Subscribe
Share
Share
Embed
Forward Thinking Founders is a founder podcast where we interview high potential founders from networks like Y Combinator, The Thiel Fellowship, Product Hunt, Twitter, etc. and brings to light what they're building for the world. Think of it like the opposite of How I Built This, where we interview founders before they are successful, then if they are, we have a moment in time we can look back on in the early days. Does the model work? Look at our early interviews and you'll have your answer.
Mat Sherman (00:01.259)
Alright, I think we're recording. We are here on the Riverside FM official recording studio with Vidhi Thate. How are you doing, man?
Vidy Thatte (00:14.604)
man, I'm doing good. Kind of tired. Yeah. Excited to be back on the podcast.
Mat Sherman (00:21.079)
Yeah, you were, I feel like you were early on. I'm gonna just search right here. You were, I, was, dude, you were just after, yeah, January 23rd, 2020. That means that was before the pandemic. That was before we knew the world of.
Vidy Thatte (00:27.97)
Was it Jan 2020 or something?
Vidy Thatte (00:39.156)
yeah, was right before the pandemic. Holy shit, yeah.
Mat Sherman (00:42.552)
I remember for New Year's Eve the night before, not the night, the year before I was in Flagstaff and we celebrated the turn of the century or the decade and we're like, oh, this is gonna be great. And then three years later, Three months later, it is what it is.
Vidy Thatte (00:58.606)
2020.
Vidy Thatte (01:02.197)
Yikes. Yeah.
Mat Sherman (01:07.978)
So what are you cranking on? I feel like when we first chatted on the podcast, you struck me like a builder, someone that is like prolific in the consumer world, and you just like ship like absolute crazy. as of now, like what are you cranking on? What have you cranked on in the last several years? Like maybe if you just wanna like, you know, give everyone an introduction to like.
who you are, what you've cranked on, we can kind of dive into various projects that you've worked on and what you're thinking about.
Vidy Thatte (01:37.922)
Yeah, dude, I mean, it's been an insane journey. So I think when we spoke, I was working on a podcast app. I think back then, like everyone was kind of obsessed with audio and then audio social networks. And I think Clubhouse hadn't started yet. It was like a month or two months before Clubhouse. That's when we did the podcast. And yeah, I was just like working on this like podcast clipping tool.
and then went on to launch that in August or September and then worked on that for a bit and then realized like, like the way we're approaching, we were basically building this tool where you can just clip podcasts, long podcasts into short clips and then like discover new episodes and do all that. But I just felt like we were building on this like RSS infrastructure and it just didn't, I was like, this isn't gonna become like.
like a social network or like something like billions of people use, it felt like a nice little SaaS product. So I was just like, okay, we should try a bunch of different things. And then we started working on a music app where you could just make, it was like a phone studio. So you could just make music on your phone. Did a lot of crazy like stuff there. We built this like real time, you could jump on a call and then make music with someone and then had like a
like a real time sync, like soundboard kind of thing. But I think we worked on that for like about a year. We'd already raised a bunch of money, but we, was like really low. We had raised about like 200K and we were like, we should, we should probably raise a bit more money because this is super hard to build. Maybe we should hire someone. And then the pandemic happened. Me and my co-founder, we flew to Hawaii, like work out of there. Cause we were in New York city. It was really depressing. And then.
Um, think yeah, 2021 we couldn't, we basically could barely could raise any money. I think we raised about 25 K, um, for the music stuff. Cause like, no, we see understood like what, what we're trying to build. Um, so I think, I think around like September, 2021, I was just like, what the hell am I doing? I'm like, you know, like, I'm going to be like 25, 26 soon. Um, and then like, I got to figure it, figure this out. Um,
Vidy Thatte (04:00.321)
I think around September I just went on this crazy, I'm just going to build all the ideas that I have in my notes. And then I think one of the ideas that I, I think I told my co-founder in December of 2020 or something that somebody should build the music app, but for memes, so people can make memes on their phone and then remix memes. I literally put that whole thing in a day. That was the third idea that I was experimenting with.
and then that started like working really well. Like I gave it a bunch of friends, like the next day people started making memes. And then by the end of the week, I think there were like, like somewhere in the tens of thousands of memes were made by like less than 500 people. and that's how.
Mat Sherman (04:43.703)
And what was the name of that? This is the greatest name of a ever. What was the name?
Vidy Thatte (04:47.694)
So it was called collage and then we changed it to Rememe after a year. Yeah, that we started working on and that was like, that was crazy. Like to see something, like working on a bunch of products that were barely working and then this like looking how many people, like people spending like, I don't know, like 16 hours a day, like weekend, like whatever, just on there. It was pretty cool. And then like every...
Mat Sherman (04:54.337)
Beautiful.
Vidy Thatte (05:14.337)
We see hard about us, everyone wanted to give us money, everything changed. then, yeah, so we ended up raising a seed round and then came back to SF, started building this, hired a bunch of people, did the whole startup thing. And then I think two years later, we did a lot of experiments, did a lot of stuff with the meme app, but it just wasn't working after a while. Their attention went down.
It wasn't growing in new cohorts and then we were just like, shit, back to the drawing board again. So actually, I think in Feb 2022 or 2023, ended up letting go a of the big part of the team, kind of scared down to just three of us. And then we were just like, you know what? Iron Masters was super supportive. were just like, yeah, we believe in you guys, it's still a seed stage company.
just try a bunch of experiments to see what works. And we started, we were just like, okay, we're just gonna like make a list of crazy ideas and just build it out and like get into that cadence of like shipping new apps. So the first app we built after that was this water drinking app. It was like Be Real for Water. We called it WAP. But you say you would basically like take a selfie of yourself drinking water with your friends. And then the cool part was you could like spam your friends with notifications asking them to drink water. That was fun.
I think we spent like a week on it, Max. I don't think we spent more than a week. And then we started experimenting with a lot of LLM stuff. So we built this like journaling app where you would start typing and then it would start asking you questions based on what you typed. So was kind of this like, everyone was building a chat bot and we just wanted to like explore some new interface where it wasn't a chat, but it was still there and like had context of what you were saying.
But this was like back in, I don't think GPD 3.5 turbo was out. It was still 3.5 and like, was like super early. Like it wasn't that good. And then we just, yeah, we like work on two other things and then we started like really getting into this like code generation with LLMs. And like, that was super cool. And everyone was building like code generation for like JavaScript and web dev stuff. And we come from like the...
Vidy Thatte (07:38.574)
iOS, iPhone app, like Swift background. So we're like, hey, we should do something for Swift. That would be really cool. And then that's when like Trace happened. I don't know if you saw Trace, but Trace is this thing where you could just, yeah, you could just like type what do you want to make? And then it would just like create the like iPhone, like front end for it using Swift. We worked on that for a while. So we launched that, that went super well. It started making revenue and everything. But I think within like two or three months in, we realized like,
Mat Sherman (07:48.66)
Yeah
Vidy Thatte (08:07.563)
We had, the way we'd set up Trace was it would like write code and then like run in like another server, like build the project. And then it would get like sent to this video streaming thing, which would like stream the actual like iOS simulator so that it rendered on your website, on like the website. And that was like costing us like 3K per month for like a few years who weren't paying that much.
And it was just like, okay, it just felt like this is a cool technology where it's like, it just feels very unsustainable. And also there was this other problem where the LLMs, like this was before Cloud Sonnet, like 3.5. So the LLMs weren't that good. And all the expert devs were like, oh, I don't think I'll use this. And all the beginner devs were like, oh, I feel like I need to know a lot more to use this. So was like this weird like market segment.
where only people like me would use it. I don't know how many people there are like me. so we were just like, okay, maybe, so like there are two, two things we're looking at here. We could build like an expert tool or we could build something for like complete noobs who don't have any, coding experience at all. so, but then they don't really care if it's like, it's like an iPhone app or a native app or a web app. so we just started like exploring like web stuff.
And that's when we started working on web clips, which is basically a way to like make apps on your phone. But these were like web apps. And we spent a lot of time working on it. We'd like build like this like whole new canvas where you could just drop buttons into like, you know, a canvas, like connect that to a backend and stuff like that. And it was pretty cool because like, I think I made like three or four apps, which all went viral on it.
this one app, which, like you would upload a picture of yourself and it would guess your height. and I think I got like 200 K uses, like people use it or something like that. it was crazy. And I was like, man, I just made that in like two minutes. This is like, this is power. but I think one of the things we discovered was that clips was it like, was, we could use it, but it's just like, when we gave it to someone else, it was very confusing. because again, like, I think the LLMs are not there. AI is not there.
Vidy Thatte (10:32.578)
for like complete noob like person who doesn't know anything about coding to like set up a system, make an app. We're still in that phase where if you know how to it'll like accelerate you like crazy. Like you can do anything, you're basically God. But if you don't know how to code, then I think like there's just this barrier still. So we realized that and we're like, man, this sucks. We spent like eight months working on this, but it just like, it just, we just can't get it to a point where.
It's super intuitive for someone who doesn't know how to code. so, and then went back to the drawing board again. Okay. Let's just list a lot. All our ideas, just ship them, see where we can go. And then we launched, so that was in November this year. so we, we launched like monologue, which was like a podcast app. we launched a pocket journal, which is like a journal app. and then now, and then, yeah. And then in like a day I built this like three JS tool, like three apps in December. And then.
Now we're working on this story, like film creation tool, which I'm actually super excited about. Sorry, super long winding. Yeah.
Mat Sherman (11:41.747)
All good. No, that's incredible. mean, we may have gotten into this on the first the first interview. I can't remember so long ago, but you're a prolific builder. the fact you built so many products in the last four years and you've probably been building before we first chatted, obviously, or else we wouldn't have been chatting. Like, when did you first start building stuff? Like, you know, it was this is this newer or have you been like kind of cranking on things since you were like five years old?
Vidy Thatte (12:07.401)
definitely not five years old. I wish I had started five years old. I don't know. I'd probably be more successful, but I started in college, like right before high school and right before freshman year of the summer. I started, I think I started watching all these super cringe Bloomberg documentaries with all these founders. And then I started reading Paul Graham's essays and I was like, yo, I want to start a company. This is so cool. It's such a creative thing to do and like, you
Mat Sherman (12:10.209)
He
Mat Sherman (12:26.326)
Hehehe.
Vidy Thatte (12:36.983)
create an impact on the world. But I figured the best thing to do is learn how to build stuff because then I wouldn't have to rely on anyone to turn my ideas into products. So yeah, this was right after high school.
Mat Sherman (12:54.411)
Yeah, that's awesome. I got into startups my senior year of college and I'm like, always feel like, damn, like if I just would have gotten into this like four years earlier or eight years earlier. But like the thing is like there's versions of the world where I would have gotten into startups four years later or 10 years later. And if it was 10 years later, I would have just be getting in right now. So it's like, there's only like so much we can do about what we know and when to jump into industries and stuff.
Vidy Thatte (13:14.442)
Yeah.
Vidy Thatte (13:20.213)
Yeah, that's true. Can't control all that.
Mat Sherman (13:22.849)
So the question for you about like AI models, it's a selfish question as like a pretty non-technical person. Like I'm good with software, but I'm not a full stack developer. Like when a new model drops, are you able to instantly know like, like this is better than that and here's why, and this is what it implies or like?
Help me understand, like you said, Claude 3.5 saw it and I've heard it's much better for coding than chat GBT. Like, is it obvious when you see the specs or do you just know when you use it? Help me understand that from a technical point of view as someone who's not technical myself.
Vidy Thatte (13:55.864)
Yeah, I think it's a combination of everything. Usually when a new model launches, I'm kind of a late adopter. I never try anything. I'm just like, we'll see. I didn't try a cloud sonnet until like, which is, I don't know if I'll ever spend this long waiting to try a new model anymore, but I don't think I tried it for like two, three months. I don't even know how that happened, but yeah, like I think I'm definitely a late adopter when it comes to things like this. just let other people try it out and see.
you know, what they have to say about it. And then if everyone's like really hyping it up and like, okay, this is it, this is it. I go try it out. I don't think I even like started using cursor, like until like middle of this year. And then once I started using it, I was just like, this is nuts. This is, this is game changing. So yeah, I think, I think usually when a new model comes out for me, at least like, I like they say, this is great for code generation and all that, specs and everything.
But I just like wait it out and see how other people use it and you know, like people play around with it and like come up with ways of using it that that you can adopt into your workflow. Yeah.
Mat Sherman (15:07.315)
like a one-person Apple to feel like Apple they like they let other companies launch new technology and try out things until they're like okay we're ready and then they they watch it usually kill them but the Apple vision pro I mean I think they'll end up winning I think they'll end up winning every market ever because I think it's Apple's Apple but like it's just funny how they just wait and you're kind of doing the same thing for me I just like I don't know if I can
Vidy Thatte (15:09.133)
You
Vidy Thatte (15:16.917)
You
Mat Sherman (15:33.431)
trust the crowds opinion, not because they're wrong by any means, but I feel like with AI, especially being non-technical, so I'm not able to do some individual research on each of these models, everyone just says everything's the best, and this is gonna kill the developers, and this is this, and I'm just like, what is actually, what can I trust? So just, there's individual people on Twitter that I follow that I'm like, okay, I generally trust this person's view.
Vidy Thatte (15:37.856)
Yeah.
Vidy Thatte (15:45.922)
Yeah.
Yeah.
Yeah.
Mat Sherman (15:59.041)
but it's hard when the crowd is saying like, this is the thing. And then for non-tactical things, you there might be, I'm sure that you and I both know companies that everyone was probably shilling that like, I thought secretly that like, this is gonna die very soon, but like, that's not popular for the crowd. And then like they die and then people are like, who would have guessed? So it's like, I don't know. It's hard to know what's gonna work and what's not.
Vidy Thatte (16:18.786)
Yeah.
yeah, there's a lot of hype for sure. do think, I do think like kind of being the adopter and then like trying it out yourself and seeing if it works for you. Then why not use it? Yeah.
Mat Sherman (16:34.61)
That was my experience with even like Claude versus ChachiBT. I use them for like sales prompts. I do sales for Product Hunt. And now I do, know, it's crazy. four years ago, I was just like launching stuff on Product Hunt. Now I'm like helping running it, is lots of fun. But I just like...
Vidy Thatte (16:40.48)
Yeah.
Mat Sherman (16:55.32)
I using quad for a bit and I'm just like, feel like I want to see what else is out there like for like, you know, for giving me different examples of sales prompts. So I've tried chat, she PT. I'm like, huh, this is a little better. Is it like a sub subconscious bias? Cause like I had like Sam Altman and I have like, is he there? Maybe, or maybe it's just like genuinely better, but like, I feel like just trying it for myself. I mean like, like I feel better with these results. I'm going to go with this versus just saying, you know, reading like, like quads better for this and chat, she PT is better for that. It's like hard, hard to follow that.
Vidy Thatte (17:24.897)
Yeah, no, I agree. I've noticed that Clark is actually pretty good for like writing as well. But if I have like anything.
Mat Sherman (17:29.919)
Hmm. Like how so? Like long form or like individual pose or what do you mean?
Vidy Thatte (17:34.766)
Yeah, think anything, anything related to writing, even if I have to craft some kind of prompts for image generation or something, I just use Cloud. But when it comes to, feel like just general, I just want to ask a question about something. And I think Cloud feels a lot more restrictive. There are things that it'll be like, oh, I can't say this. But ChatGPT, it feels a lot more free to me personally.
Mat Sherman (18:00.727)
I feel like that's a feature someone think that's a feature someone think that's about because Claude is like they're you know, Anthropic built Claude and Anthropic is like what they're trying to be like the ethical AI the AI that's like, you know gonna be good for the long term in the future of the world versus like being the AI that just is destruction and it'll be like like I do think that Sam Albin is like slightly turned into like a little bit of like a more villainy type person than he was like five years ago not to was not to like not even based on anything he did or said but I just think
that's the way things are rolling. it could be what he's done. He's trying to turn the thing into a for-profit screwing on us. I don't really know what went down. Of course, none of us do. But it's interesting to see which model, not in five years, but in 20 years, will be the one that we all use. Or will we use many? Do you have a thought on that? Is there going to be one model that reigns supreme, or is it going to be a network of various models for different use cases?
Vidy Thatte (18:41.069)
Yeah.
Vidy Thatte (18:56.278)
I feel like we're gonna use a bunch of different models. I mean, that's what I'm doing right now. And I think there's gonna be more and more models coming out and any task we have to do, we just pick like the best one for that task.
Mat Sherman (19:09.441)
Totally, yeah, it's gonna be interesting and I'm also excited for when it gets to the level where a slightly technical person, but not super technical, can use it. I think I'm gonna just kinda go crazy. Maybe that won't be good for my attention span, but we'll see. So I got a couple questions for you about the journey. If you're open to sharing, if you're not open to sharing anything, just let me know. like...
Vidy Thatte (19:24.781)
you
Mat Sherman (19:35.364)
So I feel like a lot of founders, myself included, like, you know, I had a vision for my last company, Seed Scout. You know, I raised some money for Seed Scout. I pushed hard, hard, hard. I did a lot of experiments on kind of that entity. And then unfortunately, last month I decided to wind it down for various reasons. Yeah, it is what it is, but I think it will allow me to, I think I just had to make the decision or else it just like...
Vidy Thatte (19:50.573)
That's so hard.
Mat Sherman (20:00.215)
I think it was something I had to do and I'm glad that I did it and I'm looking forward to having a stable life for a little bit with my new family and whatnot and trying to get it in a few years. But my question for you is, have you spun up just a bunch of different entities as you've tried things? Has it always been the same entity? How do you, again, you can share as much or a little as you want, but how do you manage investor comms as you're trying all these things?
Or is it kind of easy? just try on the same entity, tell them something worked, something didn't, and you move on to the next idea. How does that work?
Vidy Thatte (20:32.622)
Yeah, mean, yeah, it's been the same entity since like the last time we spoke. And I don't know what the right thing to do is if you should be spinning new entities or not. But I definitely wanna like, I mean, like we've had the same investors since, since like, you know, like the same group of investors. And I think if I start a new entity, I would have to do like the fundraising process again and like maybe get a new team, whatever, which I don't know, it just feels like a lot of work.
I think with the investors, the expectation going in was pretty much like, yeah, this might not work. We might try a bunch of different ideas. you're essentially like batting on the team, at the end of the day. So. Yeah. I mean, and think like they've seen enough companies like pivot like a million times until they find something good. and I hear all these stories too, like, so it's like not that bad. Maybe our case we've pivoted way too much. don't know, but that's, that's what it feels like right now. But,
Yeah, think it hasn't been like a big issue in any way.
Mat Sherman (21:37.816)
the same experience. I feel like we probably tried six to eight various iterations, targeting all around the same thing, kind of solve the same problem, generally one to two iterations away of what we did before we pivoted. And I think what I landed on for Seed Scout is like...
Vidy Thatte (21:42.123)
Yeah, okay.
Mat Sherman (22:01.633)
There's no world where this target market is gonna yield like a big company through revenue, through like a fund model. I don't know if I'm gonna start a fund, but I'm just trying to sell these founders and emerging managers on these products. just like, what am I doing? This is not what I'm trying to do. And it's not even my investors weren't even pushing me. You gotta keep going. They're just like, yo, you've like, I mean, keep going if you wanna keep going, but like.
Vidy Thatte (22:12.043)
Yeah.
Vidy Thatte (22:20.043)
Yeah.
Vidy Thatte (22:31.03)
Yeah.
Mat Sherman (22:31.383)
You you just got a kid and I think me having a kid was a pretty big catalyst. If I didn't have that happen, I'd probably still be going, which maybe would be still the wrong thing, maybe the right thing, you never know. No one has a compass for all of this.
Vidy Thatte (22:44.236)
Yeah.
Yeah, yeah, I don't know how people can tell. I guess you can't. just wing it.
Mat Sherman (22:51.0)
You can't, you just try to give yourself a lot of swings and hopefully one hits and that's really all takes, it takes one. So are you, are you like consumed pretty much 100 % with this work? Is there anything that you're doing outside of work? you have any time for play, any time for like, know, anything else? you kind of all in on this right now? Or I guess in other words, do you have any like hobbies to take your mind off like work when you're not working?
Vidy Thatte (23:20.545)
I think the usual, I don't know, hang out with family, hang out with my girlfriend. Yeah, that's pretty much it. I feel like this is like 90 % of my attention. I think at this point, it's gotten more and more intense where I'm just like, I just have to figure this out and I just can't sit still, do anything else.
I'm just like working and trying to try to you know spend every single like little Time that I have pouring everything into this but I do I do like work out, you know and and like make sure like I don't Neglect those things But yeah, pretty much pretty much just work
Mat Sherman (24:05.879)
Yeah, it's kind of what you gotta do when you're pre-product market fit and then when you're post-product market fit, you still have to do that because you have to keep up with the attention and the demand. just, you you get to vacation post-exit or post-failure. Hopefully it's an exit.
Vidy Thatte (24:11.795)
Exactly. Yeah.
Vidy Thatte (24:20.973)
Yeah, exactly, think like time, I don't know, I just feel like time is super valuable and you know, it's like what you said, you just want to have optimized for like shots on goal and just like try as many things as possible as long as you can possibly can so you can increase your chances. So I think like I'm in that mindset. I don't know if that's necessarily a good thing or a bad thing. Yeah.
Mat Sherman (24:50.776)
So what would you say of what you're able to share? What are you optimizing for right now? What are you cranking on? What are you excited about? I know you may not be able to share everything, but tell us what is the current version of the next big thing look like for you?
Vidy Thatte (25:05.645)
Yeah, no, can talk about it. So it just doesn't have a name and it's like two weeks old. So like a lot of things in the air. But one of the things that I got super excited about was I think I was going through this, I was going through TikTok the other day and then I saw this like person had made a whole like sci-fi like story with AI generated images and like captions and voiceover.
And I was just like, whoa, this is so cool. Like I really want to know what happens next after watching that whole thing. And then I saw comments and everyone was like, this is the best thing I've seen. Like, you know, we, this needs to be a series. Like, you know, I just followed you or whatever. And I was just like, wow, wait, I think that one thing that I've ignored, like I've been so focused on like LLMs and stuff. There's this whole new revolution happening with.
with diffusion models and stable diffusion or video generation. Sora came out, this was literally, I think, a week before Sora came out. if you go to YouTube, if you go to TikTok, there's just so many people making AI short films and stuff. And there's this whole industry that's popping up. And the models have gotten so much better compared to last year.
It's crazy. think the rate of progress is insane. And it's only gonna get better from here. So I was just like, oh, this is so cool. If I were to make, if I were to make my own film, I have to go through so many things. to generate, come up with a script, generate images, make sure they're all consistent, apply, train my own loras, create these characters so that they're consistent, and then use, and then.
convert everything into videos, scenes, and then put all that together in a video editor and add sound effects, audio, whatever. And it was just like a 20-step, 30-step process to get to something really nice. And I was like, yeah, maybe I can build a tool here and make this whole thing really simple where you have all these models in one place, you have all these editing tools in one place.
Vidy Thatte (27:24.043)
and you can just make a film using AI. So that was the whole idea. And then, think I'll announce something next week about it.
Mat Sherman (27:41.336)
by everything you just said and really the implications on lowering the friction by a trillion percent to create things, whether it be movies or creative for company. My question always goes to what are the implications? If everyone can make a movie, where does the puck go?
I think those questions drive me mad because there's never an answer. Once the puck is there, I'm thinking about the next thing. But do you think about that at all? Or do just think about, let's get this tech right and let's build this? How do you think about where the puck is going versus where you are now?
Vidy Thatte (28:21.005)
I love thinking about where the puck is going, but I've also realized that doesn't really matter at all. Just get it out there. But I love thinking about it all the time. Yeah. I was rewatching the Dark Knight movie. I don't know if you remember the opening scene where they rob the bank. And I was watching it I was like, man, I feel like one day we're going to look at this.
Mat Sherman (28:27.083)
Yep, yeah. It's fun to think about, but it's not super productive.
Vidy Thatte (28:48.268)
I think this is controversial. If someone listens to this, they'll be like, what the hell? But I feel like in a few years, we're going to look back and be like, I can't believe we used actual human beings as human beings as actors. I feel like it's going to get to that stage like really fast.
Mat Sherman (29:05.911)
Do you think, it's one thing I pontificate on, in a not helpful way, but I still think about it is like one of the benefits of humans is like there's humans that we follow. Like if Tom Cruise isn't a movie, you know, I'm gonna watch it. Or like James Bond, like when there's a new James Bond, if it's an actor, we all like, we're gonna watch it, et cetera. Like, do you think that, again, this is like going into the thing that's like less, more fun to think about than, but like not necessarily useful, but I do like thinking about it.
Do we care about AI characters like we do about human characters or do we care in a different way? Do we not care at all? If everything's AI driven and actors kind of like go away, which I don't think that's a far-fetched idea by the way, do we just build attachments to these AIs or what do you think?
Vidy Thatte (29:51.435)
Yeah.
Vidy Thatte (29:56.206)
I mean, whether or not it's AI, it's already happening. we want like Toy Story, like we all know the character, like Buzz Lightyear, Tom and Jerry, all the Disney princesses and stuff like that. don't know. So we already have, we already have, I think it's all about storytelling and world building at the end of the day. And the AI is just a tool to get that done. So I feel like it's already happening.
Mat Sherman (30:01.175)
Mm.
That's true.
Mat Sherman (30:20.823)
That's a very good point. That's kind of like when people say, we don't want to do these VR classes. And then people say, it's already happened. Everyone has their screen in front of them, everyone in the whole freaking world. It's like that, but even more so, right? Because animated movies have been, it's essentially what we're talking about. Whether or not it's created by a human in the back or not, we wouldn't know. Super interesting. Is there anything, before we wrap up,
Vidy Thatte (30:31.437)
Yeah.
Vidy Thatte (30:39.275)
Yeah. Yeah.
Vidy Thatte (30:44.929)
Yeah, yeah, yeah.
Mat Sherman (30:50.827)
anything just two categories like anything you're working on now or thinking about or you know something that's on your mind that you want to share or or that you want to promote or something that is just been like on your mind during the interview that you want to bring up or any last words before we kind of go into closing closing section.
Vidy Thatte (31:10.982)
I think one of the things that I've realized this whole year is...
I feel like if you know the basics of building stuff like programming and like what it takes to make an app, I think AI is just gonna accelerate that like massively. And you can just keep launching new products every week if you wanted to. I think you can get to that stage. That's what I feel like anyway. So I feel like it is worth learning how to code and learning how to build stuff, getting the basics right. So you can actually kind of ride this wave
and launch as many products as possible until AGI happens and none of it matters.
Mat Sherman (31:54.93)
If someone's listening to this or even for myself, they're like, great, it's 2025. I want to learn to code. How would you suggest someone learn to code in 2025? Or is it the same answer as it always was? Just learn the basics. Like, what would you say this year if someone wanted to get into this stuff?
Vidy Thatte (32:11.093)
Yeah, mean, I think, well, when it comes to basics, you don't have to go take a Python class at college or anything. I think what you need to do is build something and learn everything that goes around how to build it. So like, how do I build the UI? How do I set up a database? How do I connect everything? How do I deploy it and give the link to a friend so they can use it? I think you should try to get through that whole stage.
And I think it's so much more easier now. Like if you get stuck, you can just ask, right? versus back in the day, like we used to like Google and then search Stack Overflow and like go to that whole process, which was like now looking back, it's like, no way I can't, I can't believe I used to do all that. But, yeah, so it's, it's much easier to learn as well.
Mat Sherman (33:00.885)
And then one more question on that front, would you suggest that someone uses one of these AI forward tools to learn a V0 or a cursor, and then they learn around that, just using the tools that are at their disposal?
Vidy Thatte (33:17.325)
Yeah, mean, those are pretty intuitive. If you know how to use chat GPT, you can just use v0. So just get started with v0.
Mat Sherman (33:26.849)
Yeah, I've always convinced myself that I like the social capital and like the people element of business in the world and the sales side and less technical. And my head is going towards as the barrier to code is going lower and lower.
Where I'm going is like is it time to finally learn or is it actually time to triple or quadruple down on the on the people side on like the person to person side and that's not like there's a right answer like what I mean I don't have a company anymore, so it's not like there's like An outcome I'm going for like like obviously like I want to make a lot of money and help a lot of people and have a lot of respect but like there's a million ways to get there, but it's something I've been pondering and it's like Do I do I finally get technical or because everyone wants to get technical do I do I?
when everyone zags and it's such a useless thing to think about but I do think about it like all the time. I don't know.
Vidy Thatte (34:20.641)
I mean, yeah, you're right. I don't think there's a right answer. I do feel like it's good to lean into your strengths and like just instead of, you know, trying to make up for your weaknesses. So if you think people is a thing, go all in on it, I guess.
Mat Sherman (34:32.124)
I honestly just want to like interview. I can't now because I have a job which is fine, but like like I just want to get to a point where me or an AI version of myself maybe I don't know like interviews like 5,000 people a year or something and I just like it's just like this collection it's like you listen to my podcast because it's like a collection of interesting people and interesting conversations and access but that's I don't know something to figure out in the future. Well anyways, yeah.
Vidy Thatte (34:55.81)
I mean, that was the whole idea with monologue where you imagine you had a podcast episode.
Mat Sherman (34:59.16)
Wait, was that the one where I have AI interviewing me and then it gives me like the trans? Dude, that was wild. Are you still doing that?
Vidy Thatte (35:03.307)
Yeah. Yeah. Yeah. It's live. You can use it. Yeah.
Mat Sherman (35:09.879)
Yeah, that's like, is it gonna be live for the foreseeable future? Like if I start using it, will it be up or, cool.
Vidy Thatte (35:17.941)
It should be, yeah. Yeah, no plans on shutting it down. There is one annoying issue with it where sometimes the video doesn't get saved, so if you run into that issue, let me know.
Mat Sherman (35:27.317)
Yeah, I used that once for people that are listening essentially correct me if I'm wrong, but from the experience that I remember is like I pretty much said here's what I want to do a podcast on. Here's the questions I want this AI to ask me and then it just asked me and then I answered it asked me questions. I answered them and then in real time it like answers in like a pretty decent conversational way and then ask the next question.
and essentially it allows me to just talk about things I want to talk about without me just doing a monologue, which is hilarious because that's like the name of the thing. It's honestly a genius idea. It really is. Because sometimes I just want to share opinions, but sometimes people don't want to hear them. But if it's like an interview, it's like, oh, all this new interview, like two people, even though it's not two people.
Vidy Thatte (36:09.771)
Yeah, yeah, it feels so much more authentic to it's like, you're just talking. It's not some scripted thing.
Mat Sherman (36:14.839)
Yeah, I really think there's something there like I don't know like where the money comes from like I don't know but like but like I also do if someone gets you is using it over and over again and they're like that's where they're generating clips from like you can probably get some money for them but like is there a lot of money there like I don't know I haven't thought about it but it was an interesting product when I used it. Yeah, well cool. Well, it's a good having you back on the pod if you have any
Vidy Thatte (36:19.372)
Yeah.
Vidy Thatte (36:25.377)
Mm-hmm.
Vidy Thatte (36:31.329)
Yeah. Yup. Yup.
Mat Sherman (36:38.903)
or Twitter or I guess X or any link that you want to shill or product you want to share before we call it a day.
Vidy Thatte (36:47.457)
Don't have any product yet, but yeah, you can just follow me on Twitter, VDTHATT, last name, yeah.
Mat Sherman (36:54.357)
I feel like you're pretty out there. When you ship something new, you pretty much always share it on Twitter, don't you? So if you're trying to keep up with this prolific shipper, you've got to find them on the Twitter or the X. Well, it's good to have you on round two of the pod. And I'm looking forward to seeing this next product that you're alluding to launch and trying it out.
Vidy Thatte (37:00.17)
Yeah, all this. Yeah.
Vidy Thatte (37:16.363)
Awesome.
Mat Sherman (37:17.976)
All right, well, that's a wrap.