How satisfied are you with what you achieved this weekend? Zero five. Five. I like it. Great job. Welcome to Sea Hack. We're here in Hong Kong with over 250 developers and designers from over a dozen countries coming together to build projects over the next 24 hours. We will evaluate these projects. We're going to give them prizes and learn a lot about how people create cool projects in a short amount of time. So, this is an app that teaches people Cantonese pronunciation. Yes, you can just record your voice and it can determine if you pronounced the "u" correctly and it will correct you if you have the wrong pronunciation. Okay, so you're Jenny, and you're Coco. How do you know each other? We just met yesterday. Yes. And we created a team . Yes. Are you both from Hong Kong? Yes, we are from Hong Kong. Perfectly. And tell me a little more about your app and maybe give some more examples of how it works. I want to try it and see if I have good pronunciation. Oh, it's just a demo version. So, there are some errors, but you can just try. You can start recording. Let's see how accurate you are. Oh, congratulations. You got 100%. OK. Well, now I want to eat some Ganon Zai. You have made me hungry. We can go buy some. Perfectly. Yes, bring in the Hanonai and bribe the judges. Yes, I'll ask them to try. Yes, we also have other sections, but we are still building, for example, there is a visualization of the tone and six tones of Cantonese and some texts, but we are still building. Yes, I think the lyrics will be very popular because many people who don't speak Cantonese like to listen to Cantonese pop music. Yes. Have you listened to any canon pop songs? Well, I won't name any right away, but, uh, yeah, I listen to it a lot. Good. Yes. So I would also recommend a good canon pop to those who want to study the candidates and listen to music at the same time . Good. Yes. I think it's a really cool feature if you can essentially have them sing along and then also rate their tones and pronunciation. Oh, of course. We can add this feature. That's a good idea. Thank you. Yes. Perfectly. Continue in the same spirit. Tell me your name and where you are from. My name is Arena. I am from Physics at HQU. I specialize in machine learning. My name is Brian. I am from Vietnam. I currently work in Hong Kong at a technology company. I'm Ben. I also work at a technology company here in Hong Kong with Brian here. Hi, I'm Josh from the Philippines. I am currently a student of Economics and Biotechnology at HKUST. I am Carmen. I'm from Hong Kong , and I specialize in VIP coding of 3D low-poly 3D experiences and work in the field of artificial intelligence transformation. I'm really interested in learning a little more about what you all create. Brian , can you just tell me what problem you're solving and for whom? Okay, thank you. Uh, we're here to create a hackathon for a company called Apple Park. So, in essence, it is an individual recommendation system that will help the customer better choose dolls and toys for their children. I think the problem we're facing is with a lot of people buying gifts for other people's children. They want to show that they know the child and sympathize with the parents in their developmental mission. So this recommendation system that we've built is more conversational, so that we can see what characteristics, you know, the giver wants to express in a gift, and give them recommendations that match their preferences. We are here with Fzy Sherry Ricky, they are creating Tinder for loans. So we are, in essence, alone. We have this Tinder for loans. So we have an AI-generated plan with a given story and the amount of money we need. So when we click Send, the video is displayed where we can send it, like a video for young people. All applicants can submit it. So how it works is, lenders can watch the borrower's video and then see any video they like. And then if you click on it, you'll see the information. And if you swipe right, " accepted" will appear. And then if you swipe down and then swipe left, it's "yes/ declined." This is what is essentially happening. Perfectly. So they can just watch your video, and if the moment comes, they'll decide , "I don't want to finance this. I'm not going to give them a loan." They will just swipe it away and move on to the next one. This is how it essentially works. The idea is that it helps people who are applying for loans because it simply reduces the friction for them to present their requests to potential lenders. Yes, it also helps lenders get more information about borrowers and see how much they want. We can also show their other stories using blockchain, we created audit logs that they can review and they can see their data and whether they want to lend to them or not. We're here with Jeffrey Joe from Baltimore, who flew in to help organize and host the Mentor Mates Hackathon C Hack. How are you ? Everything is fine. Quite hectic, but , uh, really good. We have some really cool projects. Looking forward to seeing what everyone is creating, uh, what everyone has completed today. Perfectly. How many applications have you received already? Did you know? Uh, we have over 20. I think we're getting close to 30 applications. So 30 teams applied, which is great. This is great. Maybe you can tell me a little bit about what the evaluation process will be like here. Yes. So basically, when we evaluate hackathons, depending on the direction, we look for a few different things. For example, for startups, we look for whether your idea is scalable for venture development. When we look at, uh, an industry, we look at whether you're solving a real problem, and whether you're solving it correctly. Regarding community, we're looking at some value that you add to a nonprofit or potential, uh, community effort. So our judges will check these criteria. Uh, for the first round, it will be a two-minute video presentation. All the judges will review it and then we will choose the top three from each track to move on. Uh, the people who make it through will get to present their project live to the entire audience as well as the judges, and we'll pick one winner from each track. Perfectly. So, there will be three winners tonight. Is there a grand prize or something like that, or is it just, uh, the prize for each track the same? Uh, we have some prizes from sponsors, from our wonderful sponsors, that will be in addition to these. Good. Sounds really cool. And it's an honor for me to be one of those judges, to be able to, uh, determine fate. No, that would be great. I'm looking forward to some of these videos. Let's review some of them. I'll run some of them now for you good people. Yes. Thanks again, Quincy, for the rating . Ugh, I'm really looking forward to seeing what everyone has built. This is a 24-inch pizza right here in Hong Kong. You don't usually see pizza like this outside of New York City, but now we're going to delve into one of them. From Sanji S. Hak. Angela, can you tell us a little bit about what you build and for whom? Among the three tracks, we chose, uh, the Apple set, the Apple Park sets. And we determined that, uh, with all the dolls in our products, parents don't know which product to choose for their children. So what is our solution? We, uh, customize stories for each individual product so that parents know, uh, which particular doll is more appropriate for children. And how are we going to do that? We'll have LLMs, similar to Tragedy P, a chat box where parents can enter information about their children and we'll create a story and also have matching dolls, uh , that are appropriate for that child. We're here with Connor from South Africa and Cindy from Hong Kong, and they're building a tool to help recognize volunteerism in a charity . Right. Yes, that's true. Maybe you can tell us a little bit about the problem and how you solve it. Yes. So , basically, we noticed that a lot of business angels have their own system, and basically, the reason we came up with this idea: my friend works at our marketing agency, and we noticed that a lot of NGLs pay huge amounts of money to marketing agencies to bring in volunteers and sponsors to participate in their event. So we're creating a platform that can bring everything together, minimize the cost, so that all people can get the money they deserve, and not just spend it on marketing. This is what we create. Perfectly. And what has been the most exciting part of creating this so far here at Cack? Uh, taking an idea from the concept stage 6 hours ago to a finished live demonstration. It was a lot of fun. I mean, none of us are technical experts. She is from marketing. I'm from finance. Never touched code before. Never even downloaded Visual Code Studio or anything like that. But with artificial intelligence, anything is possible. So it was really fun to just turn an idea into an MVP in essentially 6 hours. Keep it up. Keep up the great work. I'll come back and check. Uh, you're going to keep refining and tweaking this in preparation for judging, right? This is the goal. Hi, I'm Quincy, the founder of Free Code Camp. Who are you? Hello, my name is China. Nice to meet you. I'm a hacker, a rapper, who develops AI, because that's what I do. Quincy from Freo Camp is coming to Hong Kong to leave his mark, as we, along with our mentor friends, create better hackathons for all of us. We are here, SEA Hacks , with Chinat, the hackathon organizer, founder of mentorm Mates, who has already held three hackathons. This is the second in Hong Kong; the first one in Hong Kong that I had the pleasure of attending, 50 participants, this year 250. How do you achieve 500% growth in one year between two hackathons? I think if you run great events, you naturally attract people to come together and help you grow that event even more. What was really interesting was that between last year and this year, so many people from all over Hong Kong outside of Hong Kong reached out to me asking, “ Hey, China, can I do this hackathon with you?” I've seen the amazing work you're doing here with Quincy, and also with some of the people from Hong Kong. What can I do to help and support?” And so this year we're really excited, as you know, we have a lot of amazing community partners, including Angel White and a lot of other local people who have really, you know, put in the effort to get people together to do this hackathon. You were born and raised here in Hong Kong, and you are a deep believer in Hong Kong's tech ecosystem. What are the reasons why people who have been sleeping about Hong Kong should come here and explore potential opportunities, such as starting a startup here or potentially investing in startups here? So I think there's a big discussion going on right now about where should I build my AI? Where is the future of AI? In China or the USA? And the answer is that the future of AI depends on both China and the United States. And there's no better place to discover it than here in Hong Kong, because we're such a global city. We are so close to Shenzhen and the Greater San Francisco Bay Area in mainland China, but at the same time, we have a global international outlook and can access AI models, AI talent, and AI infrastructure from abroad. And so Hong Kong, being a kind of natural bridge connecting two places, is a natural place where you can find east and west. We're here with Michael from Hong Kong Local, who's building a CRM tool, like a customer relationship management tool, for local small businesses that doesn't require them to have a giant marketing department or anything like that. Tell us a little about what you create and how you create it. I believe that our conversations should not be training data for the next artificial intelligence and we always want our private conversations to be completely private and never sent in large text, which is why I created REFT. It is a local, AI-driven customer resource management tool that connects directly to your WhatsApp or Telegram, tools commonly used by small and medium business owners in Hong Kong, and it will automatically reveal your chats using fully local AI, which is completely processed locally and never sent to the cloud. And it will be able to, for example, offer your answers to your clients, as well as have a summary of your relationship with them, as well as extract memories from them . So, its goal is to improve your customer interactions 100 times, and most importantly, it is completely built locally . It can run on any of the modern, ahem, recently manufactured computers, and it will run efficiently without consuming too many resources, using, ahem, the gem of E4B local models. This means that every thing will be stored entirely locally and never processed by the cloud, and this will ensure the security of our conversation, and we believe that the issue of privacy, ahem, is increasingly concerning in this modern era of artificial intelligence. What are you creating? If I understand correctly, it will use optical character recognition (OCR) on the invoice, and then once it understands the invoice, it can apply the tax. Who is the end user of this? Who will buy this app? Any companies that need to manually enter data and would like to have a single workflow for junior employees to work in a single user interface for building inspections and digitization. Will it be like a property management office, like a landlord, or someone who has junior staff who review all sorts of documents, or are you focusing specifically on billing? The flow is focused on invoices only because, uh, we have a workflow that's like a trash can, which allows employees to leave all invoices in this workspace , and then they can view them in an organized way. This actually minimizes manual data entry errors and makes the entire process more efficient and organized. Okay, Terry. Uh, tell me a little bit about what you create, what problem you're trying to solve, and how you solve it. Of course. The problem the company faces is that some individual users take a long time to write a loan application, to compile a checklist for submitting an application. And now we are creating artificial intelligence that helps in issuing loans. We can complete everything with simple chats and an AI assistant will evaluate everything that an individual user has provided and provide feedback to the admin and the admin will be able to check everything on that website and he will not need to interact with the users. Homemade cookies made by a Chinese mother. How satisfied are you with what you achieved this weekend? Um, very satisfied. Yes, we are very happy with what we have achieved, and we believe that our project can, um, change the industry. Yes, it's about insurance. Perfectly. I like your confidence. Keep it up. I would say I'm quite satisfied. We've completely, um, implemented a local layer of AI, which means that now you can have a local AI generating all your social media responses and retrieving your memories, and most importantly, it never leaves your computer. It's as secure as your messages, meaning your WhatsApp messages and, um, anything you've seen on your computer never leaks to the cloud. Perfectly. Congratulations. This is nice. Obviously, we did something like a project. It was chaotic at first , but eventually we finished. We've gradually solved a lot of problems, and, um, we're very happy with that. We just submitted our project. This is a startup track. How does it feel to submit your project? This is very exciting. I mean, you can see my hands. Good. And this is our demonstration. Here is our demo, we create a short video to show how to use our app. Yes. Very cool. Are you sending? Are you finished? So, we have a problem with our application. Unfortunately, we were unable to publish our Versel online. So it is published on localhost. I just couldn't upload it to Versel. So now we are uploading the video from the local host to YouTube to later add it to our application, but it may be too late. I'll let you focus on that. Uh, good luck. State your name and where you are from. Egor, Russia. Alina, Russia. How do you feel when you submit a project you've been working on for the last 24 hours? It's a bit exhausting, to be honest . Yes, we barely slept, so yes, it was a little difficult. What about you? I'm fine. We are deploying the project in real time on our mobile so you guys can access it. Good. So, they are now being downloaded, deployed, and sent. Sorry, my teammates are busy, because we're currently trying to deploy a prototype we have on Versell, and it's 3:30, right on deadline. So we are in a big hurry. Yes. We talk about our stress levels at the very end of the hackathon. You don't want to send your file. Um, right? Oh, here are the environment variables. Yes, because you need it to work. Ah, yes, that's true. This is true. Good. Um, you tried to click "Import". Let's see. Maybe he does n't even have an environment. But without the file, it probably won't work. No, let's see what happens. Congratulations. Oh, wait. It works. It works. Oh, I like that. Wait, wait. Let's...Let's test this first. Good. Can we send this? Yes. Let's send this first. Let's add a link. Let's show the link. Good. Where? Where did you get the link from? Yes. Just make sure everything works before sending. I think they'll probably give you an extra minute or two. Good. No, everything is fine. We can fix this on the server side. As long as the link is stable. How do I do this? Where is our presentation sharing for slideshows? Okay, let me come out of incognito mode and paste this into our slide. Then we need to share this presentation and the submitted material. Okay, here it is. Good. So, now we need to share the link to that name. That's good. Share it with me and then I'll...I'll pass it on to the organic...I'll pass it on to you. They are sending the material right here now. Good. We can...we can leave them alone. We can come back later. We can't choose the Corax track. So, we need to start putting together an application for...Oh , no. These two gentlemen here , they are having trouble choosing a track to submit. Okay, we're monitoring quality control in real time. How satisfied are you with what you achieved this weekend? Zero five. I like it. Great job. Very satisfied. I learned to program. I'm late. I'm still learning, and now I can create something. I created an overlay map for all natural resources and fertile lands. Yes, that's great. And you did this using something like a code harner, like cloud code. We did this using cloud code and from the token market that the organizer provided us with. Congratulations. Thank you. Well, the final project wasn't actually completed, but I learned a lot and, uh, there were a lot of experts there, so I'm really happy with the event. Perfectly. Congratulations. Keep working. You worked on that Yumao thing, uh, right ? Did you finish it? Not yet. Not yet. Not yet. Good. Will you continue working on this outside of the hackathon? Yes, of course. Yes, I will continue to do this. Actually, uh , I learned a lot from others today. I think this is a pretty good opportunity for me and I will continue to do it. Yes. Perfectly. I'll tell you, I've participated in hackathons where I couldn't finish the project on time and had nothing to submit. Everything is fine. Keep working on this. Yes. Thank you. Continue. Yes. How satisfied are you with what you achieved this weekend? Incredibly satisfied because we made the impossible possible. I think we're ready. We managed to do a very good job with our project, literally in the last few seconds before the submission deadline. And I'm very proud of the entire team and what we were able to create. So yeah , let's just see what's in store for everyone. Perfectly. Congratulations . Thank you. Thank you. Thank you. They got it. He was sent. Perfectly. Congratulations. And now we are ready to talk about the evaluation process. Let's check this out. I think the UI itself was obviously filled with plots and stuff, but it was all good. Yes. The likelihood of implementation, I don't think there's anything stopping this from becoming a real product. Yes. Profit, maybe four. Good. Yes. I mean, I don't know. It seems like yes. This is an imaginary problem that I cannot empathize with. I have no sympathy for this. The idea for this project arose from an educational and financial perspective. I want to be able to use this tool to invest in research infrastructure. I talked to this guy, considering the fact that they didn't even know how to use cloud resources and such. I think the execution is probably rated a four. Yes, the market is also rated at four. This is, um, interesting. Yeah, I mean, if you want to learn, like, how to trade commodities or something , it can be a really useful tool. I don't see you going back and using it multiple times. Yes. For example, how much you actually need to know, i.e. after you use it a few times, you will probably understand. So I think the main problem is scalability. I'm not sure about the market in the long term. The presentation is quite good. But she is so imperfect. Claude does a very poor job of dealing with excessive amounts of explanatory text. For example, if you look here, it's like text, and then text, and then text again. Like on a real squeeze page, then it would only be one header element. It wouldn't be like three things. Yes. It's just little things like that that drive me crazy . Heat. Heat. N. We have the first final team to represent. Everyone applauds loudly W. Today I want to introduce to you, Red, your completely private, completely local second bra for your social media, powered by artificial intelligence. What it does is you connect your WhatsApp or Telegram to it, and then it will automatically send your replies to your friends or clients. So, this is the future of social media. If you, if you are anyone, one of those people who needs to communicate, this is for you. How do you charge? Right now, um, it's a signed open source project, and to be honest , um, it's connected to the roset web protocol, which is actually our cloud service. However, our commercial plan will be, um, integration with proper records using, for example, Matter Business or, like, WhatsApp Business. And this act of, like, uh, inter, this kind of, uh, integration actually requires a real business, you know, a real solutions provider , and we will be that. So , we're essentially playing the damage agent game, right? We sell open source solutions and then we provide the integrations that the car needs. We have Candy Candies 101, how people love candy. I don't know how many of you know candy or not. Can you raise your hand if you don't know candy? Oh, many of us. So you can definitely check it out because we have a different culture. We can translate this into English so you can understand what's going on. There are five tones in can, and we can simplify that to six tones, and we can listen to the tonality. Let me recreate the key. And if you forget which song is which word, you can just, uh, you know, listen and find the right correct word of the song. And then, the most important feature of our app is that we can pronounce, we can try to pronounce the word, and we can choose a different script; For example, if you want to order something at a restaurant, you can say, "Oh, maybe I want two pieces," and you can just say, and we'll see how it goes. Is this ideal? OK. This is not useful. This is not ideal. You need to correct your homework. So you can write down, as you experiment, all the different pronunciations and translations of the word " allies" and how you do it, so you can evaluate the pronunciation, and you can see whether you're pronouncing it correctly or not; and below is what you perform and above is the unit and if you make a mistake until you say, "Oh, you don't have a G or an S" or you have a mistake and yeah it's just about becoming awesome. Thanks, okay, so we talked to lenders like Kash and they told us they were losing millions in revenue because they weren't calling their customers in a timely manner and their customers were leaving them for other lenders because they wanted their money really fast. So, we created a voice agent that is based on local rules so that they can connect and call their customers the moment they leave their name and phone number. Here's how it works. This is a multilingual regulatory voice agent. If I understand correctly, just to summarize what you said, the goal is to respond quickly to people who know so they don't switch to another provider . But by doing this, you are completely delegating decision-making to the agent, right? For example, how binding is the decision that this agent will make in advance. The agent is the interface between the credit solution and the client. So the agent doesn't make the decision. It essentially ensures that information is collected as quickly as possible and transmitted to the decision-making mechanism that already exists. Why does it matter what gift we give to someone else's child? Think about it. That a child would be perfectly happy playing with their mother's hairpin, for all we know. The thing is, gift-giving is not about buying a gift in the restroom because they wanted to buy a gift for someone else's child—it's a sign. This is the message I know about your child. I care about her and am invested in her development as a person. And if you look at the site called Apple Park , it's very confusing. This is a very generic Shopify app on a page where you select products that you think she might like. There is no personalized approach here, and that is what we aim to convey through voice. So let me ask you a question about your refrigerator, like in this app, and the user journey is very simple, just scan Chinese medicine and artificial intelligence will organize the data for you . So ask, how well is the interaction identified? Actually, we have this table. So you may know that this is called a red base, and it is also a base . So, I guess that's how you answer the question. Did you see this? I asked a question and he just demonstrated the product, answering my question well. Hey, I'm here with John V. He teaches at Stanford, and he's had about three exits as a cybersecurity founder, and now he runs Inception Studio. So hackathons play a really big role in the whole Inception Studio experience, because you have these intense 72-hour retreats, I think you mentioned them before. Yes. Uh, I mean, there's no better way to, um, find and evaluate co-founders, even to verify your own ideas, right? It's like when you have limitations, limitations breed creativity, right? When you have a hard deadline, the deadline is very clarifying because everything non-essential falls away and it forces you to come up with a creative solution, and that's where real value is created . Great, Joe and Willie, thank you so much for your time. Enjoy the evaluation process. We will now evaluate some projects. Good. So, we chose one for each category. I looked at all the projects and started to essentially rank them within the category. So overall it doesn't matter. We're not choosing an overall winner, are we? Yes. We just choose one for each track. One for each track. Good. Great demos, everyone. They were wonderful. Also, videos, we enjoyed watching all the videos that the different teams sent in. They were really creative and very well done. Good. So, we have chosen winners in each of the three categories. Karina is one team. And finally, we have another community track. There are people who were able to present their work today, and there are also several honorable mentions. We would like to share a few more honorable mentions . Yes, I was extremely impressed with the English language learning presentation. Yes , great work. We have an early save for the winner of the final community track . If there 's one thing you guys can take away with you, it's hope. Faith and startup. That's hard enough, but doing it outside the Valley, outside the Bay Area, here in Hong Kong, is something especially challenging. And yet, I hope that what we show you today will give you hope. That even in Hong Kong you can build something too. Hope in the face of uncertainty. Hope in the face of challenges. the courage to hope, to dream, to believe. Thank you to everyone who came and I hope to see you all at our next group. Thank you. I hope you enjoyed the CA hack. I learned a lot. I met a lot of ambitious people who created a lot of cool things, and this is just the beginning. There will be many more hackathons in Hong Kong. I hope there will be many more hackathons in your city as well . So you can participate. If you feel afraid, don't feel intimidated. Just go out and build.
Generated algorithmically for Search Engine
Indexing.