Ep 512: Inception Games Final – Who’s the Top NVIDIA Inception Startup?

Topics Covered in This Episode:

1. Inception Games Championship Round Overview
2. NVIDIA Inception Program Introduction
3. DMKTZ's AI Apparel Design Explanation
4. Glia Cloud's AI Video Automation Approach
5. Finalists' Sponsorship Prize Details
6. DeepChex AI Model Evaluation Summary
7. Expander AI Custom Agent Platform Features
8. Beamer's Video Compression Innovation

Keywords:

Inception games, NVIDIA Inception Startup, GTC conference, AI startups, Business leaders, Chinese AI startup Manus, $75,000,000 funding, Butterfly Effect, Autonomous agent, Large language model, US president Trump, Executive order, AI literacy, White House task force, Artificial intelligence education, Free daily newsletter, OpenAI, Lightweight version, ChatGPT, Deep research tool, Usage limits, Paid subscribers, NVIDIA Inception program, Developer forums, Venture capital firms, Glia Cloud, DMKTZ, Personalization platform, AI-powered videos, Script writing, Rendering, High impact videos, Audience engagement, Multilingual support, Seamless customer experience, Apparel design, Demand-based design, Precision-driven products.

Podcast Transcript

Jordan Wilson [00:00:18]:
Eight AI startups entered. There's two left, and only one will be victorious. Alright. We are finally here with our championship round of the inception gains. So in partnership with NVIDIA at their GTC conference, we brought you an awesome eight group of AI startups. We heard from you all on who your favorite was on how they're using AI to help everyday business leaders grow their companies and careers. And today, we are here with the final round. Alright.

Jordan Wilson [00:00:58]:
I'm excited for this, and welcome to Everyday AI. What's going on y'all? My name is Jordan Wilson. I'm the host of Everyday AI, and this is your daily livestream podcast and free daily newsletter helping us all not just learn AI, but how we can leverage it to get ahead, grow our companies, and our careers. If that sounds like what you're trying to do, you're in the right place. It starts here with the livestream of the podcast where you can learn the latest. But to leverage it, you need to go to our website, youreverydayai.com. There, you can not only listen to more than 500 episodes from some of the world's leading experts in AI across all categories, but you should also, while you're there, sign up for our free daily newsletter. We're gonna be recapping today's show, with some of the most important highlights as well as keeping you up to date and ahead, with everything that's happening in the world of AI.

Jordan Wilson [00:01:50]:
So, speaking of what's happening in the world of AI, let's quickly, before we get started, go over what is happening in the world of AI news. So first, Chinese AI startup Manus has raised $75,000,000, in new funding led by US venture firm Benchmark, boosting its valuation to nearly $500,000,000. So this new funding round saw participation from several existing investors and marks a significant increase in the company's market value, roughly quintupled from previous levels. So the capital will be used to develop and expand Manus' AI technology, which focuses on artificial intelligence agents designed to automate everyday tasks. So, the parent company, called Butterfly Effect plans to broaden its service reach beyond China to key international markets, including The United States, Japan, and The Middle East. So, yeah, if you haven't, seen or heard of Manus, it is, essentially very much like OpenAI's operator. It is an autonomous agent. You can give it a task, and it'll kinda go out and, you know, figure things out on its own.

Jordan Wilson [00:03:06]:
So right now, it's, you know, uses I I think last I saw it has, like, 26 different, tools, that it can use under its belt, and it was running on, Claude's, large language model. Alright. Next piece of AI news, US president Trump has signed a sweeping executive order, to make AI literacy and workforce skills a national priority, establishing a White House task force on artificial intelligence education. So this new order launches that presidential AI challenge, a public private partnership to develop k 12 AI resources and new funding priorities aimed at expanding AI coursework and certification for students and teachers nationwide. So federal agencies, including the Department of Labor, Education, and Agriculture, are directed to prioritize AI related teacher training, youth apprenticeship programs, and funding for AI skill development across all education levels. So this move is set to impact how Americans learn, teach, and work with AI, potentially reshaping classroom learning and career paths for students, educators, and workers in The US. Alright. Our last piece of AI news today, OpenAI has launched a new lightweight version of its very popular deep research tool inside of ChatGPT and has expanded usage limits.

Jordan Wilson [00:04:38]:
So OpenAI has rolled out a new lightweight version of its ChatGPT deep research tool powered by the o four mini model versus the o3 model to provide faster, more cost effective research responses for users. So this version is now available to free ChatGPT users. Yeah. That's a big one, and also to paid subscribers after they've hit their limits. So the lightweight deep research delivers shorter responses while preserving the overall depth and quality, allowing OpenAI to increase usage limits across paid tiers without overloading their infrastructure. So paid users on ChatGPT plus teams and pro plans will benefit from higher usage limits thanks to this cheaper, and lighter weight option. So once the original deep research limits are reached on a paid plan, queries will then be automatically switched to this lightweight version. So if you have a free plan, you get five queries a month.

Jordan Wilson [00:05:46]:
If you are on, a paid a normal paid plan, so that is including, Chad GVT plus Teams and Enterprise as well as EDU. That's 25 queries a month, and pro users on that $200, a month plan get 250 of the more powerful deep research queries. Alright. We have more on those stories and a ton more in our newsletter, so make sure you go to your everydayai.com. Alright. I'm excited for today's show. What's up y'all to all the the live stream crew in the house? Love to see it. Brian tuning in from, Minnesota.

Jordan Wilson [00:06:23]:
Renee joining us from, Tampa Bay. Michelle on the YouTube machine. Graham holding it down from Ireland. Love to see it. Fred in my backyard here in Australia. Nathan, it's already nighttime there in Australia. Thanks for tuning in y'all. Yeah.

Jordan Wilson [00:06:37]:
If you didn't know, if you're ever on the podcast, listening on Spotify or Apple, first of all, thank you. But second of all, it it's technically a live stream. We do this every day 07:30AM Central Standard Time. So if you ever wanna, you know, come in and ask questions of guests that we have on, you know, what better way to learn, than to be able to speak with the people, building all of the AI systems that we all get to use and just experts in their field. So, yeah, come hang out sometime on the live stream. Alright. Let's get into our ding, ding, ding. It is the final round.

Jordan Wilson [00:07:10]:
So, real quick, let me first tell you what the NVIDIA inception program is. You know, as we tell you a little bit more about the final round, and and we're gonna hear from our final contestants in the inception games. So NVIDIA inception is a program that helps startups grow and develop their technology. So the program provides free and discounted NVIDIA tools, software, and training. And startups in the NVIDIA inception program get access to developer forums and technical resources. So startups can connect with also venture capital firms through NVIDIA's network. Yeah. We had one of the heads of the, VC alliance there at NVIDIA on the show, as well a couple of weeks ago at the NVIDIA GTC conference.

Jordan Wilson [00:07:57]:
So at that conference where we partnered, with NVIDIA, we brought you, some of the more promising, startups in the, in, NVIDIA inception program, and we brought to you a group, what we call the awesome eight. Alright. And, through your voting, we are now down to our final two. I know it's been a week or three since we had our first round. We had to, you know, go through and tally up all the votes and, coordinate with our, finalists here. And our two finalists are Glia Cloud and DMKTZ. Alright. So in a couple of minutes here, we're gonna be hearing, from their teams making their final pitch, for our prize.

Jordan Wilson [00:08:40]:
And, the prize is going to be a sponsorship here on Everyday AI. So, yeah, not just a live stream, but, you know, going out to our entire, podcast, library of 500 episodes and, you know, reaching, our audience in, as I found out the other day, 22,000 cities or 23,000 cities, which I didn't even know there are that many cities in the world. Anyways, so, pretty pretty cool, prize for our winner, either Glia Cloud or DMKTZ. But, before, we go and we hear the pitches, I wanna quickly shout out and recap, our other five, or sorry, our other six, AI startups that did not make the final cut. So deep checks. Alright? I'm gonna give you all just the quick bullet points in case you missed, that first episode. But if you did wanna go back, and, you you know, kind of hear from all of the founders themselves. I I think most of, most of the cases, it was the founders of these companies.

Jordan Wilson [00:09:42]:
You can go back and listen to episode four ninety seven, on our website or on Spotify, Apple, wherever you're listening. So, a couple that did not make the final cut, but some great startups. So DeepChecks. So they provide AI powered, platform to test, evaluate, and compare, generative AI models. They, help companies to measure model quality, detect hallucinations, and choose the best AI version. And what makes them different, they automate evaluations with what they call LLM judges in agent swarms, and that replaces manual test sets that don't exist right now for generative AI. Alright. So you can go check them out if you wanna know more about them.

Jordan Wilson [00:10:25]:
Expander AI, and and yeah. DeepCheck and Expander were right on the cusp there. They almost made the final. So what they do, Expander, they offer a platform to build custom AI agents by connecting multiple AI tools and internal systems. They help companies by saving engineering time by eliminating the need to build connectors or fine tune models individually. And, what makes them different? They support multi agent workflows, integrating various AI platforms, giving businesses flexibility without locking into a single framework. Alright. Beamer, another one of our awesome eight.

Jordan Wilson [00:11:06]:
So what Beamer does, they use AI to optimize video encoding using NVIDIA GPUs to reduce video file sizes by 30 to 50%. So that helps media companies, autonomous vehicle projects, and content platforms cut storage and bandwidth costs. And what makes them different, they leverage NVIDIA's, NVENC hardware acceleration for faster large scale video compression with minimal quality loss. Alright. PlyOps. So they provide software that maximizes GPU efficiency and reduces operational costs for AI data centers. So, how they help, they allow businesses to serve more AI users and applications with the same hardware budget. And what makes them different? Their plug and play solution requires no PhDs and optimizes cost per AI token, improving margins.

Jordan Wilson [00:12:00]:
Alright. And then a big a big name here. Yeah. We actually had, I would say maybe an underdog or two take out a couple big names. One of the big names, contextual AI. So what they do is they build retrieval augmentative generation or rag platforms that link company data to LLMs for accurate answers. So how they help? They help enterprises get grounded up to date AI responses across structured and unstructured data sources. And what makes them different? Well, they're founded by a meta rag co inventor.

Jordan Wilson [00:12:33]:
Right? So, someone from meta that co invented the rag, retrieval augmented generation technology, and they specialize in tuning models based on user feedback for continuous improvement. Alright. So that is, some of our other awesome eight that did not quite make the final. So let's go ahead and we're gonna toss it over to our two finalist companies. And, the way that you're gonna vote, last time we did voting, via the the the live stream comments and in the newsletter. So for our final round, it is going to be voting in the newsletter only. Okay. So make sure if you want to vote for Glia Cloud or DMKTZ, make sure you go to youreverydayai.com.

Jordan Wilson [00:13:18]:
Sign up for our free daily newsletter. And in the newsletter today, there will be a poll at the very top of the newsletter that you can vote. And then you can also, at any time, you know, our newsletter is both, an inbox version. Right? And there's a web version of it as well. So, even if you missed today's newsletter, I think we're gonna, keep voting open until Tuesday. So it should be about four or five days to go vote. So you can also go to read.everyday, sorry, read.youreverydayai.com, and vote that way. You do have to be signed up, to our newsletter to vote.

Jordan Wilson [00:13:55]:
Alright. Enough chit chat y'all. Live stream audience. Hey. As always, could you give me a little bit of help here? I'd appreciate it. We have our two finalists. I wanna know, can you hear, the audio? I can obviously hear it. I wanna make sure you all, can hear it.

Jordan Wilson [00:14:11]:
So here we go with our first, finalist from DMKTZ. Let's go ahead. There we go. Alright. So please live stream audience. Let me know if you can hear our first one here, and we're gonna kick it off with DMKTZ. Here we go. Alright.

Jordan Wilson [00:15:05]:
Y'all, sorry. I'm gonna have to click click rewind here. A couple people said we couldn't, hear the audio. Apologies. Let me try this, let me try this one more time y'all. And then, we're gonna try it a different way. Give me a second. This is one of the fun things one of the fun things about, doing this live.

Jordan Wilson [00:15:27]:
Let's try this. We're alright. Take two. Take two. Let me know y'all. This is always fun. You know, like, what's crazy is is I I'm I'm doing the exact same setup here, that I did last time, in terms of being able to, share the audio. I don't know why.

Jordan Wilson [00:16:16]:
Sometimes and, you know, everyone always says like, hey, Jordan. You should do more live demos. This is this is why I I'm not always a fan of doing the live demos because sometimes, you know, the technology, hates you. Alright. Let's try it one more time. Appreciate appreciate your patience y'all, as we, give our, give our, our finalists here, some time to, shine. Alright. Let's try this.

Jordan Wilson [00:16:44]:
Let's try this one more time. Let's try it one more time. Let's do this. Alright. So now we'll see if it works. I'm going to be, loading it from my Google drive here. So hopefully this doesn't let me down. Alright.

Jordan Wilson [00:17:02]:
Since it wouldn't work the other two ways. Alright, here we go. Let's try. Let's try this again. Y'all. Alright. Thank you again. Live stream audience.

Jordan Wilson [00:17:15]:
You always save my butt. Appreciate it. Alright. So let's try this one more time, shall we? Alright. We're gonna do it this way. Let's go ahead and open. Here we go. Alright.

Jordan Wilson [00:17:39]:
What is this? Take three here. Let's see. Hopefully hopefully, take three works. Let's go ahead. Add it to the stage. This is the equivalent of, you know, restarting your computer in the middle of it. Alright.

DMKTZ CEO [00:18:05]:
DMKTZ is your imagined apparel design. My name is Jack. I'm CEO and cofounder of DMKTZ. DMKTZ is your imagined apparel design and retail through AI, fit instrument, and body scanning. We wanna deliver perfect fit it made for you close

Jordan Wilson [00:18:22]:
Alright. Perfect. We confirmed. We have audio now. Alright. Now we're gonna kick it all off. So, after some audio glitches alright. Here we go.

Jordan Wilson [00:18:31]:
Our two finalists starting with Jack from DMKTZ.

DMKTZ CEO [00:18:36]:
My name is Jack. I'm CEO and cofounder of Demortize. Demortize is your imagined apparel design and retail through AI, fit instrument, and body scanning. We wanna deliver perfect fit it, made for you clothing that can be produced on demand. With just two photo or quick body scan, we generate an accurate digital twins of each customer's body. Our ASR then recommends the ideal styles, fit, and sizes, and shows how the garment will actually look and move in realistic three d before you purchase. But this is more than shopping assistance. Behind the scenes, we automatically generate major production patterns, allowing factories to produce truly personalized garment at scale.

DMKTZ CEO [00:19:25]:
DMKTZ is not just a brand. It's a full stack personalization platform that connects connects customers to manufacturers in real time. And what makes us truly defensible is actually our vertical integrated technologies and industry partnerships. We combine high precision body scanning, realistic three d try on, and fabric solutions into a seamless customer experience. We generate production ready custom patterns instantly. We work directly with Asia's top textile and garment manufacturers, growing access to fabric data, production network, and scalable back end infrastructures. Most solution focus on customer facing personalizations, but we go from scan to submission to productions, building an infrastructure that can transform how fashion is made and sold. We will start with customized apparel for professional sports and performance wear.

DMKTZ CEO [00:20:28]:
Targeting cyclist, fitness professionals who need clothing that fits, performs, and feels just right. In 02/2025, we released our first limited production lines in collaboration with Effolis, focused on validating fit, function, and desirability. In 02/1926, we'll expand into large scale productions and broaden into other active and lifestyle categories. This is how we build a brand that leads the apparel intelligence, helping the industry shift from guesswork and overproduction and toward data driven, demand based design and manufacturing. DMKTZ empowers fashion business leaders to operate smarter, leaner, and more customer centric. No more rely on demand forecast and overstocking. No more high return rates and fees issues. No more wasteful inventory cycles.

DMKTZ CEO [00:21:31]:
With the market, brand and manufacturers can transition to a modern age of mode, where every product is driven by real demand enhanced by data and delivered with precision. We are not just making apparel. DMKTZ is building a future of how apparel is made. Thank you.

Jordan Wilson [00:22:00]:
Are you still running in circles trying to figure out how to actually grow your business with AI? Maybe your company has been tinkering with large language models for a year or more, but can't really get traction to find ROI on Jenna AI. Hey, this is Jordan Wilson, host of this very podcast. Companies like Adobe, Microsoft, and NVIDIA have partnered with us because they trust our expertise in educating the masses around generative AI to get ahead. And some of the most innovative companies in the country hire us to help with their AI strategy and to train hundreds of their employees on how to use GenAI. So whether you're looking for ChatGPT training for thousands or just need help building your front end AI strategy, you can partner with us too, just like some of the biggest companies in the world do. Go to your everydayai.com/partner to get in contact with our team, or you can just click on the partner section of our website. We'll help you stop running in those AI circles and help get your team ahead and build a straight path to ROI on GenAI. Alright.

Jordan Wilson [00:23:05]:
So, thanks thanks for the, patience there, livestream crew. So, yeah. DMKTZ, pretty cool. Right? I mean, what do you think? So this is different from, a lot of the types of of of things that we talk about using AI for, here on the Everyday AI Show. So I have a a photo up, kind of on my, on the browser here on the live stream. And, you you know, the way that they use generative AI technology is pretty cool. Right? So you've maybe seen, you you know, some of these apps or pieces of software out there that, you know, allow you to kind of try on different clothes. Right? But then it's like, okay.

Jordan Wilson [00:23:47]:
Well, what happens then? Are those fake clothes? So what DMKTZs is trying to do is to actually, put these types of clothes out to production, but using generative AI to help, you know, fit, you know, better fit, better styles, more personalization, etcetera. But something more than just, you know, trying something on on a camera and, like, oh, look how cool it is when I can, you know, use this AI tool to, you know, redress myself or something like that. So a lot of cool use cases that you heard there, you know, by starting with professional athletes and getting, you you know, close that, you know, they hope will maximize their performance. So, you know, pretty pretty cool there from from our first finalist, in DMKTZ. Okay. So now, let's go to our, second finalist here, and let's hope we still have the audio. Alright. So, here we go.

Jordan Wilson [00:24:44]:
Here we go. Yeah. A funny one here from Sandra said, so no more deals at TJ Maxx. Yeah. I don't know if if if everything's, you know, personalized, you you know, for your exact body type. Yeah. I don't know if that would end up with TJ Maxx. Right? Alright.

Jordan Wilson [00:24:57]:
Here we go. Our second contestant, we have Glee a cloud. Alright. I'm gonna play this as long as the audio is good. We're gonna let it run. So, please, livestream audience, let me know. I didn't change anything. So, hopefully, we have audio here.

Jordan Wilson [00:25:10]:
Here we go. Alright. This is this is so fun. Alright. So, everyone just said, no audio on this one even though I'm I'm bringing it up in the exact same way. Alright. Loved loved to see it. Alright.

Jordan Wilson [00:25:43]:
Give it, give me one second here, y'all. Take 500, on this one. So, let me just go ahead, restart that. Love technology. Alright. Here we go. Let's, let's try it again. I forgot what take we're on here.

Jordan Wilson [00:26:03]:
You you you know, as as as much as I try to get this, correct, first time, sometimes it's, not that nice. Let's go here. There we go. Alright. Take 20. Let me know. Live stream, are we here in

Billy from Glare Cloud [00:26:19]:
the audio? I'm Billy, and I'm the creative technologist here at Glare Cloud. So what is Glare Cloud? Well, we're first a AI company here in Taipei. We help marketers automate value by producing high impact videos through a series of AI systems. So think about your product advertisements today. You need, like, teams of creative and productions to put together a compelling narrative, and you have to execute it. Right? So what we do is we provide this sort of content creation service that helps them to shave off more than half the time at only a fraction of the cost. And by high impact videos, we mean actual quantifiable performance metrics like your retention or your CTR rates. We work with over 6,000 customers across various industries like tourism, ecommerce, banking, and media.

Billy from Glare Cloud [00:27:15]:
And we're really grateful to see that many small businesses extending their reach through our collaboration. Our vertical AI is built for publishers and marketers. What we are offering here is very importantly not a generic tool. Our AI models are trained at a hundred million daily views from real publisher data where it constantly optimizes itself for retention and engagement. This means that every video keeps getting better results. Our entire pipeline is fully automated by AI where we handle everything from start to finish. Let's say from that means from script writing to rendering and even to API powered distribution. At this point, I think it's also worth mentioning that we kinda intentionally design our platform with this principle that we think AI should amplify human creativity, not replace it.

Billy from Glare Cloud [00:28:13]:
So think, let's say, automation that preserves creative agency. Because, I guess, from my sort of, like, know how, that's where precisely the marketers can then leverage. Our AI also dynamically adapts to your brand, tone, and style, and is able to localize whenever necessary without multilingual support. And this means you can add, multiple voice overs in various languages to reach out to the intended audience. This is also a point where I think traditional agencies are unable to match the speed, and regular tools also still lacks the depth we have. Lastly, our service is also infinitely scalable and replicable across platforms and verticals. You can achieve 10 times the output with the same resource you have. Regarding the outlook of our future plan, we try to be strategic, targeted, and sustainable as we are deepening our presence in existing markets like Japan and Korea while continuing to expand in Southeast Asia, as well as entering new markets like The Middle East.

Billy from Glare Cloud [00:29:22]:
Animation in our AI technology is the backbone of our growth strategy. We've been continuously fostering a diverse open minded environment, which has been very effective in cultivating creativity going into our next phase of growth. That will see us entering the world of agentic AI, enabling even more holistic integrations of intelligence within our customer experiences. Finally, at Glare Cloud, we like to see ourselves as a place that empowers industry by helping them access the, innovation and transformative powers AI, either through helping marketer or marketing leaders, connecting with their customers, creating scalable and impactful AI powered videos, or content that drives business growth. And I think that's it. Thanks again for, giving me the chance to share a bit about the cloud today. I'm excited about where we're headed, and I'd love to hear from your thoughts. Cheers.

Jordan Wilson [00:30:23]:
Alright. So thanks for putting up with all of the, technical mishaps there to bring you our final, two. So, Glia Cloud, I gave you all a quick, recap of DMKTZ. So, Glia Cloud, very impressive as well. Right? And seemingly ahead of the trend because I think we've seen, especially in the last couple of weeks, with, you know, even GPT four o, their new image gen tool, a lot of the tools that we saw from Google, that now make kind of what Glia Cloud is doing, right, more accessible to more people. So, Glia Cloud, it looks like was a little bit ahead of this curve in terms of using generative, generative AI, specifically for, media companies, creative companies, advertising agencies, etcetera, to help them be able to use generative AI a little faster at scale, and and to, personalize, content, across multiple, languages, mediums, etcetera. So, you know, two very, very unique finalists. You know, I kind of agree with, agree with David, here from YouTube who said this is tough.

Jordan Wilson [00:31:33]:
Both concepts are very good. I agree with that. But, it started with you all. We went from our awesome eight down to our final two for the inception games, and that's a wrap. We heard from our two founders. I gave you my two, quick 2¢ recap, but ultimately, it's up to you. So which company, are we going to thrust even more in the spotlight and give some, big reach to? So, you are the deciding factor. So if you haven't already, you need to go to your everyday a I Com.

Jordan Wilson [00:32:08]:
Sign up for the free daily newsletter. In today's newsletter, you need to go vote. It's gonna be at the top of the newsletter. You have until Tuesday. So whether you check your email inbox or like I said, you can just go to read.youreverydayai.com if you are already signed up for the newsletter and vote there. So make sure you vote. If if if you really want one of these two to win, you know, share it with your networks. If you're really pulling for one or maybe, you know, you're friends with one of the founders, make sure you share this around, to really amplify their message.

Jordan Wilson [00:32:40]:
Alright. Thank you all. I hope this was a fun one. Thanks again to the, NVIDIA, GTC crew and the inception team, at NVIDIA for partnering with us to make this happen. Thank you, livestream audience, for helping me, troubleshoot all these fun tech issues. And thank you all listening, for helping us do this fun little competition. So, that's a wrap. Let us know who is the top NVIDIA inception startup.

Jordan Wilson [00:33:09]:
Alright. Thank you for tuning in. If you haven't already, go to your everydayai.com. Sign up for that free delay newsletter. We'll see you back later for more everyday AI. Thanks, y'all.