I Pitched My AI App to Investors — Here’s What Happened
Other Videos📅 2025-05-09
Transcript
[00:00] In this video, you are going to see a real AI startup pitch deck. I'm going to present this pitch deck to you as if you were the investor. And what's crazy is we're actually using this for real investors for our company here. I think the best way to learn though is seeing it live. So, I'll be cutting in and out showing you why we did what we did. Let's go and jump into this pitch video. Hello, my name is Corbin Brown, the CEO and co-founder here at Bumpups. And by the end of this pitch, I'm going to prove to you why we're going to become the go-to AI model for video content. What problems are we solving with bump-ups? Well, the first one is that
[00:30] there is no easy and streamlined way to analyze a video of AI. Don't believe me? Go ahead and try to upload a video to any AI model. You know, it doesn't exist. Therefore, we come in to solve this as we are a specialization of an AI model to only laser in for video content and provide value from video content. As the current way of analyzing video and as you'll see through this pitch deck here is manual review. Eg. You tell Timmy in your team, "Hey, Timmy, watch this 30 minute video and find the right timestamps, find the right description,
[01:01] find the right titles. Let's not have Timmy do that workload and let's have Timmy use bumpups, which would make it so that it only takes 30 seconds." If any business developer, anyone in the world wanted to even analyze video with AI, they would have to build out their own custom pipeline with a bunch of fragmented tools. While Bumpus makes it simple, provide a link, a video, and you're good to go. Which leads to the big three here. Why are we going to win? Mass integration. We are going to integrate with every single video found on the internet. Whether it's on X, whether it's on LinkedIn, whether it's
[01:31] on Google Drive, YouTube, Vimeo, anything. We're going to be the oneizefititall for all video content analyzation. In addition to this, from the data we've collected from our users up to this point, we know certain questions are asked more than others. Therefore, this is where templates come into play. Not only do we allow the consumer to ask whatever they want, prompt the video, but we also provide very specific outputs. Therefore, you don't have to prompt. You have to make a custom prompt. You simply ask, "Hey, I want timestamps. Hey, I want summaries." These are reoccurring requests.
[02:02] Therefore, we can effectively create the best timestamps in the market or the best summaries for videos in the market because we've lasered in. And the really big one here is that not only are we available to all consumers in the world, but we are available to all developers and enterprise clients. We've built out API documentation that allows any business or developer to build out any type of infrastructure when it comes to AI video analysis at scale. This is typically the standard way you'd want to start a pitch tech. You want to give the problem. You want to give a solution. And why we do this is to prove to the
[02:33] underlying investor that hey, there actually is something here and we're not just creating something just to create something. I think the heavy hitter for us in this pitch deck is the idea of how easy and simple it is to analyze a video of AI compared to what the market provides now. All you need to do with us is simply provide a link or upload your video. Other platforms are fragmented. Other platforms don't do it as well as we do it. And you'll see that later in this pitch. So, let's jump back in. So, now that we know how we're going to solve the pains within the market, what is the product? How do we do this? This
[03:05] diagram right here looks like an amazing cup of coffee in the morning. you're just like, I need to drink that cup of coffee. If you don't like coffee, maybe you like tea, so green tea. But this right here shows you everything you really need to know about bump ups and our vision. We are creating the AI model that is going to be specialized for video content. Therefore, we need to tailor our AI model to allow for any type of consumer, whether that is the general public, businesses, or developers. We're making it simple. We don't need any of those three type of consumers to have to build out their own
[03:36] infrastructure within their backend or deal with the headache of trying to find something that is super easy and simple in the market. Therefore, those consumers can simply input any video found on the internet, pass it through our AI model and get three major types of outputs. Now, the first kind of output can be a custom type of output. I mean custom. Give me an example. You got a YouTube video of a podcaster. You put it in through our AI model. You ask for a newsletter. You get the content of a newsletter based off the analysis of that video. Boom. You're ready to go.
[04:06] Pre-build outputs. What do we care about at scale? Well, we know with video content, one really relevant pre-built output that we're the best in the market for is timestamps. We perform better than any other competitor when it comes to time stamps. We are specialized for it. That's why. So pre-built outputs has a pretty big play here as we can break this down by industry. Specific outputs due to legal, specific outputs due to health, specific outputs due to finance. We can build out a full low latency pipelines for each type of output.
[04:36] Therefore, the underlying consumer puts in a video, requests it, and we can push it out in 20 to 30 seconds. We beat everyone on speed. And the third type of output can be a combination of both. Whether in the workflow you're looking for some custom outputs and some pre-built outputs, we do it all. In addition, we have successfully partnered with Zapier with our API documentation which gives users across the world who don't want to code this out the ability to do it in a no code way. Furthermore, we are a part of Google's AI starter program. This slide right here is going to be probably one of your more
[05:07] important ones as this is the slide that explains the product. What are people investing in? Therefore, break it down as simple as possible. You don't want to write like two paragraphs here. Typically, diagrams that shows the workflow of how your product is integrated into an ecosystem is the best way to go. And then on top of that, we've added ethos with the Zap Your Partnership and being a part of the Google AI startup program. It shows we're real, ready to go, and we're ready to be implemented. Now that you know the product, let me show you the business model. How are we going to make some money here? We have the web app and we
[05:38] have the API. The web app has been existent for the last 1.5 years and the API just got launched last week. These two serve different purposes. While the web app is here for the general consumer, anyone in the world can access this through bumpups.com, upload any video and get valuable information out of it. The API exists in order to entertain the ability to integrate and build with enterprise clients or B2B. But in addition to this, we have officially become the frontr runner when it comes to other developers trying to
[06:09] create software businesses on the back of us in their tech stack. For example, we could have a developer that builds out a legal software and part of that legal software needs to have the ability to analyze legal depositions with interviews with clients and we could use bumpups and integrate within their tech stack. Then don't worry about building out the custom pipeline. Just integrate bumpups. You may have noticed that some of that slide was confidential, but that is the nature of raising money in this context. Now, what I want you to take away from that slide is that we give the idea of the product in the previous one.
[06:40] But then you dive deeper into how you actually monetize a product. Now, at bumpups, we have two main ways. Whether it's the web app or the API, web app is just another way of saying like this is the consumerf facing product in the sense of anyone, your mom, your dad, your brother, your sister could use the product or unless I guess if you're mom and dad and brother and sister are developers, they could also use the API, but you get the general gist. So now that you know the product, now you know how we're going to make money. Welcome to the team. I don't know who that guy is in the top left. He seems pretty cool. But my name is Corin Brown. I am
[07:12] the CEO and co-founder. As I stated earlier, I have three and a half years of running a corporation in the past. This was Luroma Therapy, Inc. This was a high-risisk business, a vape oriented business. I've been coding since I was 12. And I found myself back coding again because end of the day, I enjoy coding. I think it's fun. Creative juices. I'm a graduate at UC Berkeley with the region scholarship, which they only give to 2% of each class. And I am currently operating and running a YouTube channel that has around 96,000 subscribers. My co-founder here, Brian Quan, who is the
[07:43] CTO, is a heavy hitter. He has worked at other early stage startups and at Capital One. With three years of experience and a graduate at UCLA in computer science, you know, we are able to lock in and understand fundamentally how to build scalable architecture when it comes to software. And me and him have been doing it for the past year and a half. At least for this company, we've been doing it for the last year and a half, but we've been coding for a very long time now. When approaching when to present the team, some companies choose to present the team at the end of the pitch. I prefer doing it at the
[08:13] beginning. If you feel like you have a solid team, you want to give more context on essentially the jockeyies the investors are betting on. A company could be an amazing idea, a great product, but if the team is just not there or the team is not as strong as it should be, typically you wouldn't want to bet on that jockey. You want to bet on winners. And to prove yourself to be a winner, build a winning team. Now, the next slide I would jump into is the market size, but as you can see with that big confidential block there, we don't disclose this information unless you've signed an NDA. So, moving on to
[08:44] market adoption. How are we currently getting customers to our product? Now, we've been going down two major avenues for this company. Organic and partnerships. On the organic side, that has to do with the YouTube channel I'm currently running, and typically bumpups will show up in the description, essentially asking the viewer, do you want to chat with this video? because the underlying viewer could grab the link from the YouTube video, put it into bubs.com and chat with the video, get information out of the video. A really good use case of this is when I do coding based videos where I show you how to deploy and build out applications.
[09:15] The viewer could build out a step-by-step guide based off that video. The other way we've been approaching marketing is through commenting on other YouTube videos with relevant timestamps. Providing free value, as you can see right there, has its ways of showing to the community that Bumpups is here to give you the most accurate timestamps. And thank you, Mary, for saying, "Please do this for every episode. We will. We'll keep providing value." The other way we've been attracting new users is through partnerships. First one is starting here with Zapier, but we plan on expanding to make and pipe dream in the future. These are no code tools that allow businesses and developers to build
[09:47] out automations that they deem to be useful. And now with our new bumpups integration, any person or business could integrate AI analysis of video right away. Which jumps into our next slide here of market validation, which is where I would show real data of underlying traffic to the site, conversions to the site, and how that equates to real money being made with a software. NDA again as you see with the big green block. So what does the competition look like? How does the market look right now? We are the only
[10:18] platform that can give very advanced insight on videos and make it as simple as possible. The alternative ways of approaching this are going to be through Gemini, OpenAI, Amazon recognition. But what you'll notice later on in this pitch is that we outperform them on essentially every metric. And why? Because we're specialized. We are the ones building the AI model for video content. And over here, an example of a possibly easy way of doing it, but you get basic insights would essentially be screenshotting every like 2 seconds of a video and asking a AI chatbot like chat
[10:50] GBT. The next slide you're about to see, I would deem as one of the more important slides in this entire pitch tech as this gives crystal clear context of why you're better than the competition. But in addition to that, it gives context of where the investor should even place you within the market. And trust me, these numbers are pretty cool and these numbers just speak volume. So, you saw the chart before, but let's dive into the numbers of a case study we did here with a 45minute video. In this case study, we compared our model of Bump 1.0 to Gemini 2.5 Pro
[11:21] to GBT40. First thing I want you to notice between the comparison of these models, our max video length is 210 minutes. Google's is 45 minutes. And Chad GBT, they don't even have a native integration for video. Their method of analyzing video is photo by photo by photo. But as you could probably assume, that is very timeconuming and ineffective. Next, let's look at a very popular pre-built output that you probably see across different video platforms, which is timestamps. We are
[11:51] the best at providing accurate timestamps. Why? Because we're built for it. We're specialized for it. When we look and compare the timestamps found between Bump, Gemini, and Chad GBT, we just outperform entirely. accuracy of timestamps from with Google and CHGBT are just inherently inaccurate because the models themselves are too general. They're not built for it. They haven't created the correct pipelines to analyze the data in this context to give the timestamps that are relevant for the actual video itself. But here is the kicker and essentially the thing that knocks this out of the park is our
[12:23] integration list. We will be able to integrate with local videos, YouTube videos, Vimeo, LinkedIn X. We're talking every single link within the internet when it comes to video content. When we look at Google, we're looking at only local videos and YouTube and chat GBT, OpenAI, just images. Therefore, because we're so specialized, we are building out the infrastructure for pre-built outputs like timestamps, summaries, study guides, practice tests, anything industry to industry. But here's a number that typically shocks a lot of different people. How fast do we process
[12:54] these videos? And this is why we have an extreme first mover advantage. You took that same 45minute video and this was the time output. How long did it take each model to process for us? 20 to 30 seconds. For Google, 4 to 8 minutes. And for Chad GBT, because there is no native way of handling video processing, 55 to 70 minutes, which leads into our mission of becoming the AI model for video content. Personally, I really like showing that slide to investors as I find when you give real data and case studies, it becomes evidently clear your
[13:26] positioning in the market and why you will win. Now, outlined in that slide was a bunch of cool examples based on that case study personally from the calls I've done and the different conversations I had, the one that always seems to hit home the hardest is that speed. And as you know, at scale, speed is everything. And especially when it comes to back-end processing as the longer it takes to do something in the cloud, the more money it's going to cost you. We don't like that. Let's jump to the next slide. So what are our competitive advantages? Well, let me just name six. First one, we're first to
[13:57] market. We are the first company that has made the mission goal of specializing for analyzing video content at scale. We're here. We're ready to go. We have the mission. The second one is our speed and precision. As you saw with that previous slide, that we just beat everyone. And that's our goal. We're going to make it as fast as possible to get insight from videos. So when an individual or a company or a business or a developer needs thousands of videos processed in minutes, we can handle it. Third one, which I really like and what was really cool when coding this out was the ease of use. How easy it is to use
[14:28] bump ups comparative to building out custom pipelines, fragmented tools, everything of that nature. You simply just provide the file or provide the link to a video, you get everything. with our fourth one come up here. When it comes to our partnerships we are creating, Zapier is just a strong player. Zapier gives us an easy way to allow developers to get access to 8,000 different apps right away and integrate bump-ups into the workflow. And with the pipelines we're building with this AI software, the idea and focus on pre-built outputs is very important, but
[14:58] the ability to actually prompt the video is still there as well. custom outputs leads to us getting more data on the type of things consumers care about. Therefore, allowing us to build out pre-built outputs for those consumers. Eg. We see there is an influx of people doing custom outputs for essays on a video. Now, we can build out a pre-built output that is cheaper and faster for those consumers for the essays. They're happy, we're happy, everyone's happy. And another advantage we have for this company is my YouTube channel as I'll be
[15:28] able to show really cool workflows integrating bumpups whether we're building a software from scratch or building it out into zap year automations everything above the board we can show how to integrate bump-ups with the next slide would cover the product road map give context to the investors what's next requires an NDA as well and as we keep going here these last couple slides are require an NDA as this is going to be financial information but the next slide we did in this pitch deck was giving a lifetime revenue to show that we're a real product, we've made money and here is our growth rate. And the next slide here is our financial slide, which is where
[16:00] we're asking for how much money we want to raise and in addition to that, what our goal is for 1 2 3 years with said money. And finally, this slide right here is really cool as this is a PI diagram showing exactly how we're going to distribute the funds that we do raise, where the funds will go to, whether that is new hires, legal, operational stuff, everything above the board there. And finally, we have our last slide here. Thanking the investor. Nice little llama on a surfboard. Pretty cool. So, there is a walkthrough of a real investor pitch deck that I'm personally currently using and I'm
[16:31] personally currently pitching to real investors. So, if you felt like you learned something in today's video, make sure to leave a like. It's completely free. Those are going to be two random videos. I think that one looks good. That's my face. I'll see you in the next