Computer Vision and Poker: Megan Gray of Moment AI Talks Large Vision Models

Megan Gray, CEO and Founder of Moment AI, explores how large vision models revolutionize sectors like healthcare, mobility, retail, and government security. She explains how Moment AI uses proprietary data and computer vision to tackle data scarcity and enhance situational awareness. Megan recounts her journey from experiencing seizures and playing professional poker to founding Moment AI. She emphasizes hands-on AI learning, and avoiding hype.

🗳️ Podcast Feedback/Questions
🗳️ Nominate Podcast Guests

Quotes:
🏥 "I woke up in the hospital. They estimated I had 10 seizures."
♠️ "At the end of the day, there's no difference between a poker table and a boardroom table."
🔮 "I see the boom of large vision models … right now, we’re ahead of it."
📚 "If you're going to go towards AI, actually learn it. Don't just rely on your CTO or engineer."

Resources:
Udacity
NVIDIA Deep Learning Institute
Microsoft Learn
Learn with Google AI

Megan Gray is driven to use technology to improve lives. As a U.S. inventor with patents in AI, she is a pioneer in the field. With only 1.3% of the world's 340,000 AI patents being held by women as of the end of 2023, her contributions are significant. She has over a decade of experience in developing complex AI systems. Megan has spoken nationwide at National Academies of Science and Engineering, Amazon Alexa, NVIDIA, Google Startups, Verizon, and several universities. She's been featured in Forbes and Amazon Developer, and has also consulted with AI and product teams. Her previous role at Google sparked her curiosity to design and build bold, innovative technology. Currently, Megan is the CEO of Moment AI, a company that has raised millions from top VCs. She's committed to building AI that empowers everyone around her.

Connect with Megan! LinkedIn
Moment AI Website/IG/

Chapters

• 0:00 Season 1 Feedback
• 0:50 Future Podcast Guest Nominations
• 2:05 How Joan and Megan Met
• 3:23 Developing Large Vision Models
• 4:45 Utilizing Large Vision Models in Healthcare
• 5:40 Understanding Computer Vision
• 8:59 Enhancing the Tech Stack with Data
• 12:28 Expanding into Every Camera in the World
• 16:02 The Future of Large Vision Models
• 19:16 Prioritizing Team Care and Work Culture
• 27:12 Improving Situational Awareness in Robotics
• 29:02 Megan Gray's Career Path to CEO and Founder
• 29:28 Developing Technology to Recognize Seizures
• 31:47 Building an MVP with Raspberry Pi
• 33:30 Funding the Startup through Poker
• 35:43 Attracting the Attention of SoftBank
• 41:34 The Parallels Between Poker and Entrepreneurship
• 47:34 The Importance of AI Education
• 52:01 Focusing on Business Outcomes in AI
• 53:43 Gratitude for Support and Growth
• 53:57 Closing and Stay in Touch

Transcript

0:00

Hey there, as we wrap up season one, we are thrilled to share with you that season two is already in the works, woohoo! We've started recording with some fabulous guests and stories I'm confident you're gonna love. But my top priority is serving you, our amazing listeners, and we would love your feedback. What questions are on your mind? What intriguing things do you wanna make sure we cover in season two? Please help us make season two even more epic together, please head to our form in our show notes. You can find it at yourairoadmap .com / podcastfeedback. Share your thoughts, give constructive feedback. You might even be featured in a shout out of a future episode. Thank you so much, Tim, for that awesome idea, et cetera. We are so excited to hear from you and grow and make this podcast even better. We have so many. cool ideas for guests in the future. And there's so many cool stories and projects around the world. I swear every episode we do, every guest I talk to, my knowledge of AI and the field expands So if you have some ideas about nominations, that form for nominations, self -nominations, friend, colleague, mentor, aspirational nominations, yourairoadmap.com/podcastguest. Okay, thank you, let's go! Hi, my name is Joan Palmiter Bajorek. I'm on a mission to decrease fluffy hype and talk about the people actually building in AI. Anyone can build in AI, including you. Whether you're terrified or excited, there's been no better time than today to dive in. Now is the time to be curious and future -proof your career, and ultimately, your income. This podcast isn't about white dudes patting themselves on the back. This is about you and me. and all the paths into cool projects around the world. So what's next on Your AI Roadmap? Let's figure it out together. You ready? This is Your AI Roadmap, the podcast. Hey folks, this is Joan popping in to say a little intro to the episode. this episode is with me and Megan Gray. I met Megan years ago when I was the CEO of Women in Voice and she was a founder being showcased by Amazon Alexa as a founder. And wow, you all are in for a treat because Megan is not only a visionary, not only trailblazing in her part of the field, not only someone who takes adversity and translates it into huge opportunity, but her story is electrifying. I dare you to listen to this episode and not cry or have chills or really see potential tragedies as the expansion into a new avenue of your career. I literally got chills in this episode and was trying to keep my cool because Megan's story is so visceral. I really am excited for you to hear it. So not only the story part, of course, Megan is trailblazing in computer vision and really thinking in very innovative ways about how we leverage datasets, how we build datasets, how they can be used by Fortune 100 companies that she can't always disclose exactly how these datasets are being used. But I am so delighted for you to hear the story with Megan. I think the world of her. Go, Megan, go. OK, let's listen to the episode. All right. Hello, hello. Would you mind introducing yourself? Hi, I'm Megan. I'm the CEO of Moment AI. Cool. Well, and what are you all doing over at Moment AI? Yes, we are developing large vision models that leverage proprietary data so that partners can build within this tech stack. Cool. Okay. And can you give us more concrete details, an example of what that looks like? Yes, a lot of people know what large language models is. Large language model is ChatGPT and they're like following ChatGPT think we all saw like 50 other companies pop up. We're in the large language model. Large vision model, we actually try to close the data scarcity between cameras and categorize it through AI. So for instance, if somebody had a health event which a health care, additional health care role, they lose $90 billion in misdiagnosis every year. You're $90 billion as well in misdiagnosis just because in a hospital, you just rely on the human eyes, rely on a few samples. So we able to actually use a large vision model that uses machine learning, deep learning, which is the way that we train models. And we do it out. a lot of breakdowns through our clinical trials. Are we able to detect when these events are going to happen in a healthcare? But also we look at healthcare behavioral things of that nature. So you think of ChatGPT but think about if you was able to put that application with all the cameras in the world, that's Moment AI. that's so cool. I think that's, we talked to another guest recently about like multimodal. and how people think a lot about like text to text, but when we open up different modalities to be, you know, so multifaceted, computer vision is a piece of that puzzle of the imagery. For those people, like, okay, so I feel like, and maybe this is just in tech, we throw around the words computer vision or vision or sight. Would you mind unpacking what that actually means, at least on the technical side for folks? Yeah, so computer vision, For the most part, it's how you use algorithms to pair down with any type of cameras. I think a lot of computer vision, people see that in autonomous vehicle, for instance. All the cameras that you see on a WayMo or Cruise or things, that's computer vision. But if you see a camera even when you use an iPhone or Android, and they actually use your face, facial recognition is like a subset of AI computer vision. So pretty much computer vision. is anything that you could think of that uses a camera and basically a computer like an AI format algorithm to enhance what the computer is seeing. And so that's like a simple way to describe it. so almost any day you use some type of form of computer vision. Right. I feel like it's built in such that people almost forget it's a huge problem. It's a hard, I'm like, cool and challenging part of the stack. Exactly. And a huge problem because you have to make sure that the data is there, that the AI is being trained correctly, that privacy being used. And then it also depends when it comes to computer vision, which type of cameras are being used. Obviously, like the cameras that you might use in a house, it's not the same that you use on an autonomous vehicle, which is LIDARS or RADARs then, know, home security, you use things like maybe like an LG or something that's kind of more basic. There's so many different cameras and the computer vision or, the way that the detail all depends on the camera, the AI, there's so many things that go into making it better. that's so cool. Yeah. Well, I certainly I think on a very practical level, most people probably are thinking about when they upgrade a phone and just the, my gosh, the images are so much better. The size of the images when you're sending them to a friend. doubled or tripled. You know, they're multiple megabytes bigger. I think just all of our tech is jumping ahead. And I think hardware is certainly a piece of that. Gosh, I want to like go down a funnel with you about LIDAR and things. And I don't know if that's exactly pertinent. As you're building Moment AI, I'd love to hear kind of your journey into what you all are building, how it's being leveraged, kind of some of that. founder product journey. Would you mind sharing some of that? So at Moment AI we develop large vision models. So we develop them for a range of use, that's healthcare mobility or other areas that we are going into, only 2025 or 2026. But we really want to help enterprises leverage this data. So think of Moment AI when you think of a house, you think of a house and they have the foundation, the cement. That is what data is to AI. We provide this application of data to enhance the tech stack. And the tech stack is the product that the enterprise is real, eventually commercialized for the end user. And so we provide that data to enhance that product. For instance, one of our partners, we right now we're testing with multiple Fortune 100 companies. They've been excited. So 2023 has been, thank you, it's been a really exciting time for us. But one of these partners, are actually in a small town. And so they able to put our, put this application that we put, and they able to put it in their cameras to help monitor the long -term care. you know, is getting older, the demographics are getting older, you know? And so long term care has become a big issue, insurance is one of the best healthcare is the best healthcare. It's probably one of the few things that the government can agree that need to be focused on. Because I've seen it pushed by Congress and the White House, so probably like five things they can agree on. But long -term care is a very big thing because the baby boomers is getting older. And on average, a home nurse costs $3 ,000 a month. Most people do not have $3 ,000 a month. So we are able to provide this application, data to increase the monitoring for this enterprise that we're testing with right now to be able to help. you know, elderly people be monitored while their adult children, you know, able to maintain, save the workforce, go and, you know, live their adult lives. And so, but again, we have our large vision model scale from mobility to healthcare, to retail, to government security, so many different areas, but you always have to train the model for situational awareness within those different environments. And that could be, it could be, complicated for training with the environment and having enough data for it. So we actually, I have nine patents granted or pending and one of the patents actually deal with data augmentation. So data augmentation sounds really fancy. you. Data augmentation is just literally everybody, if you people that they talk about generative AI, that's literally a form of data augmentation. And so I actually, what I was able to do was I realized just collecting real human data, which through our clinical trials, we are with a major research institute by the Walton family that own Walmart. They gave us this research institute this summer to even go faster. And I realized that we do not only have to real human subject data, but we also have to have synthetic data. So one of the things that myself and my team developed was a generative AI model. So we call it a GAN model that was able to actually create synthetic biomarker data. literally, like, yeah, the synthetic biometrics, we were able to create that. So that's really, really complicated. We really going to just be able to help so many different industries and both, course. gosh, what a cool application of the tech. almost anyone listening, I'm sure has either a parent or a family member or someone who's getting older in your community and like hoping they're doing well. Like I think care is, as you mentioned, the amount of boomers who are aging and three grand. Not everyone has that kind of money to have in -house nursing, even if they want to stay in their homes. the sacred and important it is to take care of our elders and how big a complicated issue. I love that you're like, even our government agrees, which I don't know. That's so cool. And that's just one of the many ways that cameras, you think about all the places that cameras are, even right now, the way that we do this podcast is doing a camera. Moment AI, we plan to be a every camera interval, like every single thing, like you have to balance, you have to balance these different models to close the data to help these enterprises, you know, improve the way they interact with humans. So, you know, again, that's human -machine interaction right there. There's so many different ways that computer vision, you know, levels of play between human and machines. Yeah. Well, and Megan, I love how you're like every camera in the world, this ambition through the roof. Patents, I mean, when you're on the frontier of it, to be putting new patents out there. I just want to remind our listeners how fast and how cool your work is. It reminds me, I don't know if you know the work of Effectiva by chance. yeah, Effectiva, they was founded by MIT, maybe like 2013, 2014, 2015. So by women founders. Exactly, yeah. stay in touch with the women founders, their AI detected emotions were in a drive hole when it was driving. I remember reading about them, you know, and just being so admired by their work. And I know they got acquired as a congrats to them. Yeah, well, that's what I was gonna ask about because they're, as you mentioned, like, super detailed about emotions that you like literally can see on the face the wrinkle lines. If you watch her TED Talk, she talks about like how they decided to code different things, which is fascinating in and of itself. But I found it fascinating that they really pinpointed about drunk driving and tired driving as a liability and were eventually acquired by SmartEye, an automotive company. And just as the automotive industry thinks about, you know, the future of automotive. If I could just daydream with you, not Not to be, but when you think about if you want to be acquired one day, your company, do you believe it might be a healthcare play? you think, do have any sense of that? So actually, see us being, if we were to get acquired, it would be probably one of the big tech companies that want to, their goal is to be into multiple enterprises because that's us. not only, we actually tested with two major auto EMS right now. So it would seem like I have bags under my eyes because we have been... It's really, my gosh. It's crazy time because we are partnered with people in the health. We are partnered with Electronics Company. We partnered with Mobility. I mean, 2023 was our year. And so we are literally tested with some of the biggest players out there. But if I was going to say that we would get to acquired, know, don't get me wrong. At the end of the day, if the zeros are right... You know, we, will talk over, if we're asked to be acquired by, you know, one of these company, but I do think, when it comes to large vision model you think about the way with ChatGPT, they're a large language models. You suddenly had Microsoft, Google, all trying to, you know, create their own. That's I see the boom of large vision model being. Right now we are ahead of it. And so I do see these companies even want to acquire or partner with us, which we saw that with OpenAI with their large language models. so, you know, at the end of the day, we would have to look at the way they impact the team, the team job, the way that, you know, what's offered because sometimes because we are gathering a lot of patents, you don't want your company to be acquired just for your patents. That's a very realistic world. know, at any day we have to come down to what we've been offered, the patents, and also, you know, my job, I'm the CEO, but there's so many people's livelihoods that depend on me making the right decision for them and their families. So I'll also have to like, know, I'm gonna make sure this person, like, you know, they walk away with something that they not, not just leave with, you know, zero dollar. So I will also want to make sure I take care of my team who has been with me over these crazy years. Absolutely. Well and I think as we're both female founders and CEOs, I think there's maybe I'm biased in this sense. I'll put that out there, but I see different types of leadership. often when women are at the helm of a company. As you mentioned, taking care of your team. I certainly talk to my team about what is the ideal work culture. Like as we scale as a company, what principles do we want to put in place? What makes a fabulous workplace for my team? And really handing them the mic because retention is one of the things I'm most keen on that my team members are phenomenal. And trust me, at Women in Voice, people... had offers to be poached left and right. So why do they choose to stay? Why this company? Why this team? Is it at least in my beating heart related to team formation and retention? I really think so. think women, I'm not a mother but I have a lot of nieces. And I do think women, we naturally have this motherly instinct almost to take care of people. And I think that, you know, like the mother, tends to, even, know, forever the mother would lead the household when the dad go out to work or, know, so even for a long time, with my team, even though they all owed it to me, I make fun of that. It's just, I always tell them they old. Megan. My team, all older, they all have this experience and they all come from these big companies like Facebook, Apple. Lockheed Martin, all these different companies and they can literally work anywhere in the world. And sometimes I have to pinch myself that they chose me, you know? it's literally, if that's the thing, a lot of people, they're like, I chose you. I hired you. I feel like they chose me to put all their trust to, you know, to, okay, this is the CEO, this is the founder I want to work for. You know, because when you started off with, when you started a company, it's very early and you're like, okay, these people are taking pay cuts. know, they take your pay cuts to come up with you early. And yeah, you give them equity, but at very beginning, no matter how much equity, know, know Jay Z says, know, well, what's a hundred percent of zero. Zero. That's the truth of it. Like very early, you you offer them equity, but until that equity is worth something, until they help you build it, you know, that equity means nothing. And they come in and they have these, they like, you know, I can tell you literally all my colleague's kids' names. They coming in because they sending me pictures of their kids. like, hey, think, know. And they was like, hey, Megan, you know, I promised my daughter I'd take her to the water park. I was like, hey, you did a lot this week. You did a close deal. Go take your daughter to the water park. And so I think that's how you create loyalty. Because there's a lot of companies, especially in the AI world right now, the money they're willing to pay for AI engineers, AI scientists, data scientists, they're willing to put the big checks in. my CTO Matt, who was senior vice president of AI at Booz Allen and at Lockheed Martin. Matt can go work anywhere he wanna, he could try to get recruited. Same for my marketing people, my go -to marketing. So the fact they choose me, I don't take that for granted. And I think they see that. And I think they're passionate about what we're building. They see that it can become this unicorn company. But more than that, I think they're like, hey, I like working with her. I think that I like this team. Yeah. Well, I love that. And I think one of the things that I think is so opportunistic or like this moment is so beautiful and special because we're seeing all this innovation on the technical side, right? Literally on the cusp of this innovation and also our workplaces, especially post COVID. And with leadership, we get to decide like you've hit your metrics and beyond, please go outside, you know, take care of your body and take care of yourself and come back next week ready to kill it. I will push back a little bit the women are I think socialized, you know, to how we take care of each other. But I do believe it's a better world when we can be more holistic. And right, people aren't just pawns. But right, they choose to or people vote with their feet about a good team. I'm voting with the feet. You know, I really, I've noticed that top talent can, as you mentioned, work wherever they want, literally. Exactly. when you bring in these people, can work anywhere you want. So like the tech world is different than most industries. So, you know, I come from a background, you know, where my mom had to work two jobs. Did she like go to Josh? was smoking. No, but, know, it's different. It's different. The tech world, like a lot of, a lot of Americans, middle -class, they have to work a job just to, you know, put food on the table, pay the bills, things of that nature. And a tech world is like, Most of them, they choose to work there because they see the vision, they like it, they getting paid well. It was the work and family time, know, works for them. And so like you said, I do think, you know, since COVID, people are like, wait, you want me to come into the office five days a week? That's not happening. You're saying you do four day work schedules and people... They like, hey, I can do my job and still be at my, with my family, you know, all the time. Not my case, with being the CEO, founder, I do work all the time. So I don't know about you, Joan, you are a founder yourself, so you have to tell me you able to have this vacation time. I still haven't been able to schedule the vacation time for myself. Yep. No, it's a, it's a challenging thing. I'm not gonna lie to anybody. I'm doing the very like scrappy to more polished founder place and it's certainly when our revenue gets to a certain place and I want things to be more sustainable. And right now we're not quite there yet. And so I certainly don't want to let people know the hours I work or, know, but I really do. My team really works on milestone based work. I am not a micromanager. I think that's legit the worst. If you hit your project. you handed in and we agree on the deadline. I couldn't care less how many days you work or otherwise. Like the work gets done, the excellence is there. That's enough for me. We actually have the same model. We work on a milestone basis. know, everybody, can, the whole team can see everybody milestones because this person to pass on their loan to them. And so we look at that milestone, like you say, it could take three days and Everything you've done, everything that we need is done, Okay, go do your Thursday, Friday off, you know, but they're all milestone based. It's whatever they can. But it's not necessarily, nine to five, you know, then are times when you have a meeting. I like, hey, I can get on this call at 1pm. Yeah, but like for the most part, it's all, you know, they have a very flexible work schedule. Right, and the transparency you're even talking about, that people know each other's milestones. I think you and I may have worked at enterprises where there was not transparency and I could make many, jokes about that. But I wanna, we're hearing about your company, we're talking more about workplace than I expected, but I love it. I wanted to talk just one more, one big question more about the technical side. And then I'd love to talk about your journey because you have one of the coolest founder stories I know of. Because you were literally seeing the frontier of the technology that you're working on, it's pretty wild to me. If you could extrapolate out and kind of talk about what you see for the next few years, kind of the shape of your field, could you just dream with me about what you see upcoming? Yes, so I do think large vision models are also going to be part of robotics. Because again, when you look at the human machine interaction, what we're doing able to categorize and provide situational awareness around any environment. And robots going to have to be one of those things that have to know the situational awareness, the environment. They're gonna have to be able to categorize, just not object detection, object detection already out there. But say if a robot, because right now, Again, the reason I keep bringing in health care is not because it's our only vertical but it's one of the most complex verticals there it is to do. Right now, ever since COVID, there's been a shortage across America in nurses and doctors. Basically, they wasn't getting paid enough to do what they did, you know, with the job. I felt like there was a lot of unappreciation for the people that save our lives where we are in our homes. One of my sisters is a nurse so I got to this first -hand. I think, imagine if a robot by 2035, I wanted to say 2030, but first somebody had to build a robot before we could provide them the software. But by 2035, imagine if a robot could save somebody on the streets that just went to a healthcare crisis, save somebody that went here because... they already have been programmed to recognize this person having a seizure or cardiac arrest, or a stroke. This person, they have a diabetic shock. And so they can be injected with insulin or things of that nature, or this person just had a allergic reaction. so let me Imagine if a robot can be able to see this type of vision and recognize it. I think right now we've seen a lot of robotic companies. I think Moment AI, again, we're here to stay. I think that we would be able to help train the brain of the robots and help train the eyes. Yeah, more the eyes than the brain. Help train what they are seeing and help train the situational awareness so they can know how to act. Because we've just seen, maybe for the last, let's just say 15 years, you've just seen the autonomous vehicles. And you still see the autonomous vehicles crash and you know why? Because even with billions of dollars, thrown at autonomous vehicle, it's still lack that understanding situational awareness being able to recognize that for events that happening, because autonomous vehicle training. if all the drivers are perfect around you, if the. pedestrians use the crosswalk and don't Jay walk in the middle of the street. And it's very unrealistic. machines cannot be perfect because humans are not perfect. And I think that Moment AI this dream, you know, in the future, I do think that we can better the eyes of robots. So right now we're bettering eyes for humans. We're bettering eyes. know, for the cameras. do think that in the future that you will see more AI large vision models and robotics. That's so cool. Well, I think that also goes back a little bit to the multimodal thing or like some people are working so hard on the robotics, right? I've seen this, you robot that can peel a banana or something, certain things, but they need excellent computer vision to be able to enable or like different parts of the stack. improving dramatically. I definitely see you being there. Well, I know we're short on time. I could talk to you for so long. Can you tell people a little bit about your career path journey to being this CEO and founder of this phenomenal company? Yeah. So my path has definitely been different than most. I used to look at Google. So my background is in engineering. But one day I was coming back from a tech event, so I was coming back on the airplane when suddenly it felt like something was kicking me in my head. I had no idea what was happening. I get the flight attendant's attention and she keeps telling me that I have motion sickness. I keep telling her that I don't have motion sickness. There's something else. just would, it's still to this day so hard to describe this pain in my head. I knew of motion sickness felt like. Everybody was telling me it's motion sickness. But I just felt like somebody was kicking me in my head and I just didn't know what to do. And she moved me to the back of the plane with the bathroom, because she keep telling me, you have motion sickness. And the last thing I remember, I woke up in the hospital. They estimated I had 10 seizures. They had grounded the plane and they thought I actually had a brain aneurysm. So they put a blood leak in my head for those that don't know what that is. During the whole time I was unconscious because I was unconscious for a long time, they tested to make sure that my brain get functioning, that I haven't broken any bones or anything of that nature. So yeah, they tell me when I wake up, they tell me that I had 10 seizures and because of the convulsions, even in hospital, I still have convulsions. They said they're not even sure the reason why that kept happening is because the turbulence on the plane was but I spent eight to 10 months in and out the hospital following that. And sleep studies. And sleep studies, they tested for cancer, they tested for everything that you can imagine I had to be tested for. So like at one point I remember like a nurse coming to the hospital room and she wanted to stick an IV in my right arm and I told her that she couldn't. She was like, but she does that on the right. She rolled up my hospital gown and she see that I already have bandages on my right arm so she had to go to my left. And then when that lady saw that the next nurse saw the left was filled up, they had to go and do a butterfly IV in my hand. But during this time, being an engineer, decided that I would build my own technology to be able to recognize these different events. And so I started building in the hospital. I started building in and out the hospital. Another thing, I didn't want to stay with my mom. I didn't go to college and go work at Google just to have to move back home. With my mom, I love my mom, that's you know, that's nobody's goal in life. We're gonna sit and like rip at the door with your parents. so I developed this MVP, using a Raspberry Pi. A Raspberry Pi, by the way, is a mini computer. So most engineers use a Raspberry Pi, when they first do an initial prototype, especially one around computer vision. And I developed it not starting a company. I developed it just to be able to recognize my seizures. And then I posted on YouTube and I started getting emails and messages from all over the world. So I'm an engineer, but I'm not like somebody that's good at maps. So I would have to Google where people who like was sending me like messages from. remember like, like one guy even called and like left me a voicemail on my phone. And like I called him back and you would have thought I would like Beyonce or somebody. Cause he was like, my goodness, I can't believe you called me back. was like, yeah, it wasn't that hard. just hit the re-dial button. It was like, was like a surreal moment because people sending me their stories. like, you know, my family member had a stroke, my daughter, she, you know, she had cardiac arrest. and how can, you know, how can we help? won't, you know, can you detect. for our events and at one point I get a phone call from Softbank and JP Morgan. I'm sitting in my living room in Memphis, Tennessee, literally trying to figure out, like building this technology, trying to figure out if I would be able to live by myself because during this time, to be able to build the technology fully, I became a full -time poker player. I know this story, this part of the story, I was like, did I hear this right when I read this? Or like, poker player? So let me back up at first I think of JP Morgan. To be able to find the prototype and hire my first engineer, I became a four -time poker player. I grew up a military kid. So I was in like, ponytails from the early age, playing with my dad and his friends. And so I needed money and a lot of jobs require a driver license. When you become epileptic, they take away your driver license. Which makes sense, know, until they know that you have control over your seizures, they don't want you endangering other people. And so I didn't have a driver license. So the jobs narrowed down, despite my background in engineering, I was like, well, I need money because people just sent me these letters. I need to be able to start this company. So I was literally playing poker in Vegas, Atlantic City, traveling all over New Orleans. And so, yeah, so I built the initial funding. I remember I hired my first engineer and I said, hey, do you mind being paid on Sundays? Because I play poker Fridays, Saturday nights. And I would bring it in thousands of dollars. even I... Like I just, yeah, it was like amazing, you know? And that's why I think everything in your life is like, life already been mapped out for you. And like, I didn't know when I was younger that I was gonna play poker full time, you know? at that time it was just something I did with my dad. so I started building, I I paid this engineer to help me code. I, you know, buying business calls, I'm able to travel. was able to move at my mom's house and down the street, all through poker money. So then, you know, one day I sit in an apartment and SoftBank and JP Morgan called me. And I remember like SoftBank, tell me they're in New York to pitch there. I'm in Memphis, Tennessee. I pitched SoftBank, but I almost hung up on SoftBank. Just to clarify, I never heard of SoftBank a day in my life before that call. And I Google, who is SoftBank? Why aren't you on the phone with me? And I tell everybody that story, like, have to realize, this, not the normal, like, Silicon Valley type girl. So SoftBank is not, like, I don't think most everyday people would know SoftBank, you know? Because I was a founder that... saw a problem and decided to develop a solution. wasn't one of those people that was in college who was like, hey, I want to start my own tech company, become a founder. And so, SoftBank asked me to meet with their operating partners and engineers the very next day. And I stayed up the whole night Googling how to create a pitch deck. now that I look back, the pitch deck was Total crap by the way. It wasn't, you know, like just so we can clarify, was definitely, it was not good, but what was good was like SoftBank had like 20 people on the call. It's me. It's me and like 20 SoftBank. When SoftBank does a call with you. They don't like to waste time. They like to get everybody on here. They had engineers, they had tax people, operating partners. And I remember one of the tax people, she was like, So how have you funded the initial company? was like, through poker. And they all start laughing thinking I was telling a joke. And then I was like, no, I being dead serious. Like this is like through, through poker. And so like for like the first two years of the company, I still continued to pay myself through poker, just so I took that because SoftBank had ended up investing. And I ended up taking the initial fund for SoftBank at other VCs. And I pay other people when I'm not paying myself. I was like, well, I'm going to continue to play poker and for the funding, but after SoftBank funded National Science Foundation, they told me about a set of labs in DC. And I ended up moving to DC two weeks after the National Science Foundation contacted me. And yeah, was like literally like people, because I get asked by founders like, how do you meet SoftBank? I said, SoftBank called me. How do you meet National Science Foundation? I think my number was written on bathroom or something, because everybody just kept calling me. it was literally like, I was so blessed. People would either call me because they found my number online, or other people to tell them about me. So I'm really grateful for so many investors and founders. There would be some times that investors met me, but they wouldn't be in the AI space. They would be honest. you know, look, we know, we know nothing about AI. We know nothing about vertical cause this is 2020. a lot of people, they did not know about AI. I even now, you know, I had, people that we probably, well, I had a meeting with them and they said, you know, just to be honest, this is a big company, a fortune 100 company. was like, they do not know too much about generative AI. And so, you know, what happened, I was lucky enough to have ambassadors, you know, like say, hey, I might not know this space, but I know somebody who even though they don't know they'd be willing to take the risk. there's like, know, that really wasn't too big in 2020, you know? yeah, so from there was able to raise it. But literally, when I say I went all in. Literally went all in, you know, on the poker table to start the company in life because like at the time you got to think I'm epileptic so I have epilepsy, right? That's what they end up diagnosing me as and my medicine I remember 2020 my medicine was$700 because I'm getting three months supply. It was $700 and My mom was like you're going to play poker started the company. She was like She was like, no, you need to get a nine to five. was like, she was like, want to do engineering. She was like, she was like, but you just can't do that right now. You know, my mom, was being a mom. But I was like, mom, I was like, I want to have to do engineering. was like, I cannot let epilepsy define my life. And so, poker gave me this way because you don't need a driver's license to play poker. poker's like, yeah. pretty much you can play. They really don't ask too much, too many questions when it's time to play poker, and I was able to pay my medicine and you know, able to move to DC and do three years of testing and you know, get this initial data and when 2023 hit people start to listen more and people start to okay, you know, If you could create some charts and show us, so show us how big your total market is going to be. Then we'd be relative back. And then these mobility companies, these health care companies, electronics, they realize, OK, we need more data. And that's what Moment AI does. We are a core data company. And so they saw that it can influence their tech stack and the tech stack then influenced their revenue. Wow, Megan, I get chills. It's literally, you have such a phenomenal, and I think so, you have experienced these pain points and then building it, and I think especially like putting it on YouTube and getting a response from people who are, you know, in the trenches dealing with potentially similar things or family members, people calling you and being like, wait a minute, like I want this, like help, help, help. It's phenomenal. And I don't forget about those people. I mean, there's no way can't. When I mean that to hear different people's stories. so one of my favorite books is the Alchemist. the Alchemist is basically talking about your destiny in your life. And I do believe that having those seizures was meant for me to be able to develop this technology. know, wish they started off in healthcare. Yeah, we have expanded to healthcare. But at the end of the day, everything, all the technology we do is the large vision models. It ought to, you know, help people in the long run. And yeah, just, I mean, when I had like moms calling me I remember this one mom story about her daughter, it reminded me, about my mom. Like my mom was speaking to me. because when I woke up from the seizure, I see my mom, you know, wake up to your mom and seeing her next to your hospital bed. And, know, and then they try to see, you know, where's your memory? How's your brain? You know, we're in the United States, that's I answered that question too many times. And so, yeah, and now to take that and develop this company, I know that. For me, it's just not about the money, which obviously we plan on making a lot through Large Vision Model, but it's like, I get to help so many people's lives. I hope that these people, when they see me, whether it be on your podcast or somewhere else one day, that they see that I remember their stories. And I do that because I actually, not only that I started writing them back, there were so many people that my marketing team had to start helping write people back. Yeah, I mean it was really amazing moment I don't think too many founders get to experience how many lives they could change like I did and I think for me at least and no no judgment against others but like founders who know the pain point themselves Personally, I feel like show up differently in their companies and how they know the problems so well to build innovative solutions. They fell in love with the problem because they're dealing with it. Well, and I know we're almost at time. got just I think two more questions at least for you. One, think I might never ask another guest, Megan. Do you think your savvy in the poker arena has shaped you as a founder and entrepreneur? Definitely. So I tell everybody. Playing poker is literally like running a startup. There's no difference between a poker table and a boardroom table. The people in the boardroom just dress down. Literally, that's the only difference. the end of the day, it's a two -factor matter. And I always tell people, at least with the poker players, you know for a fact they trying to take your money. They are out to get you. The corporate world, the corporate world, the people that smile at your face and try to steal your money. They were smiling your face while trying to take your money, trying to figure out how to undercut you. In the poker world, at least they can look you in the eye and say, I'm here to take your money tonight. but they both are very, similar. So in a corporate world, I think the poker world did get me ready because in the poker world, you have to be able to read people. You have 30 seconds to make what we call a call. So you have 30 seconds to bet, make a call or fold your call. So you playing for thousands of dollars and you have 30 seconds to say yay or nay Or get out of the pot or like, you know and in the corporate world like I think I think because the poker allowed me to meet people so fast in a corporate world I'm able to read between the lines I get to read between the smiles and they let me know if I want to work with these people and that work with them and I think You know, I think it allowed, you know, me to my team to be supportive because they had seen some of these tough decisions. Well, have to tell people, we don't want to work with you. I don't care who you are. don't care what company you're from. And it's because at the end of the day, that poker world allowed me to read people better. And read like, hey, this isn't. what I think they presented it as, that's not a collaboration that they explain it as. so I think every founder, you tend to, go through this corporate phase where you get stabbed in the back or something of that nature. I think with poker, it had minimized that happening to me. And I'm saying it never happened, but I do think it has allowed me to dodge a lot of these potpoles. Because I have been a founder, we have a lot of founder friends. And so you hear these stories and you're like, my goodness, this person did what to you? VC did what to you? But I think, know, think poker definitely prepared me for the corporate world more than any college class I took, to be honest. Wow. That's incredible, right? And I think it's a crucial thing for any founder to, what do you want to say yes to and double and triple down? And what do you need to say no to? I, in the last year, I think said no to three big contracts that people were like, seriously? And I'm like, this is not the right fit. We don't have the same aligned goals. I want to triple down in the things that are working, you know? And I think that's a really important skill set. And deciding in 30 seconds, whoa. Well, I know we're just at the end here. If people are listening to your story and you have such a unique and phenomenal story, but if people are thinking like, how, what could I learn from her? What advice might she give me? People who are thinking about pivoting to AI, thinking about becoming a founder like you, thinking about innovating on a part of the tech stack. What advice might you give people? Yeah, I would definitely say, If you're going to go towards AI, actually, learn it. And by that, I mean, there's like these opportunities. Google has Udacity. NVIDIA also has these classes that you could take for free. I think it's up to founders. If you want to go to it, just don't hire a CTO or engineer, Also make sure that you understand what you're building. So many times right now, because AI is a money maker. I keep seeing companies just, we're going to just add AI chat bots. It's like everybody's been saying, suddenly everybody adding AI to the company and trying to call their company AI, they're really not. But I think actually taking the time to learn it on the weekend, there's so many free courses, again, that Google NVIDIA offer, I'm pretty sure Microsoft, most of these big tech companies, offer free classes that you could take within the AI space. And the reason I say that I presented our technology at NVIDIA. NVIDIA chose a few of their top startups to come present at NVIDIA headquarters last year. We were one of those companies. I remember what people, after we got done clapping, it was like. You was the CEO, were able to stand up there, not only present, but answer your questions. The other founders, they would get to ask questions. Hold on one second, let me, and they were telling like the engineer, they're saying you should come up. And for me, you know, I'm like, hey, if you're gonna be building this product, you're gonna be the face of the company building, you need to know actually what you're building, how it works. how to explain that without your CTO or your engineer coming up. I have a really great CTO. But one thing about it, sometimes we have to do separate meetings. So sometimes he could already be in a meeting, then suddenly this company wants to meet with me last minute. And they're like, hey, and I'm like, hey, I could hop on a call with you. You know what? Because I'm an engineer, I know the background. I've had tons of things of that, so I could hop on. And so one thing I think people need to start taking classes to really get themselves involved. And the reason why I say take the class with one of these big companies is because also on LinkedIn, I keep seeing people say, "generative AI expert," I can train you. Don't fall for the hustle people. Don't pay somebody like $300 just because they call themselves a generative AI expert on LinkedIn. Just go to one of the free courses. And the reason I say Udacity, because again, being a former Googler, know more about Udacity. But I know that NVIDIA also offer a lot of ways that founders or just people interested in learning basic AI can learn more. Take some of these workshops. I don't think you have to go back to school to go for anything of that. But really, if you're going to be in something, do it great. Take the time to learn about it. Just don't depend on your engineering team to know all the answers for you. Because I think AI officials, think these investors, VCs, are getting tired of that. I think they saw a launch of their money into these AI official companies in 2023. And these AI official companies, they're like, well, technically, because they use an open source, they don't even own the IP. I was like really surprised in 2023 when I saw investors investing into companies that were using open source and they have no IP Defensibility and it wasn't yeah that that mean that anybody with enough resources can copy it. You know when you don't have the IP defensibility you you get ten million dollars and the right team, They could build that exact same product I do think that to expand that product, they need to start learning the actual AI at least at basic level. Yeah. gosh. I agree with all the things you're saying. And I think one of the, in the marketplaces or on social media, like you mentioned, like AI this, AI that, and companies getting funded that were like, seriously, really, that? but I think when I talk to customers and they're like, we want AI on it. Like, how do you AI thought, know, just throw machine learning on it. I'm like, What is the actual outcome you want? How do we measure success in ROI? My team, we're like a dehype type of team. It's like, what moves the needle for you? What is actually going to change your cogs? I love dreaming of the case study that we build back from. And sometimes it gets kind of boring or practical, but so tangible to business outcomes that we become invaluable to our customers. as they scale. So I couldn't agree more. Megan, this has been such a delightful conversation. I hope to have you back one day that you could share your journey along the way. Where can people find you in Moment AI online? Yeah, so you can type in moment .ai into your web browser or you can... Go to LinkedIn. My LinkedIn is linkedin .com, slash, Miss Megan Gray. Very easy. yeah, I was able to get that, so I'm really glad about it. But yeah, we all plan those going. We're gonna have a lot of demo days and live streaming coming up. And so you want to follow and see how we're changing the way that people view vision. Yeah, just follow my LinkedIn as we continue to post some of those dates. Awesome, beautiful. And my team will put that in the show notes and hopefully I can't wait to see these live streams and demos, Megan. Thank you so much for taking the time. It was delightful to speak with you. Thank you so much Joan for having me. And you you was one of those, you were one of the first women in AI that I met, like maybe like two or three years ago. So, you know, just to see the growth of your company and just to be on this. podcast, you know, I'm truly grateful because you have been in my corner for a long time. So thank you so much for having me. absolutely. Well, a win -win. Okay. Have a good rest of your day. Bye. Cheers. Oh gosh, was that fun. Did you enjoy that episode as much as I did? Well, now be sure to check out our show notes for this episode that has tons of links and resources and our guest bio, etc. Go check it out. If you're ready to dive in to personalize your AI journey, download the free Your AI Roadmap workbook at yourairoadmap .com / workbook. Well, maybe you work at a company and you're like, hey, we want to grow in data and AI and I'd love to work with you. Please schedule an intro and sync with me at Clarity AI at hireclarity .ai. We'd love to talk to you about it. My team builds custom AI solutions, digital twins, optimizations, data, fun stuff for small and medium sized businesses. Our price points start at five, six, seven, eight figures, depends on your needs, depending on your time scales, et cetera. If you liked the podcast, please support us. Can you please rate, review, subscribe, send it to your friend, DM your boss, follow wherever you get your podcasts. I certainly learned something new and I hope you did too. Next episode drops soon. Can't wait to hear another amazing expert building in AI. Talk to you soon!