Pod Insights How I AI Using Veo 3 to create AI-generated music videos, like a Tiny Desk Concert with Notorious B.I.G. and Kurt Cobain About this episode Anish Acharya is an entrepreneur and general partner at Andreessen Horowitz, focusing on consumer investing and AI-native products. In this episode, he demonstrates how AI can be used for creative and personal projects beyond typical work applications. He walks through creating an AI-generated Tiny Desk Concert for Notorious B. I. G. and Kurt Cobain, building a book cataloging app using video analysis, and using browser automation for personal finance insights. Anish shares how these technologies allow anyone to bring creative ideas to life with minimal technical expertise, transforming what would have been impossible projects just a few years ago into accessible weekend activities.
What you’ll learn:
1. A step-by-step workflow for creating AI-generated music videos featuring artists like Kurt Cobain and Notorious B. I. G.
2. How to extract vocals from existing tracks to create unique audio combinations for your AI-generated videos
3. A simple method for cataloging your book or record collection using video analysis and Gemini Flash
4. How to use Comet to analyze personal finances and get investment recommendations without manual data analysis
5. Ways AI is transforming childhood learning and play by enabling interactive storytelling and creative exploration
—
Brought to you by:
Notion https://notion.com/howiai—The best AI tools for work
Lenny’s List on Maven https://maven.com/lenny—Hands-on AI education curated by Lenny and Claire
—
Where to find Anish Acharya:
• Andreessen Horowitz: https://a16z.com/author/anish-acharya/
• LinkedIn: https://www.linkedin.com/in/anishacharya/
• X: https://x.com/illscience https://x.com/illscience?lang=en
—
Where to find Claire Vo:
ChatPRD: https://www.chatprd.ai/
Website: https://clairevo.com/
LinkedIn: https://www.linkedin.com/in/clairevo/
X: https://x.com/clairevo
—
In this episode, we cover:
(00:00) Introduction to Anish Acharya
(03:05) How AI transforms creative constraints in music and video
(06:00) Creating an AI-generated Notorious B. I. G. Tiny Desk Concert
(07:36) Using GPT-4o to generate still images
(09:27) Using Hedra to animate still frame images
(10:40) Adding custom audio to video
(11:30) Using Adobe Audition to clip and sync audio
(15:42) How to use Demucs to extract vocals from any song
(16:36) Using Hedra to generate a Tiny Desk Concert featuring Kurt Cobain
(19:40) Creating a ’90s-style Nirvana music video with Veo 3
(27:40) Building a book collection cataloging tool with Gemini Flash
(35:35) Using the Comet browser for personal finance analysis
(37:20) How AI is transforming childhood learning and play
(41:23) Tips for getting better results from AI tools
—
Tools referenced:
• GPT-4o: https://openai.com/index/hello-gpt-4o/
• Hedra: https://www.hedra.com/
• Adobe Audition: https://www.adobe.com/products/audition.html
• Demucs: https://github.com/facebookresearch/demucs
• Perplexity: https://www.perplexity.ai/
• Veo 3: https://deepmind.google/models/veo/
• Kapwing: https://www.kapwing.com/
• Cursor: https://cursor.com/
• Google AI Studio: https://makersuite.google.com/
• Gemini Flash: https://ai.google.dev/gemini-api
• Comet: https://www.perplexity.ai/comet
—
Other references:
• Anish’s Notorious B. I. G. AI-generated Tiny Desk Concert: https://x.com/illscience/status/1935721063876550939
• NPR Tiny Desk Concerts: https://www.npr.org/series/tiny-desk-concerts/
• Notorious B. I. G.: https://en.wikipedia.org/wiki/The_Notorious_B. I. G.
• Kurt Cobain: https://www.kurtcobain.com/
• Robinhood: https://robinhood.com https://robinhood.com/
—
Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email jordan@penname.co.
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Speaker 00 It's like the most creative satisfaction I've had in my whole life. So I generated all these clips in a pretty straightforward way. I used GPT-4O to help me with the prompts. Said, hey, help me capture grunge 1990s Seattle, inspired by some of these music videos. And then, as you can see, it gets progressively more like camcorder, grimy. So I generated all this stuff and then I threw it together into a music video. All right, let's watch it.
0:28 Claire Vo? You get the patented Claire raised hands reaction on this one. I cannot believe this is AI-generated. It's so high quality. It's so specific in aesthetic, in a wardrobe, in emotion. You have inspired me after this podcast. What music video am I going to make? It's so much fun.
0:50 Claire Vo? Welcome back to How I AI. I'm Claire, product leader and AI obsessive, here on a mission to help you build better with these new tools. Today, we have a fun and inspiring episode with Anish Atria, general partner at Andreessen Horowitz and AI consumer investor. But we're not going to talk about portfolio companies or the future of AI. No, we're going to use AI to build music videos, analyze our bookshelf and help us plan our personal finances. Let's get to it. To celebrate 25,000 YouTube followers on How I AI, we're doing a giveaway. You can win a free year to my favorite AI products, including V-Zero, Replit, Lovable, Bolt, Cursor,
1:35 Claire Vo? and of course, Chat PRD, by leaving a rating and review on your favorite podcast app and subscribing to YouTube. To enter, simply go to howiaipod.com/giveaway, read the rules and leave us a review and subscribe. Enter by the end of August and we will announce our winners in September. Thanks for listening. This episode is brought to you by Notion. Notion is now your do everything AI tool for work. With new AI meeting notes, enterprise search and research mode, everyone on your team gets a note taker, researcher, doc drafter, brainstormer.
2:17 Claire Vo? Your new AI team is here, right where your team already works. I've been a longtime Notion user and have been using the new Notion AI features for the last few weeks. I can't imagine working without them. AI meeting notes are a game changer. The summaries are accurate and extracting action items is super useful. For stand ups, team meetings, one on ones, customer interviews, and yes, podcast prep, Notion's AI meeting notes are now an essential part of my team's workflow. The fastest growing companies like OpenAI, Ramp, Vercel and Cursor all use Notion to get more done. Try all of Notion's new AI features for free by signing up
2:59 Claire Vo? with your work email at notion.com/ how I AI. Anish, I am so excited to have you here and let me tell you why. It is because I have spent the majority of this podcast talking about enterprise B to B product management, how to manage your manager or manage yourself as a manager, or how to vibe code. That has been the topic of how I AI. And today we are just going to have a little bit more fun. So why did you start to come to these AI projects that are a little less like work related or technical? And, and actually just a little bit more fun. How did you, how did you get here?
3:42 Speaker 00 Great. Well, I'm excited to have some fun today. I mean, I've been passionate about music forever. I think most of us are. I've been DJing and making music for 30 years, but music is very constrained, you know? There's only so many ways you can work with it. An example of that is if you look at a track that has all the instruments mixed down into a final MP3 or WAV file, there's no way to just extract the vocal or just extract the drums. So you're really limited by a set of choices that were made in the studio. And with AI, you can do all this crazy stuff, like disentangle a track into just the vocals and just the instrumentation. So what really got me excited at first was everything you could do with AI and audio. And then that, of course, fed into all of the new video models and video gen and lip sync
4:24 Speaker 00 and all of the new technologies we're seeing. So it's just, it's like the most creative satisfaction I've had in maybe my whole life.
4:32 Claire Vo? Yeah, I agree with you. One of the things that I have so much fun with AI on is people are really worried that it takes away the most fun, most human, most creative parts of not just building things, but creating music, creating art, creating writing. And I, in fact, feel like it just gives me so much more tools, so much more breadth, so many more things I can play with and, and build. And so it really opens up this like creative artist side of me in a way that has been really hard to access as an adult, also with limited time.
5:02 Speaker 00 Yeah, no, and it's, it's actually a fun conversation we'll have over a glass of wine sometime. But if you look at music culture, music culture's kind of been defined by remix culture for the last 40 years. You know, like the mixtape was the first time that you could take the music and do something, you know, the cassette tape and do something of your own with it. And then that, of course, evolved into, you know, hip hop, which also sampled and which also had a lot of suspicion on it. But sampling was the foundation of hip hop. And I think AI is just the next manifestation of sampling and it'll be as important for music as hip hop was.
5:31 Claire Vo? Well, and we'll, we'll stop opining about AI and the arts. But the other thing that this remix culture makes me think about is kind of the next step that we've seen in the past couple of years, which is kind of audio and video remixing, this like TikTok memes, these dances, these things where you're taking a snippet of creativity, turning it into your own thing and then releasing it to the world in a, in a new version. So I definitely think we're seeing this, not just the audio side, but also at the video side, which brings us to your use case. So tell me what you built or what you created, maybe. And I'm excited to walk through how you got it done.
6:09 Speaker 00 Amazing, amazing. Great. Tiny Desk is the best. So if you haven't gotten into Tiny Desk, most people have seen it. It's just, it's so cool. It's so fun. And of course, you know, like creativity loves constraints and the constraints of Tiny Desk are incredible. There's a really good one from Clips that just dropped last week. And I mean, anyway, there's an infinite number of them. It's a fun format. It's sort of like the unplugged format of the 90s. So I love Tiny Desk and I got to thinking about all the artists I'd want to see on Tiny Desk. And of course, some of them are no longer able to be on Tiny Desk because they're not alive anymore. So that got me thinking about how I could do a Notorious B.I.G., Christopher Wallace Tiny Desk. And do we have the tools and technologies?
6:53 Speaker 00 And of course, can we do it in a way that's respectful and not derivative? And I did it and it seemed like it kind of worked. Maybe we can cut to it so your audience can check it out. And the workflow is pretty simple.
7:04 Claire Vo? We'll do a little clip of it, I think, and then we can work through how it got there.
7:35 Claire Vo? OK, we love it. It's great. And you made that.
7:38 Speaker 00 I did make it, yes.
7:39 Claire Vo? OK, show us how.
7:40 Speaker 00 And it took surprisingly little time. Yeah, so let me show you exactly how I made it. So I started with 4O. 4O is the best general-purpose multimodal model, in my opinion. I use it for everything. And I just asked it to generate an image of, we're going to do Kurt Cobain. That'll be fun today from Nirvana, of course. That's from when I was in high school, playing a Tiny Desk concert. So let's see what it comes up with.
8:04 Claire Vo? Wow, this is loading. You know, you mentioned that 4O is the best multimodal, all-purpose model. I generally agree. You know, 4O ImageGen had this super viral moment a couple of months ago when they released it. What do you feel like 4O ImageGen is particularly good at compared to some of the other ImageGen models?
8:26 Speaker 00 It's very good at prompt adherence. So you can do things, and I think that's because of the infrastructure underneath it. It's a different infrastructure from the diffusion-based models that preceded it. And BFL, Flux, a bunch of others do this now as well, and it's great. But I think it was just the most productive image model because you could manipulate it in such a fine-grained way.
8:46 Claire Vo? Yep, and I remember the biggest improvement when the 4.0 image gen came out is that it could actually spell things and write letters out. That was a magical moment. So I have to call out the NPR in the top corner. This image is actually done correctly. Look, there he is with his cardigan.
9:04 Speaker 00 Okay, I'm going to remove the guitar, actually, so that it is a cappella. Because I think that might work a little bit better. But look, this is the vibe of Tiny Desk. You know, it's as if you're seeing a photo from the 90s in a Tiny Desk studio. So I just, I love this. And I think that we've become so attuned to what's possible. We forget that this would be, you know, witchcraft three years ago. Witchcraft, right?
9:27 Claire Vo? What is the purpose of this? Are you storyboarding? Are you creating an asset that's going to go into another tool? Why start with an image on this file?
9:35 Speaker 00 So I'll talk through essentially what I'm going to do. So there's this product called Hydra, which is the best way to, I think, the best way to take a still frame and add custom audio to it. So create a video that has sort of animated from the still frame and includes the audio with the right lip sync. So, and there's a bunch of amazing tools to do this. Sync Labs is one of my absolute favorites as well. But Hydra is nice because it actually generates the video. So it does the text to video, or the frame to video, and then it also adds the audio. So what we're going to essentially do is take this frame. We're going to get the audio from YouTube. We're going to stem-separate the audio so we get the audio track we want. And then we're going to put them together in Hydra. And that's it.
10:21 Claire Vo? This really is remix culture.
10:23 Speaker 00 It's amazing, isn't it?
10:24 Claire Vo? It is amazing. Okay, so the asset that you really need to go into this video gen lip sync tool are two things. You need a still image that can be used to generate the video. And then you need some sort of audio to sync this to. So I know we're looking at this music example, but what other examples have you seen people use this kind of workflow for?
10:46 Speaker 00 I think we underestimated how useful it would be to add custom audio to video. And there's been a bunch of great, you know, one of the early examples was taking a speech that somebody was giving. I know Javier Millay did a really famous one. And essentially lip syncing, changing the language to English and lip syncing it. That went really viral a couple of years ago. So we've seen, and then of course, you can imagine a character, a photo of a character that you generate, and then you want to animate them doing something and speaking at the same time. So, you know, stories are told this way, and these technologies make it really, really easy to do so.
11:21 Claire Vo? Oh, we got him. Great. Okay, so now he's got bad posture, but we'll, we'll allow it. It's very grudgy.
11:27 Speaker 00 I think he always did.
11:28 Claire Vo? Yeah.
11:29 Speaker 00 Exactly. Okay, so now we've got Kurt. Now, what I would do if I didn't actually have a, so Tiny Desk has got a really specific acoustic aesthetic, which is it sounds like live instrumentation. So for the Biggie example, I actually found a Biggie cover band playing live in Brooklyn, and I pulled that down from YouTube, and then I extracted the actual vocals from the Notorious B.I.G. and laid them over. But in this case, Nirvana did a really famous New York City unplugged concert in 93. So there's video of them playing in the way that they would, and audio in the way that they would on Tiny Desk. So that is right here.
12:10 Claire Vo? Even in the same cardigan.
12:12 Speaker 00 Even in the same cardigan, isn't that amazing?
12:14 Claire Vo? Yep.
12:15 Speaker 00 Okay, so I used this nifty little tool called 4K Video Downloader, which is slightly sketchy, but that's okay.
12:23 Claire Vo? I love these little utilities that you just, you know, you Google like, how do I get audio out of YouTube? And then you go to the scariest website possible, and you just cross your fingers that your computer won't go up in flames, and you download 4K Video Downloader.
12:38 Speaker 00 Yes. My, yes. My-
12:42 Speaker 00 Data is definitely going somewhere sketchy as a result of this.
12:44 Claire Vo? So for the vibe coders that are listening, I have a request for startup, which is go, go find all these slightly scary little utils and build me ones that are less sketchy looking.
12:55 Speaker 00 A hundred percent, a hundred percent. It's a great idea. Okay. So now we actually have this. So we've got the video. Yep. Now we're going to open Adobe Edition. Okay. So this is, this is a tool that people who have been working in computer audio have been using for 30 years plus. It used to be called Cool Edit Pro. It's completely beloved and it's very, very easy to use, which is why so many of us use it. It was of course acquired by Adobe many years ago. It's now called Edition. So I go to Edition and I take this video and I just drop it in. So here we actually have the audio from the video, which is really, really cool. I'm going to zoom in and I'm going to see the first few seconds of it are blank. So let's just cut that out because we don't want to hear that.
13:41 Speaker 00 Then we're going to zoom out and we're going to take, I don't know, let's take 15 seconds. And you can kind of see the audio, the video in the bottom left corner there.
13:50 Claire Vo? Oh, got it. So it's combining the audio and video, just so you know exactly what you're syncing up to.
13:56 Speaker 00 Exactly.
13:57 Claire Vo? And I'm going to pretend like you're doing 15 seconds because we're doing a very efficient podcast here. But one of the limitations I know, having used some of these audio and video gen tools is you're getting small clips right now with what we're, we're working with. And so, you know, what I'm looking forward to is the day where I can have the, you know, hour long Nirvana Unplugged Tiny Desk concert. But you know, do you feel, do you ever feel constrained by the kind of length of assets being generated or the quality?
14:30 Speaker 00 I mean, sort of, but again, I think creativity breeds constraints. So to, not to over rotate on hip hop, but if you look at the reason that so many samples were used in hip hop and creative ways in the eighties and nineties was the actual drum machines and samplers had very limited sampling time. So you could only sample a second of anything. So you couldn't really sample four bars. And that's why so many producers put tracks together that use these many one second samples in surprising ways. And once we actually got the technology to sample for more time, we actually got less creativity, I would argue. So I sort of love the constraints that the technology gives us today.
15:09 Claire Vo? Well, I also love my complaints. I'm like, isn't it annoying that you can't provide Nirvana and overlay their audio and generate a completely fictional concert for longer than 15 seconds in probably under a 30-minute podcast live?
15:24 Speaker 00 Totally.
15:24 Claire Vo? My complaints are so ridiculous because the idea of creating something like this even a year ago sounds so, as you said, impossible that we get so spoiled once we get used to these tools.
15:37 Speaker 00 100% right. No, exactly. Like, I mean, this stuff, we would have called it witchcraft three years ago. It would have been. Okay, now, there's two things you can do with this. If we wanted to do an a cappella-only version, for example, we can use a technology called DMUX. So DMUX is this amazing technology that allows you to to extract the vocals from any song. So here, I've forgotten what the actual command line is. So I just do this. I looked it up in perplexity. What's the actual way to extract two tracks with DMUX? We do this, DMUX two stems vocals, and then let's go find the path.
16:17 Speaker 00 Okay, so this command is going to take that audio file we saved of the first 15 seconds of this concert, and it's going to extract the vocals from the instrumentation. So this will be Kurt Cobain singing "Come As You Are" a cappella, which, as far as I know, has never happened, which is pretty cool. And then we simply come back here and we say, "Start frame, upload an image." Let's use this. Okay, that's our Kurt Cobain. Audio script, upload audio. And let's use, actually, the full audio with all the instruments. Add to video, and then we just say, "Man singing on Tiny Desk."
17:00 Claire Vo? What I love about your prompting compared to other HowAI guests is every prompt has been sub-six words. Six words. You're very simple in terms of describing what you want and get high-quality outputs there. So I don't know what that says about the prompt engineering industrial conflict. But proof here that you can use simple prompts to get pretty cool stuff if the tool behind the scenes does the work for you.
17:27 Speaker 00 I think you've got to give the AI the space as well. You know, if you overly constrain it, it just really struggles to satisfy you. Whereas if you give it less constraints, you know, sometimes it has unexpected results, but often they're unexpected, you know, delightful.
17:41 Claire Vo? Well, that's what I've heard a lot from folks that come from the more creative backgrounds. Designers in particular tend to be less precise in their prompting because they want that exploration space that then they can narrow in on. And so I really think it also comes into play, your prompting technique can come into play based on kind of what profession or what background you're coming from. Engineers want like the most precise. They not only want the code to work, but they want the code to be written exactly how they would write it. And so they're very precise in their prompting. Where I've found designers and more creative folks building different kinds of assets really like that wide open space.
18:20 Speaker 00 Totally. Yes, exactly.
18:23 Claire Vo? And while we're waiting for this to load, it might be interesting. I'm just looking at some of the options at the bottom here. So you have different kind of models that you can use, including one that looks like that they specifically fine tuned for this. Different aspect ratios, orientation, length, probably based on the script. And then, you know, the prompt says, prompt your character with emotion and gesture. So I am very curious if you put like
18:50 Claire Vo? angsty man singing versus cheerful man singing, if you'd get a different, a different version here, even if the audio and video were the same.
18:59 Speaker 00 It works really well. Absolutely. Yeah. No, this is such a useful storytelling product. It's amazing. And when you combine it with other video gen models like VO3, you can start to tell real stories.
19:11 Claire Vo? Yeah. Okay, let's check it out.
19:12 Speaker 00 All right.
19:25 Speaker 00 All right. Pretty cool.
19:27 Claire Vo? It's very good. It's very good.
19:31 Speaker 00 Very satisfying.
19:32 Claire Vo? He even manages his mic well, you know, pulls back on some of those
19:37 Speaker 00 Totally.
19:37 Claire Vo? notes. That's incredible. And so, you know, could you take this and take different clips of the video and sort of generate a string of these, these videos and maybe put them together in a longer form version?
19:52 Speaker 00 A hundred percent. Yeah, I actually was inspired by this. So I put together a music video, a little mini music video for a different Nirvana track. Can I show it to you right now?
20:03 Claire Vo? Yes, we would love to see it.
20:05 Speaker 00 Okay, I used VO3 to generate the clips and it turned out great, I think. Hold on one moment.
20:10 Claire Vo? Yeah, and I think if you haven't tried VO3, it is pretty incredible. I mean, I can only generate like two and a half videos every day of three, you know, seven second length or whatever. I'm still capped on, on usage, but the quality is really good. The physics are really good. It's one of my favorite video models to play with right now. Just as a, just as a consumer, it's, it's kind of, it's, to me, my experience with that model has been, was very similar to my first experience with Mid Journey, where just the breadth of things coming out of the model were so incredible to me. So highly recommend folks give that, that model a little spin.
20:55 Speaker 00 It's amazing.
20:55 Claire Vo? Yeah.
20:56 Speaker 00 You've, you've got to get on Gemini Ultra, Claire.
21:00 Claire Vo? We have a, a household Gemini Ultra account.
21:05 Speaker 00 Okay.
21:05 Claire Vo? But my husband is the, is the video gen guy. So he, he's up there. By the time I get to it, we would burn through some tokens. But you know, I read all the, I spent all the money on cursor. So, you know.
21:19 Speaker 00 Fair, fair. I know. My wife, for the first time this month, was like, babe, what is cursor? I'm like, ugh, don't worry about it.
21:29 Claire Vo? I know all these like little secret AI tools popping up on the credit card.
21:35 Claire Vo? How I AI is now on Lenny's List with my personal selection of the best AI engineering courses on Maven. You can spend months thinking and playing with AI before really integrating it into your workflow or shipping an actual AI feature. If you want to start building, then these hands-on Maven courses are for you. Learn directly from Aishwarya Naresh Riganti, MIT instructor and AI scientist at AWS, or Sandra Shuloff, who has authored research with OpenAI, Hugging Face, and Stanford. To pivot into an AI role or successfully
22:13 Claire Vo? lead your company's next AI initiative, visit maven.com/Lenny to enroll now. Use code Lenny's List for $100 off. That's M-A-V-E-N dot com slash Lenny to get ahead in the AI era and start building.
22:35 Speaker 00 So this is, these are all the videos that generated Google Flow. So I was trying to capture like a 1990s high school band auditorium, you know, a little dystopian energy. So I generated all these clips in a pretty straightforward way. I used GPT-4O to help me with the prompts because as you can see, this is actually the beginning of my generations. This doesn't, this is like the complete wrong energy, you know, this, I don't know what this is, like early 80s, you know, synth pop or something. So then I went to GPT-4O and said, hey, help me capture like grunge 1990s Seattle, you know, inspired by some of these music videos. And then as you can see, it gets progressively more like, you know, camcorder and sort of grimy. So I generated all this stuff and
23:19 Speaker 00 then I threw it together into a music video and I put the music behind it. I'll show it to you right now.
23:24 Claire Vo? Amazing. So just restating this, 4O, helping you refine your prompts to get the aesthetic right, the phrasing, the prompting right, give you some keywords, Veo to generate these like shorter clips. And then you put it together in like Final Cut or something like that?
23:40 Speaker 00 I put it together in Capwing. Capwing is so easy and so useful. I highly recommend it.
23:45 Claire Vo? I'm a Tiptop girl, so I use CapCut.
23:48 Speaker 00 Yeah, gotta get on Capling. Alright, let's watch it.
24:57 Claire Vo? Okay, you get the patented Clairvaux raised hands reaction on this one.
25:02 Speaker 00 Love it.
25:03 Claire Vo? I'm gonna tell you the real truth. Something like this makes me almost wanna cry because I really got into technology. I wanted like everybody, I wanted like make video games and like make movies and work for Pixar or direct like and it always felt so inaccessible to get these like amazing ideas that I had in my head
25:23 Claire Vo? into a thing. Like, could you film it? Could you access the people? Did you have the time? Did you have the music? Did you have to create? And you just put together this amazing, amazing music video.
25:36 Speaker 00 Thank you.
25:37 Claire Vo? I'm so impressed.
25:38 Speaker 00 Thank you. It was so fun. It was so easy. And I also like music videos are a lost art form.
25:44 Claire Vo? Totally.
25:45 Speaker 00 I'm so excited to see, you know, everybody making music videos for all their favorite tracks because what a cool way to contribute, you know? And in no way does it actually dilute from the original. I think it's a
25:55 Claire Vo? No.
25:55 Speaker 00 It's a testament to the original and our appreciation of it.
25:58 Claire Vo? No, it looks like a love letter. And I have to, I have to call out, when I was watching it, there's a lot of it that I think is incredible. I like how the cameras, you know, like pan and zoom in. The part that really got me was the sequential shots of the teenagers in the hall. And I was like, I cannot believe this is AI generated. It's so high quality. It's so specific in an aesthetic, in a wardrobe, in a emotion. And it got me until, and again, they are three good at physics, until there's like a guy with like a pack of Camel cigarettes on his arm. And like the cigarettes are like halfway coming out.
26:34 Speaker 00 Yes, yes, yes, totally. That's right. Well, the actually, and the other funny artifact is if you look at the end, when the band is playing and a bunch of people are jumping out of the crowd. Four people jump out of the crowd at the same time. They look the same and they're making the exact same like, you know, like they look like acrobats at a circus or something.
26:52 Claire Vo? It's like the end of like an eighties TV special where they all jump up.
26:55 Speaker 00 Totally. Yes, yes.
26:57 Claire Vo? Um, that's amazing. You have inspired me truly after this podcast. I'm like, what, what music video am I going to make? It's so much fun.
27:06 Speaker 00 Do it, do it.
27:07 Claire Vo? Music, I mean, music videos, you could do like fake movie trailers.
27:12 Speaker 00 Yes.
27:12 Claire Vo? Also documentaries, I mean, we're doing the fun art, you know, heart, heart and soul filling stuff. But I also think the ability to create educational materials that are compelling and interesting with this technology are also right there.
27:29 Speaker 00 I mean, if you look at fanfiction, fanfiction's enormous because people want to contribute to the things they love. And now we get fanfiction for every medium. It's so cool.
27:39 Claire Vo? Okay. Sold. All right. That was, that was just workflow number one. We're going to go pretty fast through workflow number two, which I think is a little bit more of a practical, practical one, but still connected to, to the arts. So walk us through what your second, second workflow is.
27:56 Speaker 00 Cool. Yeah. So one of the things that I think is really under hyped, underappreciated, underused is all of the multimodal capabilities. And the model that does this really well actually is Flash, Gemini Flash. So it's just, it's great. It's one of the very few models that can do video analysis and ingestion. It can do all kinds of amazing things and yet I don't see it being used out there a lot. I thought I would use it to create an app that would help me catalog my record collection because I've got, you know, like every DJ, I've got so many records and it's such a pain to keep track of them and know which ones I had and which ones I didn't. So I did a very quick app on Friday that let me take a video of flipping through my record collection and then using Gemini to extract artist names, album names, photos.
28:41 Speaker 00 It's, it's really, really cool. So I thought today we could do something similar except for books.
28:46 Claire Vo? Ah, this is amazing. And we were talking before we started recording, this is going to help me because over here I have like a hundred books and a hundred records piled up on shelves that have definitely not been catalogued. So I can't wait to see what this looks like.
29:01 Speaker 00 Perfect. I got you. Let's share. So here we are in Google AI Studio. So I'm sure folks are familiar with AI Studio, but if you're not, actually, I think it's the best product surface to interact with all the Gemini models, one of the best anyway, because it doesn't have all of the kind of overhead and links and constraints that a lot of the other Gemini products have. This feels like somebody just took a blank piece of paper and brought the best manifestation of the Gemini models forward. So I really love AI Studio. It's my starting point for all of these things. And then in AI Studio, you can see here, you can of course chat, you can stream with your phone or with your webcam, you can generate media, and you can build apps. This is a very good app builder and
29:45 Speaker 00 this is the best way to build off the shelf apps, I think, that integrate with Google models. So here I've, I've typed, you know, create an app that takes a video of a person flipping through their book collection and extracts the author and title of every book shown. Then I give it a suggestion for how it could do it, which is you could do this by taking the video and first extracting the frames that show distinct books and then have a vision model analyze those frames to extract the information. Make sure you extract every book shown sequentially.
30:20 Claire Vo? What I have to call it here is, you know, what's interesting is people know that these models exist and they generally know some of the capabilities, vision, you know, text to speech or speech to text, all this stuff. But what's really hard for people to do, and I appreciate you showing us, is think of novel ways you can access the abilities of those models. I would have, actually, I thought you were going to show us, like, you took a picture of it and you cataloged it. But this idea of a video and then extracting the frames, I just haven't changed my mental model to match these multimodal models in order to take, you know, take advantage of things that can be more efficient, allow you to do things. And so I really think it's great that you're coming to this from how
31:05 Claire Vo? could I solve this with audio? How could I solve this with video? How could I solve this with text? And knowing that the models can do kind of the hard work on the backend.
31:14 Speaker 00 Thanks. Yeah, look, I completely agree. And a video is just, of course, it's so much more rich than image. And this is the way that we build, we bring a lot of the outside world online, I think. So I've been really inspired by video. I saw something on Twitter where somebody had set up a mini app that watched him shoot free throws and kept count. You know, you could, I mean, there's just so many ways that this will be productive. I'm very passionate about AI for parents. And I've got kind of a neat video idea in there as well. So to me, there's like the sort of skeuomorphic technologies, which is using the new technology with the old assumptions. And then there's the native ways to use it. And this feels like a very native way to use the models.
31:52 Claire Vo? Well, to connect the two things that you said, the, you know, basketball shooting analysis in kids, my husband did upload every single one of our eight-year-old's basketball games to a video analysis to get like each kid's staff.
32:06 Speaker 00 No way.
32:07 Claire Vo? Shooting percentages, all the, they actually don't even keep score at this age. So you got to like get the score.
32:14 Speaker 00 I love that.
32:15 Claire Vo? Yeah.
32:16 Speaker 00 I totally love that.
32:16 Claire Vo? Okay, so now we have an app.
32:19 Speaker 00 Yeah, so I'm going to take a video here of me just flipping through my stack of books.
32:32 Speaker 00 Okay, I've taken the video.
32:34 Claire Vo? Okay, and that took all of seven seconds.
32:38 Speaker 00 Exactly, yeah. Now, you know, the one edge here that's kind of interesting is this is really, it's really easy to get something working. But if you want to publish an app that a lot of other people can use, it then becomes more work.
32:50 Claire Vo? Yeah.
32:51 Speaker 00 So I probably, it took me 15 minutes to create this for my record collection, at least create the working demo in primitive. But then it took me half a day to get it live so that anybody could use it.
33:03 Claire Vo? And what's interesting about that is I feel like a lot of individuals are just going to build their own tools and presume other people are going to build their own, their own tools. And so maybe this will just inspire somebody to build their own record collection extractor, which might be faster than trying to find yours online and reusing something somebody built.
33:22 Speaker 00 I mean, the era of personal software is upon us, you know?
33:25 Claire Vo? Totally. Okay, so what it's doing, taking this video, it's going to do frame by frame extraction of the, again, something that is just so time consuming. And then it's going to use the vision capabilities. What model do you know is behind the scenes of all this? You say Flash?
33:40 Speaker 00 It's Flash, Flash 1.5.
33:44 Speaker 00 And I can kind of skip ahead and show you what this looks like. So here is, here's one that I built yesterday and which with essentially the exact same prompt.
33:52 Claire Vo? Yep.
33:53 Speaker 00 So let's, let's run it in parallel and see if this one's any happier with us.
33:57 Claire Vo? Okay.
33:59 Claire Vo? And I did notice one was light mode and one was dark mode.
34:03 Claire Vo? Was that just the AI?
34:03 Speaker 00 Yeah, this is just some of the randomness of the models. Yeah, exactly.
34:08 Claire Vo? Oh, I do, I do have to say I like the progress indicator of the second one. It told me how many frames it's extracting. Oh, look at this.
34:16 Speaker 00 So here we go, you know, this is the Chris Dixon book, the Paul Graham book, this very nerdy book that Mark asked me to read when I was hired. This is a really good Thomas Sowell history book. Anyways, this is my entire stack of books, every single one of them. You can see a photo, it's extracted, the author and the book name. So it's like, you know, this is just a couple of prompts. That's it. And it generated it. So this is what's possible. And then if you go here to deploy with Cloud Run, you get a deployed version of it that's actually running on the cloud. And now you can send this link to anyone. Now, this is going to cost you API credit. So maybe you want to be a little bit deliberate, but you're pretty much ready to go with this really sophisticated
35:01 Speaker 00 video processing app that would have taken, I don't know, a month of time previously.
35:06 Claire Vo? Yeah, amazing and so useful because now I can figure out which of these also very nerdy books we have, we've read. I also see some duplicates up there.
35:16 Speaker 00 Totally. Yes, yeah. It's not perfect. Yeah, exactly. Well, actually, in this case, the photo is duplicative, but it detected the Ben book and the Chris book separately. So, but yes.
35:28 Claire Vo? I need this, man, I need this for the pile of kids books I have up in my kids closet so they even remember what they have.
35:36 Claire Vo? Okay, this is great. Well, thank you so much for showing us these fun use cases. I have to call out, as we hop into our lightning round, one thing I noticed, which is you are using Comet.
35:48 Speaker 00 I am using Comet.
35:49 Claire Vo? Tell me a little bit more about why that new browser is your browser of choice? And what are you getting out of it?
35:58 Speaker 00 Comet is so good. I mean, I've been skeptical of the new browser thing because it just feels like the ways to improve the browser in the past have been very incremental, ambitious, but there just wasn't that much surface area for new browser features. And now with Comet from Perplexity, it can do a bunch of really incredible things. My favorite thing that it can do is what's called RPA, which is where the models operate your browser on your behalf. So you've seen a bunch of examples of this of like, hey, go find me a flight and pay for it, which is interesting. The way I've been using it is in my finances. So I'll go into Robinhood and I'll say, hey, why don't you tell me how my portfolio is performing? Why don't you tell me where I could get stocks that have similar upside at a lower cost basis?
36:44 Speaker 00 What stock should I buy next? Are any of these memes, I mean, you can just go so deep. And look, I could probably figure that out by clicking around the website and downloading the data, but now I don't have to. So this assistant feature in Comet makes every website dramatically more useful.
37:02 Speaker 00 And it's been a big unlock for me.
37:04 Claire Vo? I love this whole episode because you've actually shown a couple of use cases, including talking about personal finances with Comet that really are consumer use cases. Again, as I started at the beginning, we're doing a lot of like, how do you work this inside of an enterprise? How do you write code with it? But I think the real, you know, underappreciated transformation is going to come in consumer experience. I think we're so early. I mean, as somebody who does a podcast, trying to educate people, I just realized we're so early on consumer adoption of AI. And so I have a question for you, which is if you could get, you know, like my mom or, you know, one of my friends that is less, you know, not in Silicon Valley, less in
37:47 Claire Vo? the middle of this, in a room and say, you know, let me show you three things in 15 minutes, there's going to totally change how you think about your life or things that you never knew were possible. What would, what would be those things? What are the consumer side things that you're excited about?
38:01 Speaker 00 So I have kids and parenting is on my mind all the time. And the ways that my kids use models are amazing. So for my four-year-old, ChatGPT reads her a bedtime story, but not just a bedtime story, one that where she can ask infinite questions, you know? So what, what, what was the king's dragon's name? What color was it? Where did it come from? Did it have any kids? You know, she's really into unicorns and alicorns, like tell me a story about an alicorn and a golden egg. And so she can just really interact with the bedtime story. And ChatGPT is far more patient and creative than we usually are. So that's one way. And look, she can't really use a computer otherwise, other than watching YouTube.
38:43 Speaker 00 And then for my son, he'll set up two figures like Sandman and Spider-Man. And then he'll take a photo of them in, in ChatGPT or one of the other models and say, hey, who would win? And then it'll do this whole, you know, oh, Sandman would win in these conditions and Spider-Man, but maybe Spider-Man does this. So they're just, they're able to kind of play with the technology instead of just being broadcast to from technology, which is really new. That's like the near term stuff. I think in the longer term, you know, I think that the models can really help with a lot of social emotional learning. If you look at the classroom, a part of it, of course, is academics, but part of it is just teaching children to be, you know, good people for the world. And a lot of that comes
39:26 Speaker 00 in observing how they're sort of behaving and interacting. And we never had a technology that could do that. If your kid went to a great school, there might be a second teacher in the classroom focused on social emotional. So I think that's how AI shows up in the classroom. It's probably less like homework helpers and assignment generation and more observing the social dynamics in a classroom and helping kids be better people.
39:49 Claire Vo? Yeah. Well, calling back to what we were saying earlier about trying to identify the AI native way of doing things. I watch my children so much. I say that my children form my consumer AI feces for me. Because the other day, my six year old was playing Minecraft and he wanted to know how to do a command. And he literally went to my purse, picked up my Meta AI glasses, put them on and said, hey, Meta, how do I transport to the woodland mansion in Minecraft? And I was like, wait, this is like, it's not type into chat GPT. It's not even ask Alexa. He took this physical device and put it on his face.
40:30 Speaker 00 Amazing.
40:30 Claire Vo? And ask this personal AI a question. And that just really opened my mind to, again, I think multimodal is going to change. I think hardware is going to have a real place to play here. And then this like AI native generation is going to think about accessing information. And building things in a totally, totally different, totally different way. So I am, I'm with you on all of that.
40:56 Speaker 00 I love that. Yeah. And it's, it's interesting because we have been taught what computers can and can't do. But they haven't been taught any of those things. So when I generate an image of, you know, a Harry Potter image for my son, I'm like, wow, did you see how I just generated that? He's like, dad, of course a computer can do that. So they just assume that everything's possible. And now everything kind of is.
41:15 Claire Vo? Oh, my gosh, we had it. As I say, when I had to walk uphill both ways for my internet, like.
41:20 Speaker 00 That's right.
41:21 Claire Vo? Okay, well, we'll get you out of here. One last question I have to ask. You have had such success with generating these complicated assets. But when AI is not listening to you, when it is giving you really poor results, what is your prompting technique to get it back on track?
41:39 Speaker 00 I mean, I don't know if it's a prompting technique, but it's a sort of, it's a, it's a mindset. Two things. One is go with it, you know, like let it take you to some strange, unexpected places and you might be amazed at the results. I, I think the other is just reducing this sunk cost fallacy thing. Where, you know, you create a GitHub branch, you try to do something really ambitious. It's just like falling over, over and over again. Just abandon the branch and start over because you didn't actually do any work. You feel like you did work because it did work, but that's not you doing work. And, and I think being a lot more willing to abandon sort of approaches that aren't working is the sweet spot.
42:17 Claire Vo? I completely agree. Well, thank you so much for showing us all these works, close. It was totally inspiring. I want to get off this podcast so I can go play. So thank you for making my day and I know everybody's going to love the episode.
42:30 Speaker 00 Thank you, Claire. Super fun.
42:32 Claire Vo? Thanks so much for watching. If you enjoyed the show, please like and subscribe here on YouTube or even better, leave us a comment with your thoughts. You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app. Please consider leaving us a rating and review, which will help others find the show. You can see all our episodes and learn more about the show at howiaipod.com. See you next time.
Previous How Amplitude built an internal AI tool that the whole company’s obsessed with (and how you can too) | Wade Chambers Next How to digest 36 weekly podcasts without spending 36 hours listening | Tomasz Tunguz (Theory Ventures) Page updated November 1, 2026 · transcript and summary are machine-generated