Startup Lessons from Asia’s Leading Founders: Ibrahim Orhan & Salma Dias Saraswati

I think what I can, what I can think of clearly is that start with the people close to you. I mean, when it comes to scaling to a whole new place, a whole new country, what will help you is the one person, having one person that really knows the market, really knows the place. So before going there, you have to have at least one person that you can talk to for that market, about that market, about the users, about the problems, so that you don't waste your time doing experiment, doing, try to do this or that, that doesn't work. So I think having, having like a very, a person that knows that market really well, and then have relationship with them. And then at least, or if you can have them on your team, for you to scale or just being an advisor or something like that, I think it helps.

Hello everybody. So let me introduce about myself. I'm Lan Chi, I'm the country director of Foundry Institute Vietnam. So I'm actually, I was I started my startup journey since a long time ago, 12 or 20 years already. So in the country, I am considered as an ecosystem builder. I joined in a lot of startups, so recently a lot of AI startups as well. So today I have the chance to talk with our two FII alumni. We will talk firstly to Ibrahim Ohan, the CEO and co-founder of Canaria Tech, an AI startup for robotics developing perception models for mobile robots. Can say deep tech, and he introduce about himself later. And welcome Salma Dayats Sarawaswati, the CEO and co-founders of Senar AI, the number one empathetic AI-powered mental health co-psychologist for Gen Z in Indonesia and Southeast Asia. So hi, Ibrahim. Hi, Sanma.

Hi, Jiang. How are you doing? Good to see you.

Good to see you.

Hi, Jiang.

I think, ladies first. So I will ask Sanma to share about yourself, right? And about your company, about what you are working about, right, recently to the audience. So please, it's on you.

Okay, yeah, thank you, Gian. Hi, hi everyone from Foundry Institute and all the people that's interested in here. So my name is Salma, my background is actually a psychologist as well, and I've been also building Tanang for the last 2 years in Indonesia. So Tanang, it means calm in English. So in short, yes, we're building an empathetic AI co-psychologist for Gen Z, and we started in Indonesia. And what we're building is that you can basically chat with our AI as if you're talking to a psychologist. So we are trying to create a first step of care for people to seek for mental health support so that they don't have to schedule beforehand. They can get help literally 24/7, immediately, and in a fraction of a cost. So it's very cheap and We believe that in terms of mental health, it's a problem, especially in Southeast Asia, knowing of the stigma. So that's why we're building this for the last, uh, going 3 years now. And I think we are serving more now, not just B2C but also B2B, in the university hospitals and also companies. So that's basically pretty much about what we're building.

Thank you, Salma. So now it's your turn, Ibrahim.

Yeah, thank you, John. Um, so a little about myself. Originally, I am not from the field of AI or robotics. I started my journey from synthetic biology, which was very new when I started back in the day. So it was more academic research and turned into some projects. And actually, funny thing, it even, it even involved some fundraising. I was I was 15 years old and we needed a sponsor for our project to be able to do. And we were doing fundraising back then. And then after synthetic biology is very related to genetic engineering and bioinformatics. And from bioinformatics, you have to use a lot of AI and you have to use a lot of machine learning. That was my entrance into the AI world. And then in time, more, I found myself more in the AI realm, AI world where started to work in an AI startup from America trying to do business in Japan. And I was in Japan. And after that, that company bankrupted, which was a great experience for me to see how a company can be born, fundraise, and fail. And then that gave me confidence to build CanaryAttack. And with CanaryAttack, after a couple of pivots, we found ourselves in a very sweet spot where we are happy where we develop perception models, like foundation models, almost, you could say, almost like a GPT for mobile robots. And when I say mobile robots, I mean robots with wheels. And the reason we need foundation models for these robots is they are, you could say, they're currently like 5 years old in terms of human intelligence, like 4, 5, 6 years old. They do all these like very innocent mistakes. They're confused most of the time. So what we want to do is we want to make them adult intelligent with foundation models so that we can put them in a more difficult situation, more different environments. We could start using them more efficiently. So that's what we are doing at Conary Tech right now.

Thank you. So as you may know, I Most of my startups, actually all of my startups are in Vietnam, you know, can say that. My first startup was founded since I was in France, but anyway, it is all still a Vietnamese startup. So I would like to know, you know, about how and why did you choose the country? So for Ibrahim, it's Japan, for Sanma is Indonesia, to start, why not other country? Why not Europe? Why not America? But Asia and maybe your own country like Sanma case. Sanma first, go ahead.

Okay, I think one of the basic reasons why we started Indonesia is because of my own experience as Indonesian psychologist living here. I think what brought me into building this is because of my own experience. And I think I know this market very well. I know the users, I know the problems. And because this is something that's pretty new to build— when we say AI mental health, it's not very common in Asia. It is well developed in the US and also in Europe, but in Asia, I think we, we can say that we are actually the first at building this. So I believe that in order for us to be able to see if this is actually a good solution to be scaled, we— I gotta start from where I am familiar with, which is Indonesia. And yes, getting, getting, um, going forward 3, 2, 2, 3 years, I think Indonesia is a good fit because the problems is there, the pain points is there, and I think we, we kind of we kind of start looking at other markets in Southeast Asia as well. The reason why is because mainly the similarity of the problems when it comes to mental health. I think it happens many in Southeast Asia countries, especially the neighborhood of Indonesia, as well as I think Japan as well, Korea. I think they, they have a very big pain points and stigma around mental health. I think that's where we also come into place. So yeah, the basic reason is the familiarity, the pain points that we see here in Indonesia, and also for the scale, scalable for the market in Indonesia. We have quite a big market. I think that's a good start for something that's pretty new in the technology.

Okay, so Ibrahim, Israel is her country, right? So it's much easy for her. But how about you, Ibrahim? Why Japan?

So why I started in Japan would be, you know, that answer would involve this personal story why I ended up in Japan as a person from originally from Turkey who went to university in the States. Why did I come? That story would be a different one, but I want to actually give maybe a valuable, like something that would be more useful to the audience here. That story might be something for some other time. So I was actually giving a speech on CIC, which is this famous place in Tokyo for startups to give pitches and all these different speaks. There I was talking about what we are doing and foundation models as well. And then the question I also focused on was why Japan? Why Japan is a great country. Actually, personally, and kind of logically, we believe Japan is the best country on earth to create foundation models for robots. And the reason is, there are a couple of reasons here. One of them is, first of all, Japan is, as you can imagine, especially Tokyo, is very hardware-focused environment where you can find so many hardware companies. And When you think about foundation models, you might not really think that, oh, hardware foundation, like, aren't you going to do the software? Aren't you going to do the AI? Why hardware is that important? Well, first of all, you need customers. And also, you're going to have to, even if you develop something that can work on simulation very well, you're going to have to test it outside. Otherwise, it doesn't mean anything. You have to show that it works in reality because there's a there's a fact called simulation to reality gap. So to be able to fix that, you need to actually use real hardware, real robot to do testing. So that's one thing. Another one is Japan is a very robot-friendly society. Like, you can see some videos from other countries, and I saw a couple of them from different regions where some delivery robots were being attacked by people because they were like, oh, you're stealing our jobs, kind of. Vibe. Japan is kind of the opposite. Here they need so much people, so many people, because it's an aging population. They, they're like, we need so many people. When people see robots, oh, thank you for saving our society, you know. They're more thankful for robots. They give the ways. They're always respectful. As a startup owner, imagine this, like, as a startup owner, you don't want to have the fear that, oh, some of my robots might be attacked at some point during the day, you know. In Japan, you don't have that fear. And also, when you're trying to build a model, a foundation model, a perception model, a behavior model where robots need to behave more intelligently so that they can be similar to human-level social intelligence, then you need data for that. Because it's impossible to teach it by just coding. You cannot do that. You have to show data. You have to show the real good example of What is a good behavior? And I would say, as a, as a person from, uh, Turkey, stayed in states and then visited other countries as well, Japan overall is the best country on earth to have the good example of people walking around in a very narrow, narrow environment with a lot of people. It's high population here, but places are not that wide. Which means people have to really do those little maneuvers to, you know, not hit each other and everything. It's perfect. And good thing is, when I came here, I wasn't— now I consider myself a good walker. And the reason I can say this is whenever I visit my own country, I'm like, people don't know how to walk, man. Like, I feel that. I wasn't feeling like this years ago. But after living in Japan for a long time, I also realized, oh, Japan kind of changes you. So as you can imagine, as humans, we can learn from the society in Japan. Why not robots? Why not AI? So that's why we say that Japan is the best country on earth to teach AI models about social navigation and being acceptable in the society to move around as robots because They have to improve their social skills as robots. You know, they cannot be just obstacles all the time.

Okay, thank you, Brian. So surely we can understand that you start from that country, right? But how about the future? So now we start and we really need to focus on the market, surely. But in your opinions, this country won't be your first market or your testing market, or you will be your main market, or it will only be your call centers, you know. So what should be in the future that you consider the current country will be in your future project? So maybe Ibrahim, you can go first. First time of your first market or testing market or main market, you know, or just, just like a production market.

For us, it could be all, all what you said is possibility, but we really believe, um, the branding of Japan is so strong in terms of quality that we can really use that on our advantage. And production market for sure, because we're producing here. It's going to be very Japanese intelligence, almost like you could say Japanese AI, because it's going to be— even though using data from all around the world, it's going to be locally produced data, locally produced AI. So production part for sure. When it comes to customers though, It's gonna really depend on a couple of things. First of all, what we are developing as a product, as we are trying to reach level 4 auto navigation in the world, which is not reached yet. The most that was able to reach right now is level 3 for mobile robots. So when you reach something that doesn't exist, it's a disruptive innovation, as you can imagine. And Generally, disruptive innovations are not for mainstream customers. Where I, for example, warehouse owners would not probably need our intelligence because what they have is probably already enough. But delivery companies doing delivery, like robotic deliveries, like they do deliveries and everything, even current existing ones, they need this technology. They need a level 4, level 5 auto navigation so that they can diminish the number of people who have to look over all these robots all the time. They have to monitor them all the time, first of all, and that's— that becomes a bottleneck. So for us, who's going to be the first customer is going to depend on that delivery, delivery company. And in Japan, a good, good, good thing for us is there are so many companies, big ones like SoftBank, Rakuten, Panasonic, and all the other couple of companies, they are trying to build their delivery infrastructure. They're already doing some trials and everything. So for us, Japan probably will be our production market and then expansion market. From there, as a Japanese brand, we can expand to the world with so much more momentum. That's, that's how we approach this, this, uh, this market expansion situation.

Yeah, you got your reason, great reason. Okay, so how about you, Salma?

Um, I think in terms of other market, uh, we are still going to have it in Indonesia first, but, uh, now that the last couple of months, I think we started to see users coming from other countries, uh, other than Indonesia because of the problems that are rising. So yeah, that, that's fair enough, that's understandable. But I think in order to scale, we are starting off with our neighbor, which is Malaysia and Singapore. I think that's a very close market expansion that we're looking for, because we already have also initial partners there that we can do pilots in order to scale there. I think in each of the markets that we're scaling, we're still going to have our own production team or our own core team there, because At the end of the day, we are going to develop this empathetic AI to be localized in terms of not just the language that we serve, but also the approach, the solution that we approach. We don't want to just build an AI mental health that's just able to help them with mental health just as ChatGPT do, but we also want to make sure that the approach that our AI do is very localized as if they're talking to a local psychologist there. So we want to make sure that the approach is really tailored and very culturally fit there. So we want to also make sure that we have an internal local team to develop our AI. So we have our own community, we have our own people there to eventually develop the AI. So we're looking at that market as well as we've been seeing Korea and Japan as well, because the mental health stigma is very high. It's a big possibility for us as well, but we're still kind of looking into it as well. So that's, I think, how I answer that.

I think I can connect you with a couple of startups here who are trying to do similar things.

Happy to do that. Yes, let's connect up. Sure, sure.

And not only Ibrahim, the Fado Institute can help you as well.

Definitely.

Right, right. I think Arjun would be happy to introduce you to some of them.

Okay, so, uh, now, uh, you're talking about scaling, right? Uh, so I would like to know, how do you define scaling in your startup context? You know, more users, more countries, more use cases, or like a deeper market penetration? So how about you, Ibrahim?

Hmm, this is a bit tricky because our answers to this question change through our journey. And it is, in a way, still keeping the changing method. It's still, you know, in time where we are finding ourselves in a different approach to scale— scaling. And The— as a deep, deep tech company, as a deep tech startup, initially you gotta have the technology, right? Without technology, it's not a service, it's not SaaS, it's not— it's not something that I can just do, uh, you know, what they call concierge MVP, where you do everything manually on the page.

It's—

it probably wouldn't work in that way as well. Like, I cannot probably do this, oh, I'm selling this foundation model for robots, but on the background, people actually drive robots. That's not scalable anyway. For initial customers, it could work, but it doesn't solve the problem of building that foundation model for your product. So that for, for robot. So initial thing is going to require having that foundation model, then trying it with a couple of customers or partners. Then scaling from that point on would mean, since we want to become— maybe some of the audience might not know, but still I'm going to use this example. In the engineering world right now, one of the most popular products is Cursor because it helps you with coding so much. So what we see Kanaria Tech as potential in the future is cursor of the robotic world. Where all these robotic companies where they have robotic hardware, their, their own robotic fleet, for different reasons— it could be construction robot, it could be last mile delivery robot, it could be hospital robot— all these different companies, probably some of them, not maybe all, yeah, let's, let's correct it, not all, probably some of them are going to really need that level 4, level 5 intelligence. Those customers, those are the customers for us. So scaling up means actually having a ready-to-plug-and-play intelligence where you can just easily package it to another customer and sell it. And then that would mean selling it to licensing it to more delivery companies, for say, or hospital robotic companies. So these hardware companies are going to be our customers. So scaling up means from Japan all around the world, finding those customers, convincing them, and then licensing the model out. And there, there's always this question that comes to us: how about your own robot? Why you don't you do your own hardware and then sell your robot? That's always an option, by the way. That's always a way to approach this, and that could be a little nice horizontal expansion. But currently we are just considering us on the AI side, the software side, that provider of the software and AI. So our scale would mean providing our model to more hardware companies.

So your model is a B2B model, right? Based on the licensing.

Yeah, yeah, B2B model where— but of course this is very early to say certain details, but current approach we have that we are exploring is B2B model where there's a license licensing fee per robot. So you can use— it's almost like this per user we pay for ChatGPT per user, we pay for Cursor. In this version, per robot, you pay that service. And that's the current trajectory that we are exploring with potential customers.

Thank you. So I think Salma's model will be much different, right? So how about you, Salma?

Yeah, I think for me personally, scaling up Yes, it means more users, but for us, it's not just getting more users because eventually our AI or our model doesn't actually do much if people don't come back the next time they needed help, or even that we don't want it to just be there for them to— we don't want to just be there for them whenever they need help. We want to be with them along the way to actually help them increase their well-being and also get them to know themselves a little better. So in terms of numbers, yes, we do want more numbers of users, but also we want to deepen the quality of the conversation with Neesha, our AI, as well. So we want to make sure that they don't just use the AI for the sake of that they Google things up or they just needed like one-time thing help. But we want to make them like engage with the AI and able to use it again. And we want to have it like a one app and then people just come in and we actually customize their experience based on their country. So we want to make sure that wherever that they use this app from, they can still get to help tailor-made to their culture or where they are. We want to really make sure that it's not just them getting help, but we want to make sure that the quality of the AI that we're developing is really tailor-made and really culturally fit to them as a person. I think that's what we meant by scaling up.

Understand. So I just want to conclude that your business model is B2C and B2B2C, right?

Yes, correct.

Okay, so now we'd like to know, you know, about your GTM strategies, go-to-market strategies. Which one that you think that will go best with you, right, in the current countries or in Asia? Digital channels, partnership, partnerships, community building, or anything else? Okay, maybe, uh, yeah, Salma.

Yeah, I think I'll go first on this. Uh, relatively to go-to-market, I think we combine a little bit of here and there and everything. Yes, uh, definitely you have to be— for me, because we're targeting Gen Z, of course you have to be— exist on social media because people wouldn't, wouldn't figure out who you are or what you do if you're not existing on social media. So that's the main part of, of the digital things. But what's interesting, interesting with us for this digital marketing is that we actually haven't spent a single dollar in this digital marketing because we are also working with several key opinion leaders or quote-unquote influencers in order to for them to try out this app and then exchange with them creating their, creating their experience out of, in a video. So we combine a digital marketing and community base in order for us to be able to lower the, the CSP cost, the cost of acquisition. So we try to make it very low. And it works for us. So we've been collaborating with some KOLs to actually create a quote-unquote free video, and in exchange of their good experience with our AI. And then the second part is that what I learned from building something in Indonesia and also Southeast Asia is that you have to build a very good relationship with the government because you have to know what the regulation is. And if you're not in their radar for the regulations and for the like existence, you will not be, you know, you will not be, you will be— what I say, you will not know what the regulator is. And because we're building something that's new, we have to know what the regulator is to be able to build the product. And also because the adoptions in Indonesia really depends on how the government support. So I think that's actually what I need to— what we do, what we have done, and what help us as well for the go-to-market.

Yeah. How about you, Ibrahim? This may be more difficult.

So for us, this question has two parts. One is answer for Japan and one is answer for global. And the reason, as you can also imagine, the Japanese business culture is very heavily focused on personal relationships you have. So without having that personal relationship and trust bound, it's not very easy to do B2B businesses, B2B deals. So for that reason, we have been already, even before having the product, we have been already trying to build those relationships, which some of them, some of them building some of those relationships can, you know, take around 6 months to 2 years. And before you gain the trust of that customer, potential customer, it's not very straightforward path. So it's not very strategy involved per se. It's more like whatever that current situation with that person specifically, or that company that you're trying to do the deal with, is going to depend on very specific condition that you are currently dealing in that time. So that's why for us, it's, it's going to be always trying to perceive those customers and trying to build those relationships. Right now we have very, very strong relationship with a couple of big guys here, I would say. And that part is determining very easily that, oh, for Japan we need to build the relationships. Or the global though, similar to Salma, we are also— if you wanna, if you wanna go global and if you wanna, like, for us doing social media marketing would not increase the customers per se, because you cannot just, you know, easily get individuals to buy something like this. But creating that, that hype, creating that image recognition, creating that, oh, this is interesting, what are these guys are doing, let's explore, vibe is very important. You know why? Because in our case who makes the decision to what technology to use is, it generally comes from mostly technical guys, engineers or CTOs. And the reason is, even for our engineers, we ask them, which AI platform do you want subscription for, for your coding? Cursor, GPT, Claude, what is it? Choose one, tell us and we will get it for you. So Now imagine a scenario in the future where these robotic companies, they're like, they're trying to find out those new technologies to make their robots better. Then who's gonna decide? Who's gonna try? Probably engineers and CTOs and all those technical teams. So to be able to get their attention, so in the— using social media is very interesting for us as well, creating that hype and creating that, you know, image and brand recognition so that we can capture more attention. With that, we believe we can capture some of those different companies' attention a bit easier. Image— the brand recognition would also be helpful. Then expansion would come as a Japanese local company to the global should not— should be very, um, very easy for us as long as we have that superior technology, in our opinions.

Okay, so let's focus more, you know. So let's say going global is, is the main focus, right? So how do you, or have you ever thought about the benefits from foreign Right, you have big network, you have, you know, thousands of mentors, you have a lot of chapters all over the world, you know, 200 cities, 100 countries, a lot of country director like myself, and we can like create a lot of potential customers, right, from the FI alumni. So how will you, or have you ever thought about how to make benefits from that for the global expansion?

Oh well, I think I have already done it. I mean, I'm not from the foreign city of Vienna, but I already made friends with with you coming from Japan and you also from Vietnam, I think Poly Institute is not only— you're not only benefiting from where your cohort is. I think it's way beyond that. I mean, once you're in and you are graduated, I think you got a lot of— I think thousands of mentors and also connections. I think once you try to find the alumni and you mentioned that your founder from another institute and you just need like a, like a connection, referral, or anything in between, I think it's very easy for you to do that. And I think that's how I would definitely benefit from different institutes alumni as well. I think the dashboard really helps with all the connection and potential referral and also alumni. And yeah, I think that's how I'll do it.

Thank you. How about you, Ibrahim?

Yeah, first of all, even if we cannot find, let's say it's impossible to find customers through AFI alumni, let's say that doesn't work, right? Let's say that's the rule. Even then, if I could just talk to you, Gian, like if I could just say, dude, we want to enter Vietnam and we want to like, we want to do this right. We want to do this right way. Like if I try to, by the way, if I try to arrange a call with you, as a Foundry Institute alumni, it's so much easier than doing just randomly because we already talked as well before, Tiang, and it's so much easier, first of all. And then after that conversation, I could easily ask you what are the things that I should not do in Vietnamese market? Because I have no idea about Vietnam. I have no idea about the Indonesian market. I have no idea about so many markets. I only know a little bit about Japanese market altogether. I'm trying to learn as well. So it's always helpful to ask opinions from different people. And those mentors are very helpful. Those alumni, some of them, by the way, I would say as a personal feedback to myself that I don't really know a lot of alumni from all around the world from Founder Institute. The reason might be also that while we were doing fundraising, and at the same time trying to build the product, we went through that hell phase, you know, that very intense phase of no break, no holiday, full working and trying to get it done. During those times, I missed a lot of fun events I could not join. Some of them I intentionally did not join to not distract myself from what I was doing. But now I can find some of these you know, very good opportunities to talk with different, uh, alumni. And I'm very glad we could, we could talk today with Salma, and then this, this can become a very good connection for the future. And maybe next event, someone else, and then I can meet someone from Germany or United States or whatever. And even without all these customer acquisition paths, even just learning about the markets would be very valuable. But of course, there's no such rule as you cannot get customers through a founder institute. So I'm, I'm pretty sure we're going to be able to make very good— we are going to be able to get very good introductions from some people of like headquarters of founder institute in, you know, in California, you know, introducing us to one of those robotic companies or whatnot. Those are very good possibilities for us. I'm, I'm excited for that.

You know, it's just not only interaction because the, you know, I am, I used to sit in a board of one outsourcing company, big company in Vietnam, and we always try to find more partners because, you know, nobody can do everything. Yes, but customer, they need everything. That's the point. So one, founders, maybe they are not be able to be your customer, but they can be a contact point, you know, or a partner, can be surely be a partners because their customers, they need everything. And the founder, they cannot provide everything to their own customers. That's the point. So we just will need to talk to each other, discuss, you know, discover what we are doing. And, and from that, I think we can find out how to create partnership between the, the founders.

Right.

Okay, so, uh, now, uh, I would like to remind our audience that, uh, please, you know, uh, raise your questions in the Q&A.

Uh, there are already some questions.

Yeah, yeah. And after one question, we will go through all the questions from the audience, okay? So I think one last question from myself is about the team, you know, about the team. Because when we trying to expand globally, surely we will need people around the world. So have you ever tried to, to get the global people, you know, for your, for your team? Or if you haven't, what do you think, how will you do it? You know, how will you expand the team globally? Do you mean like having branches in other countries, or having cross-relevant team or having global partners, you know, having people in other countries as a member of your team. So how will you go through that challenge?

Well, I think for me, as for the team, I personally have a really thought about like an internal team for me, but what I have been doing so far is we've been trying to get some like mentors and also advisors from global that help our internal team to grow, and some of them are also from FI. And I think in order for me to get a talent from global, I think there are several sources on how. First is always, I think, I think if I'm not mistaken, the internal dashboard of FI is able for us to just share if we are actually currently open hire. And then the second is, I think, is always coming from the, the, like, the channel, the platform, the hiring platform globally. And I think the third one is always the referral from others. So I think that's the initial how to scale the team. I think, hope it answers.

Yeah, thank you. How about you, Ibrahim?

So for this, actually, I'd like to— this is going to be a bit marketing for him, but I'd like to shout out to Kyo Kobayashi. He is one of the most famous Japanese entrepreneurs in Silicon Valley. And there, I would recommend this platform to everyone here, by the way. They have a platform called Noxx, N-O-X-X. It's for hiring, it's for recruitment, and they're using all of the newest AI technology to make the hiring process so much more smoother. And we've been able to find some engineers from there. So with all these different technologies coming into the play, the hiring and recruitment and building— maybe I'm talking about hiring first, first of all, in this part of the question— where hiring and building teams in different countries can be even maybe easier than ever. And for that, we would probably— we're gonna probably use these kind of technologies more and more. But on the other hand, for example, Giang, if you would introduce me a specific person in Vietnam that you think that is a very good match for my business, Probably that's going to be already superior to any other candidate that I might find from AI. So those kind of— so you got to balance it out in a way where you cannot partner up with everyone, you cannot have teams all around the world. So you got to choose a place, you got to do it in a specific location that is going to suit your, um, the interest the best. So For that reason, I, I'd like to explore my own, um, network, professional network. At the same time, these, these cross-platforms with AI for recruitment and everything, and global expansion is something that sounds very exciting.

Yeah, yeah, thank you. And anyway, let's get back to Funnel City, right? Big network. You can— because I, I, I used to try already. You can get everything from the Funnel City network, you know, uh, co-worker, uh, co-founders, advisor, mentor, investors, business partners, and customers. Everything you can find from our network. Okay, so now we go through the Q&A box. We have one question from Kamashi. How you people have introduced AI in your startups and made it successful?

Does this mean AI tools such as GPT and other stuff?

I think so. Okay.

Well, I don't have to introduce all the engineers and all the people are always asking more stuff. Oh, can I purchase this? Or can I have subscription for Adobe, whatever? And can I have this Google Video AI creation stuff? Then I can, can I have cursor? Can I have GPT-3 Pro or whatever? Like they're always asking.

So it means that they have to ask for the budget, right?

Yeah, they have to ask for the company credit card to purchase them.

Okay. How about your team, Salma?

Oh, I think for me it's kind of different because back in the first 1 or 2 years we're building, introducing this kind of technology like AI as a psychologist, it kind of blows their mind somehow. So I think in the early years we have to expose ourselves a lot with many, what do you call it, many events locally, especially so that people at least get a sense or get to know what is the Nang AI, what it does, and then just get a bit of exposure about what we do. And then I think we we collaborate more with also still the KOLs to help them share their own experience with our technology in their own language so that the other people with similar persona can get also an understanding of what we do. So that's what have been helping so far with us.

Okay, so let's get to the next question. Philippe, he will bring us back to the world, you know, back to the real world. How did you acquire funding and find investors?

Okay, I think it's a classic question, but For us, our funding comes from two parts. It's from a typical VC one, which we are introduced by. Second, we get also one from like grants. I think it doesn't— grants as well. And the interesting part is that we also got a follow-up funding from loyal VC. So Loyal VC is a VC that's, you know, I don't know how to explain this, but Loyal VC is a VC that Founder Institute have. So only Founder Institute alumni and also in SEAD are eligible to be invested by them. So they invested in us $10,000 and they got a follow-on almost $200K. So that's where we got funding. So yeah, thanks, Founder Institute.

That's great. Okay, how about you, Ibrahim?

Yeah, well, obviously, like everything in life, fundraising is also like playing a game, and you gotta learn how to play the game well to get somewhere. And initially, even if you think you know how to play it, it's, it's, it turns out very difficult. Sometimes. It serves out very easy as well. And everyone says, oh, fundraising is very difficult, it's the most difficult thing. I mean, I wouldn't say that, because when you do it kind of right, it becomes very straightforward in a way. That's why serial entrepreneurs can get fundraising very easily. This was my first funding experience, so I would say it wasn't very straightforward from the beginning. I had to do a lot of failures. I even— there were some meetings that I was just There was this VC and then I didn't like his questions and I said, I don't like your questions, man. I don't want to get money from you. We can finish the meeting here. Like, you know, you even have those kind of experiences on the way. And sometimes you feel after talking with, I don't know, 20 different VCs and all of them are asking very weird questions about like, why are you even asking me this? This has nothing to do with my business. But I realized some of them are trying to see how are you going to react, like all these things and everything. And after some point, you lose that confidence quite a bit that you're like, are we going to be able to do this? Why no one is believing in us? That's where actually all these honorary institute mentors were very helpful. Anytime after those things happened, you could talk with them and they would give you a little boost on the confidence. But in the end, what turned out to be a very um, big revelation for me was as soon as you find the right investors for you— because not all investors are for everyone, not like— there has to be— it's like, it's like a little bit like dating. There has to be a little vibe matching, you know? And if there is a match, ah, it's easy.

It—

our fundraising was very fun after we found the match. So that, that's what I could say. So find your match.

Yeah, thank you. You know, finding an investor is the same with finding a customer, right?

Oh yeah.

So to sell your things, you need a GTM strategy. So also to find investors, you should also have a GTM, go-to-market, right, strategy. Without strategy, you cannot have. Okay, let's go to the next question. This is for Ibrahim from Gran Cuma. How did you move to another country completely new and started a company? Did you incorporate first?

I don't know. My Japanese story didn't start with Kanaria Tech at all. I came here years ago, 8, almost 8 years ago as a Japanese language student. I went to Japanese language school. Then my initial plan was that we— I like moving from US to Japan, that initial transition very in the beginning. In my head, I'm like, you know, this is probably temporary, you know, I'm gonna just learn a little Japanese and go back to America, you know, just trying a new country. But then I happened to like Japan more than I thought. It became very— and I was living in Osaka, by the way, which is particularly friendly Japanese people live there, and I made very good local friends and everything. So I loved here. And then I'm like, okay, let me go to university here as well. And then I went to university in Japan too. And then I found— I figured out that I don't want to work in a very big Japanese corporation because of the working environment doesn't feel very good. And I always did something new anyway, like even if it is a research or project. So my journey to this actually took time. It wasn't like, oh, let me just move to Japan and do this in Japan. It wasn't like that. It became a bit natural, organic, organic decision where I ended up being an entrepreneur. Like I could even say all paths for me Like, even if I did something else in Japan, I would probably become entrepreneur again. Like, all paths were merging into entrepreneurship at some point. So yeah, it wasn't just like starting with incorporation. It started from student life.

Thank you. And the next one is for Salma. In your AI backend— oh, it's a bit tech, right? Does it monitor users according to some modified input run against the DSM-5 to look for warning signs of serious possible health problems that should be referred to a real doctor?

What is DSM-5?

It's like an ICD-10, something that the doctor use for diagnosis. So yeah, yeah, it's that, that's for psychology. So yes, um, for us, what our AI only able to, to solve the low to medium daily stress. So if we, our AI actually detected that, that the problems or the message are in a high or severe cases, we are prompt you to actually go to a psychologist or book a session with one. So basically we, yes, we do have already like a a system, simplified like a system to sort or to filter if this message actually in a high risk or not. And one of them is based on DSM, but not because our tools, our AI is not for diagnosis. So it's not, we don't diagnose anything. We don't give any kind of diagnosis. So we only do like a summary of what you've been feeling and what you've been talking about. And yeah, I mean, in short, it is, it's, it's, it's in our engine, our technology, but we don't do diagnosis and all that.

Yeah. Okay, that's the great question, is from Peter, uh, we benga, and he's also have questions for you, Ibrahim.

Yeah, maybe summarize the question very fast so that we don't have to read the whole question. Okay, he's— yeah, he's trying to ask about the— the— like, our company faces two challenges and issues. One of them is the capability of the AI, of course. The other one, he's curious about if our AI is going to be able to work in different robots directly, or does it read— doesn't need a specific hardware requirement. And Or this is, this is a very great point for us as well. Internally we discuss about this, and the something that we are very excited for our vision is we want to have a platform-agnostic foundation model that can work any kind of robots. Let's say any kind of robots in this case is any kind of robots with wheels. Let's start with that. Of course you want to increase from there, then you— it becomes a bit more generalized intelligence, obviously, but Let's not get there first. That's the next step, right? So we want to have a platform-agnostic intelligence where it doesn't even require any prior knowledge about the hardware. How this could work is something that we are exploring right now. Imagine you get a canary model. You put it on your robot. Your robot has 4 wheels. It has a delivery box area. It's like 1 meter high or something. Like 50 meters, 50 centimeters high, and then you're trying to use our model. Like, what we want is, what we really want to achieve is, imagine that that model sends random electric signals to the wheels, to put it very simple, in simple terms. And then, like, let me send a little bit of electricity, a little bit of input signal, to right wheel and then the left wheel and then see how much the robot is going to move actually, how much my environment is changing actually, so that almost like a little baby, the model would explore its potential while trying, oh, I must have this kind of wheel, I must have this kind of body shape, I must have this kind of torque power and then electricity, all these things like we want it to be, we want it to be so that everything can be automatized. And it would be, you just wouldn't get the product, the AI model for your robot, and that's it. It would handle everything else. And that's the future, probably. That's how is it going to happen. And for that reason, yeah, for Peter, right, Peter's question, yeah, it will be platform and shape agnostic.

Okay, thank you. So now we go to the questions from Anh Bui. So this name is from Vietnam. He may ask that when did you raise a fund, you know, from which Which phase? From when you have a POC or when you have a prototype with some numbers or else, you know, so when did you raise a fund?

For us, even though technically we had customers, even actually to be very open here, we had SoftBank as our customer providing our AI model to them. But we still not— we still consider that that relationship with SoftBank we had was a very small part of the fundraising. So we raised as a pre-revenue startup, I would say, with no customers, no technology, that, that no working technology. Of course, we had some of our own infrastructure, So we had to convince investors really about the potential, like team, team-only raising here, which is very difficult to do in Japan, which we are very happy to be able to accomplish so far.

How about you, Salma?

Oh, for me, our first fundraising is quite difficult. I mean, I have to talk to nearly 50 thesis at that time only to get a fund from grants eventually. So it was quite difficult because given my background as a first-time founder and also AI mental health, something that's new. So it's been quite a challenge for us to actually convince that this is going to work. So yes, that's our first one. It's actually pre-revenue. I think we only have a proof of concept. To some of the small numbers of users here in Indonesia. And then I think after we get that grants, we get a few more numbers paying customers, we got our next funding. So yeah, that's how it goes.

Okay, so I think the last question for this session is from— there are two questions. Okay, it's from Giovanni. Maybe you need to explain again, right? Uh, Giovanni Bazzi, he would like to know how did you make your first customer and which kind of strategies did you use?

Um, for, for making our first customers, when we try to make customers through proper ways, you know, trying to do cold calling, cold emailing and everything it wasn't very successful. And the way that we were able to, you know, land big customer was I, I, I, I saw a robot and I was like, I was talking to my co-founder and I said, can you check who's the responsible for this robot? Let's call them and ask for a demo. And then they arranged a demo for us and then they were showing the robot. What I did was I, I wasn't even, I didn't have any intention actually. I was just seeing this robot and I, I was like, this robot is stupid. Like I was just telling them honestly, why you are doing this way? And they were selling out because of this. And I said, that's wrong. You should do it this way. And then really, and then they were doing more stuff. And I was like, why you are doing this way? And they said, because of that. And then I explained again, that's wrong as well. You should do it that way. So that's how they got interested. Can you actually, would you like to meet us again? And then we met and then during the meeting, there's another guy that actually, no, there's another manager came. A woman from SoftBank. And then I told them, yeah, this is actually how it could be done. And then when being very honest about those opinions, they were very interested, and that's how we landed it. And it was quite the contrary of how I read on the internet, how I heard from the people. So just being brutally honest about what's going on within the tech side worked that time. And I am probably going to keep this method a little bit more.

Yeah, I think it works also for me. My first customers are actually my own, my own university where I work at. So I was a university, I was a psychologist in this university. And I think I know what like in their counseling system. And I just, just give a shot to help them with that. And it works. I mean, what I learned about being the first customer is many times is it coming from your own people, your own friends, your own family, something, someone that knows you. So, so, so that they don't have to wait until your product is fully ready, fully deployed, like complete. Perfect in order to use that. No, they're willing to, you know, grow with you, iterate with you with the product. And I think we are pretty grateful to have such first customers as well.

Thank you. So I think we should sum up with your advice, you know, final advice. For the founders who joined today, how they— what do they should think about, you know, when they started to scale, you know, and then what they should think about when they started to expand globally, originally. So scale and global expansion. You go first, Ibrahims.

I think I'll go first.

Go ahead.

I think what I can think of clearly is that start with the people close to you. I mean, when it comes to scaling to a whole new place, a whole new country, What will help you is the one person, having one person that really knows the market, really knows the place. So before going there, you have to have at least one person that you can talk to, uh, for that market, about that market, about the users, about the problems, so that you don't waste your time, uh, doing experiment, doing, um, try to do this or that, that doesn't work. So So I think having like a very, a person that knows that market really well and then have relationship with them and then at least, or if you can have them on your team for you to scale or just being an advisor or something like that, I think it helps. And other than that, I think you can also start with having like a pilot with a certain group of people in that country just to test the water if the solution is actually needed there before scaling big. Start small with just some, some of the people there just to see how it goes and the feedback as well. I think that's, that's all for me. Thank you.

Thank you very much. Okay, Ibrahim.

What I want to leave the last message here is When we were in Founder Institute cohort in Japan in 2023, initially I realized there are so many things that I have to learn. I'm always reading articles, trying to watch videos and trying to learn. And then of course you have these mentor office hours, right? With your mentors and then you talk and you ask questions. And initially each mentor, I was asking different questions regarding like focusing on their field, right? Then after some point, some of my questions were already answered and I realized I kind of like some of my questions overlap for different mentors. Then what happened, which was very big, you know, realization for me, it was when I asked the same question to different mentors, they were answering differently. One was saying A, the other one saying B, and then the other one said— then I was like, on purpose, I asked the same exact questions to every single mentor I had the meeting with. And all of them had different answers. Like, I even had this actually. I had one answer coming from one mentor. I tried to challenge that mentor, and that, that mentor didn't, of course, but you know, they are strong. And then I went to another mentor, and I, I asked the question, they answered, and I used the first mentor's answer for the second mentor challenge, and they were like, they didn't know how to answer. I— okay, but I realized no one knows the answer. There are not only one answer, there are probably so many different ways you can do it. Do it. So one thing, no one knows your business and your idea and what you want to do and what you can do more better than you. You're the only person in the world who knows the best about your capability, your desires, your excitement, everything. That doesn't mean don't listen to your mentors. Always listen to them, always ask questions and everything.

I do.

But don't take their, uh, messages and answers as the as the verse of a holy book. You know, it's not like that. It's just a nudge for you to think a different perspective, different part of it. And in the end, no matter what, go with your own gut, like your own decision, because then it's going to be your decision. And even if you're wrong, so what, man, you learned. But if you follow someone else's advice and it didn't work, that will be painful. That will be painful because you didn't listen to your gut, right? Listen to your gut. Do whatever you think is the best for your own sake and your company or whatever you want to do. And I think that's going to be the easiest way of peace of mind for you. Always follow that instinct that you have, and it's going to be fine. Even if you fail, you're going to feel like this was a good failure. I'm glad I failed. I, because I failed on a couple of things and I'm glad I did. Like, because I decided on those and I failed, I'm very proud of them. So follow your own decisions and instincts. That's, that's the last thing I want to leave here to the audience, I guess.

Yeah, thank you. So, you know, your product is from your own, but the market is not on your own, right? The market is for a lot of people, right? So I think two things that you need to do right away, right after the session, is try to find Ibrahim and Salma on LinkedIn, try to connect to them because you know them well now, right? After listening to all of their stories. Okay, thank you, Ibrahim. Thank you, Salma. So, Julie, we try to have coffee together, right? In Japan, Indonesia, or in Vietnam.

Of course, of course, of course.

Or maybe in the— how can I say— in the yearly annual event of Finance Institute University. Okay, so audience, this is the last minute, you know, but You still can stay and join the networking session. So I hope Ibrahim and Salma will join as well, right, in the networking session. So it's all for the panel session, but please go to the networking session. Thank you very much for joining us today. Thank you, speakers. Yeah, and see you very soon.

Thank you.

Thank you, organizers. Bye-bye. See you soon.