Edison. You know, Edison took his credits for a lot of stuff that he did. And then, you know, that's a separate story. But the he he but the thing is, he's still remembered as one of the greatest men that live. Right. Like one of the greatest engineers, you know, then there's countless examples of that. It just I if you if you if you if you enjoy the risk and if you enjoy the if you enjoy the getting the you know, getting to the market. Right, right. This is where I think there's a little more difference like I you enjoy that you enjoy solving people's problems so much, that you do now want to you know, you do not want to you want to start something actually, actually, you know, to to, to to actually solve a specific problem that when you work for someone, they may not give you the opportunity not to. That's when you because they give you look at certain companies. Right. Like especially like especially FAANG type of companies they have internal entrepreneurship. Um so within the company they, they, they, they basically reward you for finding they're building out different tools internally and then possibly like, you know, possibly like you basically compete for the political upcoming because you created certain things within the company like Chrome as best example right. This is just like that's created within Google. And that was one of the biggest problem as Sundar as such the CEO of Google now. But that's you know that as we go you know that some that some people enjoy that things and that it's great and they're just like they're just like they have a risk tolerance within the corporation itself. But the thing is, most people don't when most people don't, they just, you know, you look at so you look at other tangents of how can you be great at what you do if your engineer is like, you know, GitHub is sort of your your bragging rights, right? Is like this is what you do. And also the amount of new libraries that comes out that we all look up to, like which really this is the stuff that or a new model like ChatGPT is like this is usually a new type of model that you generate. This is the stuff that you know, we really look up to. But if you have ability to monetize those or thinking about monetize, I think in ways of monetizing a certain type of technology, this is where you can start considering entrepreneurship. And by thinking, I don't mean that thinking about how to all your ideas. Great. I'm talking about like, thinking how this specific type of technology can solve somebody's problem that's willing to and that can sustain a little team as well as eventually a bigger team and bigger team.  

Tony [00:32:25] All right. So that's almost about all the time we have today. So like with most of my guests, I want to ask you, where do you think technology is going in the next couple of years? And what are you excited about coming from technology the next couple of years?  

Peter [00:32:38] So in the next couple of years, I think the next big thing you will see is a text text based video generation. This is what NVIDIA just annouced look for them for the newest demo and then you can Google just you know, Sundar kind of did that in the in the video. We can see because this is where this is where you can see that you it's kind of where you can see the computing powers again pushed to another limit because you need the ChatGPT to generate the text for thing for the entire dialog and then you need that stable diffusion to generate the textures and then you need to go back to ChatGPT to really generate the, the script for, for Unreal Engine or other gaming or, you know, or other Unity or other engines. Right. And then deploying that and then render that through the gaming engine like Unreal. So and then that will be the final outcome will be a little video and now will be that. That's pretty exciting to see. And of course it will take it will take another maybe couple of years before this thing can perfected itself down to the grain, which is like a you can have like a Pixar level on it, like animation in a matter of no time. Right. That's that's kind of the that's kind of where I see this kind of exciting opportunity to then and people who again, when that comes out you will see like a thousand startup doing something like that. But the thing is you have to see what people that's doing that right now. Right? And this is way I have to backtrack how who's doing that right now, because they but they will possess the expertise and know how of how because they are already meet the majority of the engineering is really a problem that you meet. And then once you go cross those problems is like your asset as human assets is how much stuff that you can deliver and the how much how fast you can deliver in the how, how, how much better you can deliver. Because the people who are doing that right now is definitely running the type of problems that, you know, when these things become public, like you have tools, like, you know, when oneAPI can generate that stuff, it's like, you know, because that give access to everyone. But then again, the people have been doing it right now. They will have like they will have a special they will have specialized tools like too specifically like shading or something that's like and then then they can, they can literally just use a AI to generate like a 3D games within, you know, this space because they already have the tech that's like the foundation of the tech and they have understanding inside of the technology. I think that's the most important part and I think that's most exciting to see as well.  

Tony [00:35:04] Yeah, that's very exciting, actually, said one of the producers of a Marvel movie said In the future, Marvel movies could be generated by AI. Well, thanks for joining us, Peter. We appreciate all your time and your insights on being an entrepreneur and learning a little bit about your company Sitamana.  

Peter [00:35:21] Thank you for having me, Tony.