Creative Flux

Pierson Marks (00:00.604)
kicking things off on episode four. It's pretty exciting.

Bilal Tahir (00:05.858)
Yes, it has. Can't believe it. Time flies.

Pierson Marks (00:07.898)
So I know time flies and it's recording this on Thursday instead of Friday like normal because tomorrow's the 4th of July. you know, we're like happy 4th, happy 4th. Yeah, well, if it's the first time here, welcome to...

Bilal Tahir (00:17.806)
Yeah. Happy forth to everyone.

Pierson Marks (00:26.01)
Episode four of Creative Flux, the podcast where we dive into how we're actually using AI, focused on creativity, media, gen, and workflows that are, you know, working for us. I'm Pearson Marks.

Bilal Tahir (00:39.438)
and I'm Bilal Tahir.

Pierson Marks (00:41.162)
And yeah, and it's just a conversation. you know, well, I think it's been fun so far. We're four episode, this will be the fourth episode. We've talked about, what, the first episode we talked about, was that ImageGen? Was that number one? I forget now. VO3.

Bilal Tahir (00:57.934)
Yeah, Viewer 3 was the first one. And I think the second one was 11 Labs Text-to-Speech. And third one, last week we talked a little bit about image editing and what else? feel like it was random. We did a lot of random post-AGI and physical philosophical musings, I feel.

Pierson Marks (01:17.424)
Right. It was kind of cool. were talking, talking about like post-reality. actually brought that up last on, on Tuesday. So I had this like a conversation with these, group called the AI salon and it was pretty fun. They're all like old, later in their career, but they're really embracing AI. It's really cool to see because, their executives or they're just enthusiasts that are really

passionate about this space. And it's not like young people that, you you think about new technology, think a lot about young people. And I'm not saying it's like a bad thing at all. It's like really cool to see an organization that's catering towards more of like people that are more experienced in their careers that just like, well, this stuff is amazing. I'm using it for like, for fun, or I'm using it to like help me accomplish more things in the day. Um, but, uh, yeah, I mean, uh, I brought up this whole thing where

Bilal Tahir (01:45.197)
That's awesome.

Pierson Marks (02:13.893)
we were talking about post-reality image gen. Like when the videos and when the images get to the point of being like perfect, everything we generate could be perfect. Like what happens then, you know?

Bilal Tahir (02:26.99)
Yeah, it's so interesting. And by the way, I love the name AI Salon. It gave me a picture of you just sitting there, like getting a manicure with humanoids and just talking about AI. love it. But no, it's so interesting, kind of going off that. I was watching this talk by Fai-Fai Li, who's one of them.

Pierson Marks (02:35.249)
Right. Right.

Bilal Tahir (02:45.582)
know, OGs, you know, in the AI space, you know, she was a Stanford professor and her lab made ImageNet and then ultimately AlexNet used that ImageNet data set, you know, to win the 2012 competition. And she is like big on spatial intelligence. She thinks that's the next frontier, but she was talking about how like...

Pierson Marks (02:47.066)
Right.

Pierson Marks (02:54.886)
Right.

Bilal Tahir (03:05.496)
what we think of reality, it's actually like 2D projection. We take these inputs and our eyes actually make a 2D projection of that. So even what we think is real is actually not real. It's a fascinating Plato's cave kind of a experiment where, know, because we just assume whatever is around us is the baseline reality, but what is reality? It could be there's so much going on around us, but our eyes are just not enough to...

take that signal. so maybe AI can actually process a lot more information and get a richer picture of what's actually a reality.

Pierson Marks (03:43.378)
Right. I mean, it's like that thing where is my blue the same blue that you see? I mean, I don't know. And is it like, it kind of combines all these really interesting aspects of communication.

Bilal Tahir (03:48.588)
Right, right.

Pierson Marks (03:57.458)
because language is limited in that sense where you can't really communicate exactly what you see. Like I can't explain to you, like what is this color that I see on my plant right here? And it's green, it's a dark shade of green and...

could I explain that thing to you perfectly and you were able to always capture the right sense of green? Like I could reference to, it looks like an olive or it looks like a palm tree leaf. And then you have your interpretation of what that is. But it's like a limit in communication ability that like prevents us from actually seeing the same thing. So.

Bilal Tahir (04:23.534)
Mm-hmm.

Bilal Tahir (04:37.186)
Right, yeah. No, it's super fascinating. Like, and what happens when that, what do call it, the Tower of Babel, you know, where, you know, invented languages and created this friction between us, you know, that goes away and we actually do, can communicate seamlessly, you know, what happens. It's like, what do you call it? A hive, we're like almost become a hive mind where, you know, everyone's thought just can be instantly transported to each other.

Pierson Marks (04:56.304)
Right.

Pierson Marks (05:01.073)
Totally. No, it's such an interesting, I've had this conversation and just thoughts just, you you go down the rabbit holes of, okay, yeah, what if we could instantly communicate and if I could just transmit via my brain to your brain and then there's no more like issues in communication. We just kind of have context. And that's kind of crazy. Like, I mean, today Cursor just released their 1.2 update, which

general like GA memories and improves a context awareness and doing like a better sort of look up for your code. And eventually I mean like.

This is what they're working on and what they're doing, but our entire team is going to have like, Pearson writes code this way. Buala writes code this way. Team wants to write code this way because of the comments on those PRs. And how can we like write code through cursor that like adheres to kind of a, the standards from the team. And it's kind of like, it knows.

Bilal Tahir (05:42.862)
Thank

Pierson Marks (06:04.705)
all this context and kind of takes you as a unique individual contributing and then puts it through this like filter sort of like, I want to do this thing and cursor and AI will take it and then, you know, adhere to all those guidelines has the context of like how code has been written in the past and actually ultimately produces something that is higher quality without the back and forth. Like you giving me a feedback. Hey, why did you do this way? We do it this other way, you know, so.

Bilal Tahir (06:30.478)
Right, Yeah, no one's asking. Yeah. Well, hopefully.

Pierson Marks (06:34.767)
And for anyone listening that doesn't know what cursor is, it's a code editor. So with AI sprinkled in everywhere, it's really cool.

Bilal Tahir (06:41.1)
Right, yeah. Yeah, it keeps getting better. I'm excited about using the latest feature. I think they have like to-do lists now and PR, like you can get a PR review, et cetera. So, yeah.

Pierson Marks (06:45.275)
But.

Pierson Marks (06:53.103)
And they stole the Cloud Code PMs, right? Or the engineers on Cloud Code.

Bilal Tahir (06:58.367)
yeah, I saw that. was it. Wait, no, thought plot stole cursors. What was it?

Pierson Marks (07:04.697)
No, think no, Kerscher stole Anthropix, Yeah, and everybody.

Bilal Tahir (07:07.214)
It's all of us right now. Speaking of that, maybe this is a good point to jump into the probably the spiciest news of the week. A guy named Soham Prakash. Yeah, apparently, I don't even know how to set this one up because I guess it started with this tweet from this guy called Sohail, who I think is a former, he did, what was that company?

Pierson Marks (07:17.593)
Yes, let's explain this.

Bilal Tahir (07:37.666)
that analytics company, I forgot his name, Meta, Meta something, Mixpanel, Mixpanel. Yeah, I think that was his, yeah, he went to IC with Mixpanel, they had a good exit or whatever. I guess, he was a first time founder, or at least that was his success. And then he moved to...

Pierson Marks (07:44.047)
Mix panel.

Bilal Tahir (07:58.21)
I think Mighty computers and stuff, doesn't really talk about that anymore, but which kind of flopped. I thought it was an interesting idea. It was basically a cloud in the computers, where you can have basically a cloud server running your OS, and then you access it. And his whole thing was like, you don't need to upgrade hardware and stuff. You can just access the latest cloud instances and get the software upgrade.

Yeah, maybe it was too early in its time because of the latency issues and stuff. Google Stadia was also a thing during this time when everyone's thinking it was going to be over the cloud, but just didn't pan out for any of those. And then he moved to Playground AI, think. Mostly that's what he's doing, which I don't know where it's at right now, but it's basically you can, he started with stable diffusion, but now they host a bunch of models. TLDR was what happened was he put out this mess tweet that said that he was hiring this guy and

Pierson Marks (08:29.455)
Alright.

Pierson Marks (08:44.345)
Right, right.

Bilal Tahir (08:52.814)
apparently he was a scammer and what he does is like he will apply to you and not do any work but he also simultaneously working like a dozen, like half a dozen other jobs. And this concept is called over employed actually. I actually know about this since COVID because it became a thing when COVID happened because remote work became a thing and it was possible for people to get multiple jobs at once. And so they would just do that and...

There's a spectrum. There's like one side is like you get the job, you don't do anything and you get the pay and you you move on to the next job and it's basically a scam. That's one end. On the other hand, I mean, actually people are like, you know, I can do this job in 10 hours instead of 40. So what am I doing for the other 30? And so the people who are multiple jobs and you know, they actually get promoted and stuff, know, so I've like, so it really depends on who you are as a person. It's a murky thing. You know, some people actually.

Pierson Marks (09:30.16)
Right.

Bilal Tahir (09:47.598)
do the job and sometimes even their employers know. Other than it's more of a scam. In this case, this Soham Parkeesh guy apparently was definitely a scam. In fact, he took it further where apparently what he does is it's not even him. So what he does is he's really good at interviewing with lead code or whatever. He goes, he does an interview, passes an interview and then he just gives access to what he has a dev shop running in India and he's from India. And he, one of his junior guys like takes over and he'll do like one PR every three weeks or something just to try to do the bare minimum and try.

Pierson Marks (09:48.944)
Right.

Pierson Marks (10:12.558)
You

Bilal Tahir (10:17.552)
to keep the paychecks coming until like it's over. So anyway so so Hale puts out a tweet like this guy beware of this guy he's been doing this for a year I told him to stop he's not and then it just exploded and suddenly people were DMing him and pre-tweeting he's like shit I hired this guy last week my god yes I hired this guy or we interviewed this guy and it became a thing and it was just so funny because on the one hand obviously what this guy did you know he's a scammer this is a grift I don't

support at all but the fact that so many Silicon Valley startups just hired this guy or you know just had this a process that kind of hire for this guy is kind of a tell you know that there's this monoculture of you know requirements of these companies who supposedly are supposed to be out of the box thinkers and are trying to disrupt the world so I just found that funny

Pierson Marks (11:09.841)
It's wild. I mean, I have such a strong opinion on interviewing process. It'll be interesting as we grow at Jelly Pod to really create a process that hopefully these types of people never even get past because like, like, code. mean, you could hack the leak code. If your whole gig is like just passing interviews, you're going out leak code. You've seen every single problem.

Bilal Tahir (11:27.81)
Right.

Bilal Tahir (11:34.477)
Right.

Pierson Marks (11:38.306)
There's just so many like really especially with the age of AI like, you know being able to communicate design sort of like why you do this What like why do you do it this way? Like how would the user think about you know, and I have a whole list of interview questions that kind of You know, most engineers are just engineers wouldn't really be able to think about

Bilal Tahir (11:48.622)
Thank

Pierson Marks (11:59.984)
because it's like less of engineering, which is expectation, more of like design and more of like product and, you know, something that I think we're both pretty decent at.

Bilal Tahir (12:07.778)
Yeah, yeah. And I think even before just going that we like the Silicon Valley index is so much on lead. Well, basically, if you have an X fang label or late code and the resume, like he put up his resume, it was such a joke. And I'm like, you know, one great way to avoid that is just before you even

Pierson Marks (12:21.359)
Right.

Bilal Tahir (12:26.414)
get to the interview, filter for that by looking at their GitHub or what kind of projects they do, et cetera. And I feel like in the age of AI, you can be actually be very sophisticated about it. You can literally scrape the profiles, go through the commits and actually see if, know, it wasn't like he just worked like, you know, whatever, like Kubernetes and just made a read me update or whatever commit like, you know, cause that's a scam to people who are going to be contributors on big projects. You know, I've seen that happen, but there's, have tooling now to kind of look.

Pierson Marks (12:46.993)
Alright.

Bilal Tahir (12:54.804)
and get a smart signal, at least on that end for people who are active, who have open source projects, et cetera, who contribute on GitHub and social media, you can do that. does that, that's an incomplete solution because a lot of people, you know, they work for Google or they're not.

they won't have time for open source projects. And so I totally understand if your GitHub, just because your GitHub is empty doesn't mean you're not coding or you don't have the skillset. I'm just saying, but that's one way you can actually find good people that I feel like is so underused, utilized. Like even today, I mean, feel like today, if you look at it on Twitter and stuff, the...

Pierson Marks (13:24.913)
Totally.

Pierson Marks (13:28.529)
Totally, no, I absolutely agree.

Bilal Tahir (13:33.324)
The best way startups find is like, they're like DM me, you know, or who do I need to hire product guy, marketing guy, who are, and people will like recommend people. mean, this, especially in the age of this automation where there's so many bots, this word of mouth, you know, recommendation has just become probably one of the strongest signal of how to get good people. So very underutilized.

Pierson Marks (13:51.972)
100%. I mean, I have so many friends that I think so. I studied computer science in undergrad and then I graduated in.

Uh, five years ago, and then I worked for Amazon for a while and a few of my friends that were younger than me or decided to change career paths went to like the CS masters route. Like there were engineering smart people and then they got like a master's from like somewhere in computer science. And some people still ask me today, like friends and younger siblings will be like, Hey, like what do you, what's your thoughts here and everything. And like, you can go do that cool. Like don't expect that to get you a job because I'm like, Hey, if you, job market's hard.

right now for engineers as well. I think there's like a small pool that are getting the 100 million dollars obviously from Meta and like that's crazy and then there's like the start of people and they have word of mouth but like the entry level market for just like a baseline engineer is the rug's been pulled and I'm telling every one of my friends like you know can code like it's like you have to prove it and how you gonna prove it it's like hey find a open source project that you enjoy or like spend a week.

It just find one to find a project that like maybe there's some issues that are out there or like Claude code. That's a great like if you if you're interested at all in AI right now.

Claude code is an open source SDK that is accepting, like has a bunch of issues. It's accepting PR, it's actively maintained. Like how much cooler would it be? Like, hey, like I built like a significant portion of that, that Claude code. It's like, you know, 500 lines of code, a thousand lines of code, add this new feature, do it. Like.

Bilal Tahir (15:30.68)
Yeah, yeah. Yeah, and you got to talk about it. Like it's not enough. Like it's hard, especially, mean, as someone who, you know, it's been hard really getting into the marketing distribution game. have to, you know, if you're, you know, obviously, you know, if you do videos at YouTube and stuff, that's great. But if you, even if you're camera shy, you need to be posting about it on LinkedIn, whatever your social like, Hey, I did this, you know, check it out. I know it sounds cringy, you know, maybe you're showing, but honestly it's

Gregory and Glen Ross out there. Always be closing, always be selling. You kind of have to sell yourself. But the good news is most people don't. It's literally a cheat code to getting the opportunity you want in life. Just put yourself out there.

Pierson Marks (16:07.174)
Mm-hmm.

Pierson Marks (16:14.001)
All right, get your, like...

by your domain name. This is something I did a long time ago. I have Pearsonworks.com and it's like, yes, I have Pearsonworks.com and I've kind of like over the years played around with it. It's like right now it's a basic blog with some of my writing. I'm actually gonna rip that down and put something else up with like Creative Flux, like put the podcast there and then also like put writings and then, you know, do all the above and make it a little bit better. one, it's like, it's your authority in the world.

Bilal Tahir (16:17.742)
Yeah. nice. Wow.

Bilal Tahir (16:33.207)
Hmm.

Nice.

Pierson Marks (16:45.989)
Especially when AI becomes like a major contributor to everything on the internet Like who are you and what's like? What's your social proof? It's like hey, this is me This is what I write. I hear my thoughts like my uniquely human thoughts and I'm putting them on the internet Maybe one person maybe zero people ever come across it. Maybe one person does Eventually they get sucked up into the next iteration of training data of all these new models and then you know, it's like

Bilal Tahir (16:50.924)
Right.

Pierson Marks (17:15.859)
Hey, Pearson Marks, he's the CEO of JellyPod and they work on AI podcasting. it like, I think is a huge opportunity here for people to just put their thoughts out on the internet and it's hard because you have to sit down and actually write. And most people don't like writing or like reading, like what we're doing here. mean, it's easy to have a conversation for sure. Easier than writing, I think, but I don't know.

Bilal Tahir (17:41.582)
Yeah, no, for sure. It's about, I think the world is gonna be, it's gonna be about agency is gonna become so important. It's always been important, but it's like, maybe it was 10 to one before it'll become 100 or 1000 to one where the person who just takes action is just gonna get so much more outsized results versus a person who doesn't. And that delta is just gonna be insane, right? So.

Pierson Marks (18:07.826)
Totally, this is pretty, no, for sure. I mean, and this relates to kind of like this whole podcast and this conversation here about Creative Flux and kind of what we were trying to do is we'll talk about the news and images and video and audio and all these cool tools. We'll talk about how we're using them. And this relates because...

Bilal Tahir (18:08.812)
So you want to be on the right side of that.

Pierson Marks (18:30.61)
A lot of people have this like really niche domain knowledge. mean, I think about it. You go, you go to college, you have a, you get a bachelor's degree, master's, a PhD in some niche content and you're going deep, but you never really spread outside of that. But while going deep, you know, you start a little bit, like it's like a funnel. go like broad, you have your general eds, your master's, your PhD all the way down. And at the very bottom, like you're like one of, if you're a bachelor's, you're

one of a few hundred thousand people, million people, whatever, then master's even fewer, PhD even fewer. But you have such domain depth in something. But now with AI, you can start to spread. You've one, been able to think about problems logically, solve complicated, unsolved things. And you can start to ask better questions and start to spread out your depth. It got narrower and narrower, but AI

starts to grow that and start to make it easier to connect the dots between things because you don't need to know necessarily the right question you just need to be in the kind of right direction and AI kind of will help you like like hey if you're dealing with I don't know some weird protein folding and you're dealing in medicine you're a smart person but you're so like in your bubble and AI kind of helps you get out of that bubble and that's really cool as long as you're using it and actually kind of you know playing around.

Bilal Tahir (19:53.774)
rain.

Pierson Marks (20:00.38)
with it and exploring your creativity and exploring what you're interested in outside of what like your day-to-day job is.

Bilal Tahir (20:05.176)
Yeah, absolutely. Yeah. There's something so magical about the cross-domain. I think that's one of the things I think people are understand like AI, the way AI is going to help us. Like you said, most people are in this one field or maybe two, but AI will connect the chemistry problem to the physics problem, to the bio. Like there's all these moments. Like you hear all the time about some discovery happened, but it was actually in the biology department 30 years ago. But the guy who had the physics.

expertise needed to see it and that conversation didn't happen until this weird conference where they bumped into each other and they're like, you wrote a paper about that? That's exactly what I wanted, right? And I feel like that's a lot of scientific discovery and stuff is gonna come out from that, you know, cause AI will be able to connect those dots, you know, across fields and help you, you know, if you're an expert in one field to broaden your horizon. So very exciting times for sure.

Pierson Marks (20:50.867)
absolutely.

Pierson Marks (20:56.21)
Totally. It's like I said this the other day and it's like these three sort of A's in AI. It's like what should AI do and like what can it do?

It want to amplifies your ideas. like it allows your voice to go farther with less like less work. So, you know, you can take a podcast and transcribe it and create that into a blog and it amplifies you. It's amplification device. It's also a acceleration device. So, you know, you had some idea you wanted to go from A to B. You could go there completely manually and it's completely possible, but just accelerates that process. So can get from A to B quicker. So it's amplifies, it accelerates.

and it also augments. And so you have some sorts of ideas that you're thinking about it in this way and it could be like, hey, I know you're thinking about it this way, but there's something over here that you're not really considering. Can I bring that forward to you? Does this make sense to play? you're like, wow, that's...

Bilal Tahir (21:53.656)
Yeah. Yeah. I love that acronym, by the way. You should own that. It's Accelerate. Sorry. Amplify, Accelerate, Augment. know? To make it a chant or something.

Pierson Marks (22:06.354)
Yeah, I thought it was just cool because it's just like, really is, you you want it to be proactive in ways that, you know, you're not like covering your blind spots. It's a helpful like companion that's not getting in the way, but it's accelerating your process and then it allows you to do like more things that you would have never even done. So the amplifying part, I don't know. There we go. Triple A. So.

Bilal Tahir (22:15.458)
Right.

Bilal Tahir (22:24.29)
Right. Yeah. Love it. Love it. Triple A. They might sue us, but.

Pierson Marks (22:32.388)
Yeah, yeah, yeah, yeah, totally. Well, I mean, we talked a little about Soham, which is crazy. So I was kind of just blown away by that whole thing. I was like, people are joking now. It's like, has to be fake.

Bilal Tahir (22:43.502)
Yeah, it is crazy. yeah, but you did you know about, um, like the over-ployed thing before this or, okay. Cause yeah, that subreddit is wild though. mean, it's just like craziest thing.

Pierson Marks (22:50.864)
Yeah, yeah, I didn't know. I knew some people.

Pierson Marks (22:56.614)
It's really, yeah, mean, like, it shows a structural flaw in sort of employment and expectations. And, you know, there's a part of me, it's like hustle and hustle. It's also a loyalty part of me that's like, like, screw you. Like, you know, an employer should rightly, I mean, if like, yeah, you get fired. I mean, you're lying to your employer and you're, you're

Bilal Tahir (23:05.678)
Right.

Bilal Tahir (23:10.904)
Yeah. Right.

Bilal Tahir (23:21.346)
Yeah, right.

Pierson Marks (23:25.522)
their trust and everything, it's like also, mean, so him thick, it's crazy. It's nuts.

Bilal Tahir (23:31.244)
Yeah, no, it's crazy. But I also think it's a, because I mean, I was born in Pakistan. So I understand the, there's a mentality, unfortunately, in South Asia, which is a, because you grew up in a very hustler kind of mentality where you got to outwit the system. We call it jugard. Like that's the word for it. Where it's all about, because, because everything is kind of set up in that way. And so if you don't, if you don't have the street smart, whatever, right, you just can't get in. there's this weird

Pierson Marks (23:36.016)
Right.

Bilal Tahir (23:58.104)
kind of fucked up mentality. Unfortunately, a lot of people there, which, you you see that in the stereotype and unfortunately, you know, the Indian scammers, stereotype, et cetera. It's not all, you know, all of them obviously, but there's that mentality. And unfortunately, you know.

Pierson Marks (24:06.898)
Bye.

Bilal Tahir (24:13.438)
I learned, you know, it actually, obviously it's worse for, you know, the other side, but it's even worse for you because actually, if you think about it, the best way, if your goal was to make a lot of money, it's better to have one job that can make you millions. The unfortunate piece is like, these guys are so smart. If they had just taken that talent and applied to one job.

they would have been actually able to make a lot more money with a lot less stress than haggling 10, 100k jobs. They could have made millions of dollars in one job and gotten a great exit if they had just focused their energy on it. But because they have that hustler mentality, it's almost like they enjoy the hustle more than the results. know, it's actually somebody, there was this viral post last year about this. Some guy said, I just, want to do fraud. There's something about fraud that just makes my adrenaline rush. it's like, it's not even about the results. It's like, but that's as sad as it is.

Pierson Marks (24:52.775)
Mm-hmm.

Bilal Tahir (25:05.387)
Some people just have that in their head. if you're anyone, you guys listening, you kind of understand what I'm saying, I'm saying it's actually better for you financially as well. Just focus, play by the rules. You will get your carrot.

Pierson Marks (25:19.378)
Yeah.

Pierson Marks (25:25.766)
Yeah, 100%. Yeah, it's interesting. It's like compounding growth too. It's like, if the growth rate is like very little, which is what you're doing when you have all these little jobs, it's, I don't know, it's interesting. But yeah, very cool and not cool at same time.

Bilal Tahir (25:29.731)
Yeah.

Bilal Tahir (25:45.902)
Yeah.

Pierson Marks (25:46.384)
I think so we have like this whole agenda of like all these things come covering like foul and, music, generative media. one of the things I just want to touch on real quick, just because I thought it was interesting. maybe we'll talk about it on a different episode. I know we mentioned this last time as well, but like AI music, it's something that we haven't really talked about. I'm a musician. I love playing the keyboard. I've been busy and haven't been playing around with it that much, but I have decent, you know, musical sense. the theory has slipped away from me now.

Bilal Tahir (25:49.494)
Yeah, yeah I know. Generative media.

Pierson Marks (26:16.308)
but I can explain things pretty well. But, Suno is an AI music generator and this week there was a band that kind of, I don't know if it's actually used Suno or not, but band has gone viral. get 800,000 monthly listeners on Spotify called Velvet Sundown.

Bilal Tahir (26:37.33)
Maybe I saw that. I saw there was a K-pop movie that wasn't exactly AI, but it was animated characters and they blew up a lot on Netflix. this is different, right? Probably. But Velvet Underground, interesting. I think I did see some of that. This is the future.

Pierson Marks (26:45.255)
Really?

Pierson Marks (26:51.398)
Vel, yeah, Velvet Sundown. It's crazy, mean, 800,000 monthly listeners. I listen to the songs on Spotify. And so it says that, like you can look it up, I mean, just go on Spotify, Velvet Sundown, and listen to songs. And actually, they start off really good. The instrumentals were great. It was like an acoustic rock type.

Bilal Tahir (26:58.998)
Yeah. On Spotify? Wow.

Pierson Marks (27:14.502)
whatever and the lyrics not that good they were still kind of cringy I listened I was like when the lyrics came in I was like bummer because I have heard songs that actually the lyrics are great

Bilal Tahir (27:25.46)
Interesting. So these were like then one short song. It wasn't like a musician samples you, Suno and then crafted a good song or something.

Pierson Marks (27:34.416)
I'm not entirely sure of the process. I didn't research and dig into it that much beyond like the high level, hey, look, there's like some music that people are listening to. it was, it kind of blew my mind, but because I knew this would happen. You're see a whole new generation of artists that are going to be generating music.

Bilal Tahir (27:58.798)
It's insane. No, seriously. There's something we don't understand. Our future celebrities and stars are going to be AI. And the K-pop movie was just such a... One of those, like, okay, fuck, it's happening.

Like it was a movie, it was on Netflix, I forgot the name, but basically it's a bunch of K-pop stars and they're actually like zombie fighters or stuff. I just saw the trailer, but saw that on Netflix. I'm like, okay, not really into that. I like some K-pop songs, but I'm not into it. I'm not gonna watch a movie, especially an animated movie about them. But then I see a Netflix post about apparently it's, in the movie they do songs, so it's a musical. they do songs with like the animated character, kind of like gorillas. They're doing a song, they're a story.

Pierson Marks (28:13.359)
Really? Yeah.

Pierson Marks (28:40.933)
Mm-hmm. Right.

Bilal Tahir (28:42.944)
and they were like the number one or number three top YouTube videos of the week or whatever. The songs were crazy. Millions of views. People love these characters. And I'm like, this is the future. This is digital identity, you know, because think about it. You guys, with Tuna, you can generate a song, but that's just a song. What people connect with is the artist, the story behind the artist, right? And so you can create like, so what they did was...

Pierson Marks (28:50.501)
Totally.

Bilal Tahir (29:07.726)
And this wasn't like AI generated, this was actually human actors behind the scene. But think about it. Like they have a storyline. So you already have a story. You have characters. So you're attached to the character and then the character singing these songs. And you can generate each of those phases. And suddenly you have a storyline, characters in a universe and a story. It's like, that's killer. That's like what audiences want. They want that connection. And so I wonder where you're going to be able to be able to spin up

these worlds, then make a movie out of these characters, have a music track, and then based on the audience's reaction, maybe have a spin-off sequel of a single standalone character, et cetera. You can build a whole universe and shut it down in basically days or weeks, and then move on. So it's crazy.

Pierson Marks (29:57.374)
It's so exciting. mean, like, because think about it. mean, people are, then people always fall back to this whole sense like, oh, but it's not real. And then the reality is, I mean,

It is real, just as much as Star Wars is real, just as much as Iron Man is real, just as much as like your SpongeBob is real. I mean, when you're watching SpongeBob, it's an animated TV show that, you know, it's not a real person, it's a sponge that talks and they do weird things. And it's like, who cares? It was entertaining, at least for children. And it's like, cool. And somebody behind the scenes wrote the scripts and wrote the, did the lines and everything. If AI does more of that process, cool.

Bilal Tahir (30:12.492)
Right. Right.

Pierson Marks (30:38.259)
If it's not good, it's not good. It doesn't matter if it's a human that wrote it or an AI that wrote it. If it's good, it's good. I don't care about the process or the technology used in making those big movies. I listened to this podcast, Jack Altman. So Sam Altman's brother, Jack has, it's like alt capital and he has like this podcast like everybody else, just like us, we have a podcast. He has a podcast.

Bilal Tahir (31:02.83)
I wrote it as a podcast. Hopefully everyone will have a show and a movie here, so maybe that's the next.

Pierson Marks (31:07.969)
Right. But he was talking to Vinod Khosla and they were talking about just the Hollywood industry and they're talking about disruption.

And, you know, earlier this year with the Actors Guild and they all went on strike to like prevent the use of AI in movies so they can all keep their jobs. So Disney and Paramount, all the people that signed that Actors Guild strike essentially, they're like, okay, we won't use AI at a high level. They won't use AI to replace actors. But you know what's going to happen? It's like Disney is going to spend a hundred million dollars to create some movie that makes 50 million dollars at the box office. It's going to net them negative five

negative 50 million, they're gonna lose money on that. But then you have some indie shop that's gonna be like 20 people that can spend a million bucks using AI and make $20 million on that movie. And it's like, you're gonna see this, I hate how this happens with, I don't know, I have to turn it off somehow with the thumbs up. The thumbs up went up when I was recording. But I mean.

Bilal Tahir (32:02.158)
What was happening to you?

Pierson Marks (32:10.991)
Yeah, it's going to be such a crazy world where, it's a good world, where it's gonna make more movies. Like I would consume, if Disney came out and made a Star Wars movie that was good about exploring all those characters rather than taking a year, they could do it every month. I would watch every single one of them. And it's just like, it'd be fun. And I don't care as long as it was good. And it was consistent with characters.

Bilal Tahir (32:34.85)
Yeah, yeah, no, it's so interesting. And I wonder in this age, like, how would you stand out when you can just create Star Wars level content like, you know, right there? And it's almost, I was watching this podcast, Colin and Samira, I think I mentioned to you earlier this week about this guy, he was like, called the bank ski of marketing viral videos and stuff. And he made an interesting point. And I think I'm wondering if this is true. He says that people are less interested in the story.

Pierson Marks (32:50.416)
Right.

Pierson Marks (32:55.515)
Right.

Bilal Tahir (33:02.86)
and more interested in the meta narrative of the story. So yeah, you created a cool show, but people almost are more interested in your struggle to make that show, because they want to connect with you. That's why I show speed or guys like people are just watching people just like sitting there talking or playing video game because they just want that connection. And I wonder maybe in the age of it, I like with this, when everyone can just create cool content, maybe it's about that.

connection to the influencer or someone and how did they come up with that story is more important. That becomes the story, right? So it's fascinating how we evolve.

Pierson Marks (33:39.748)
It's yeah, you're right. It's it's it's wild. I'm excited. I'm excited for this next generation of people and kids that grow up kind of this and this is native to them.

How lucky, I mean, we're lucky to be, you know, relatively like, like if we were at the end of like my parents or my grandparents, they won't see the fruits of all the change. And I said this to my girlfriend once in a while, like, I just don't want to die. Like, I cannot die in the next 20 years. You know, I do not want an accident to happen. was like, please, it's so exciting. We live in such a crazy, crazy time.

Bilal Tahir (33:52.76)
Yeah. Yeah.

Bilal Tahir (34:15.432)
It's a classic thing. You don't want to be like, there's will be some guy who dies the day before long longevity escape velocity or ESI or whatever. And I'm like,

Pierson Marks (34:25.849)
Yeah, totally. mean, and I was on this like the same call that I was on the other day with the AI salon, the head there, one of the organizers, he was talking about how recently in the last like one or a few weeks, last few weeks, all the big major AI labs haven't been talking as much about AGI.

Bilal Tahir (34:46.542)
Yeah, and more about ASI. Right, right.

Pierson Marks (34:47.281)
It went straight to superintelligence. Kind of the AGI stuff is like, they haven't talked about it that much, but they're going superintelligence, superintelligence. So what does that mean? I mean, that's interesting.

Bilal Tahir (34:54.124)
Hmm. Yeah. Yeah. It'll be I don't know. I the more time the more I'm a little I feel like and thinks I'm opposite the same thing where we used to think naively that AGI once AGI arrives or something, everything changes that day. And I feel like ill and the word I guess Akarpati uses jagged intelligence where it will be like super in some areas, but not in other areas. And I wonder, we'll just have this weird phase where we'll be able to get

superintelligence in certain areas but not in other areas and then the question becomes is that even ASI because technically if you ASI true ASI once it is ASI would be good once it's good enough in one domain it should be smart enough to get good enough in pretty much everything that is actually true superintelligence so maybe that's why it changes and that's where AGI is interesting because I mean read even today's LLMs I mean they're

Pierson Marks (35:25.478)
Mm-hmm.

Bilal Tahir (35:46.782)
AGI and writing essays and like there's you can name dozens of tasks that are intrinsic to what we do today that they basically are there from for any metric we had you know and we keep moving though that go post but if you go back five ten years and you show it to them they'll say yeah that's passes the Turing test passes that then all that got blown out the water but nothing really changed I mean things change on the fringe but fundamentally we're the same you society and so the question becomes like you know

When these things get to that threshold, what will happen? How will we adapt, et cetera? Will it be a silent thing that just spreads until one day we wake up and go fuck or it'll gonna be a gradual thing?

Pierson Marks (36:27.323)
I think that, I mean, yeah, I think humans.

This is one area where like where history repeats itself and you hear with technology, you hear the same thing over and over again about one day this technology is going to come, whether it's the steam engine, electricity, the internet, the personal computer, and it's just going to change everything overnight. It's never happened that way. Solar electricity is never happened that way. I don't think it will ever happen that way. And it'll be a slow adoption. be a slow integration. And for the people out there that are like either in this space and like, why should I start a company?

Or why should I do this thing? Why should I study computer science? Why should I like do this when this is right around the corner? I mean, okay then like why should you even? You know, do anything like when something new is gonna like it's like a bad argument I think that you don't do something today because tomorrow something might happen Yeah

Bilal Tahir (37:11.373)
Yeah.

Bilal Tahir (37:16.461)
Right.

Bilal Tahir (37:21.902)
Yeah, you gotta be optimistic about the future, but at the same time you gotta be almost pessimistic in the present where you're like, I'm gonna plan for, I'm not gonna not save my 401k or something just because AGI is coming tomorrow. Just have a balance of both, wolves I guess, and you're right. So I I think of it the same way like you, it's like, I don't wanna die tomorrow, but then I hear these stories, people are like, oh, but AGI is coming, so I gotta be safe. I'm like, yeah, but.

I'm gonna go to Europe, I'm gonna take those trips, I'm gonna live my life. don't wanna be in the very chance that I'm six years old and I've been waiting for AGI or something, right? So, you know, live your life, you know, maybe, maybe, you know, don't go skydiving or maybe do, but with a certified company, right?

Pierson Marks (38:00.284)
Right.

Pierson Marks (38:09.326)
Right, totally.

There was a quote, forget who it was, maybe if you remember who it was, let me know. it was, know, pessimistic people sound smart, but optimistic people make money. And it's like, I thought that was a great quote because it's really true. I know people in my life who are sometimes more on the pessimistic end about the way the world's working. like, oh, you can always look at the dark side. It's like, it's very easy. We're humans, we've evolved to be hyper aware.

Bilal Tahir (38:22.573)
No, yeah.

Pierson Marks (38:41.172)
of all the predators out there and you know that's why news is always the bad news and that's what drives views. It's like okay cool like you could get you could get so caught up in that world or I mean hey this is the way the world is today. What are the opportunities? What can I build so to make my life better make me a happier person and live the journey? I mean that's that's that's the point of life is living that journey so.

Bilal Tahir (38:48.557)
Yeah, yeah.

Bilal Tahir (39:08.334)
Yeah, exactly. It's like, I don't know the exact word, I'm paraphrasing, but I love that word. I think it's by Burial. And he says, you know, don't have sex, don't do drugs, don't ever push, get out of your comfort zone, don't travel, don't do anything that jeopardize your safety. You may or may not live to a hundred years, but you'll definitely feel like it. So yeah.

Pierson Marks (39:32.538)
Yeah. Totally. Yeah. Well.

Bilal Tahir (39:37.646)
Anyways, feel like we haven't really talked about gender media and this is supposed to be a gender media podcast. So maybe we should like touch on a couple of those items. Yeah.

Pierson Marks (39:44.701)
Totally, mean, yeah, we could touch on some of the items. I know, and I know we're coming up on time even too. So I just wonder like, what should we talk about? We have all these things. I mean, this is not something, I mean, we can push a lot of stuff to next week as well. Was there anything that you wanted to touch on?

Bilal Tahir (40:00.77)
Well, one thing I thought was really cool was there's a couple of, a model, we've talked about video generators like VO3 and stuff, but there's another transformation which is video to video, which I think is pretty cool. It's still expensive, but basically it's style transfer for video. So what you do is you take a video of, let's say, us talking, and then you did this kind of like with our first episode, you made a clip of us in claymation. So you can give it a claymation style and then you get the same video, same audio, but you know, I'm like a puppet.

Pierson Marks (40:24.786)
Mm.

Bilal Tahir (40:30.742)
or a Muppet, whatever, cetera. And it leads to all kinds of cool, I think, variations of the original content. So there's a bunch of models for that. But one that came out recently was Rayto Modify. And Luma is a company behind them, Rayto. Decent model. A little expensive. I think it's like 30 seconds per second of output. So it's expensive. But it was pretty decent. And so.

Pierson Marks (40:31.984)
Right.

Bilal Tahir (40:57.438)
what I did was I was trying it out with some memes and stuff so it's on fall we'll link it you know so you guys can check it out there it's a little like

I wish they would just do it under the hood, but you do have to need a workflow. What the workflow is, because it needs basically a frame, the first frame of the video, and then it needs the new style. So what you have to do is you almost need a pipeline where you take the video, you extract the first frame using FFMPEG or whatever, and there's a API for that in fall, and somebody actually built a pipeline so you can just run it. And then you take that.

image and then use something like flux flow context or whatever your favorite image editing model and make it into anime style, claymation style, whatever you want and then you take the new style first frame and the original video and you feed it and you get the final video. So thought it was a cool workflow. I played around with it, you know work decently for

videos that don't have a lot of crazy movement. So especially for still shots, stuff you know, it does pretty decent. The coherence is pretty good. So check it out, I would say it's pretty good. And then if you want to add sound to that, there's a variety of sound effect generators. 11 Labs has sound effects, but also there's models like MM Audio and stuff, which you can just give it a video and it'll automatically infer. Like if it's a guy sitting down playing guitar, it'll just add.

Pierson Marks (42:04.988)
Right.

Bilal Tahir (42:21.91)
guitar sounds or you know somebody cutting fruit you know you'll add an ASMR like cutting sound so if you the reason I wanted to like bring this up was because people have been using Vio3 a lot but Vio3 is expensive and this is a great cheap alternative because almost for a tenth of the price you can just get the the video

either by generating the video using cling or whatever, or using video to video, and then add sound for, and in total it'll be a 10th of the price of your three. But basically, for certain clips, it basically, you know, gives you the same quality. So, you know, it's a great hack.

Pierson Marks (42:55.162)
Right. This is like the stuff that I feel it's there's so many creatives and people that are in this space like even me. I didn't know about Ray to tell you put it on our agenda. And when I click on the link and see foul there it's like what you just said.

There you have to have the video. You have to have an image. You got to take that image and convert it into a new style. I mean, I even look at that thing and I was reading right now while you were talking about this, I was looking at it like how does it work? It's not user friendly. I mean, isn't that crazy to me? It's like we have on the tips of our fingers such cool technology.

Bilal Tahir (43:27.308)
Right, Yeah.

Pierson Marks (43:34.45)
And it's just like, no, you got to do this and you to do this. You got to this together. You got to take it over here. You got to do this thing. And that's going to cost you 35 cents per every second of a video. it's like, what? Like, how is anyone expected to ever leverage these tools until there's a company that comes around and just simplifies it?

Bilal Tahir (43:48.546)
I know, it's so frustrating, but yeah.

And then that's where the low hanging fruit is. When you can optimize, you can create wrappers. actually lets you create a pipeline using a GUI tool, like a comfy UI. But then it actually, once you make it, you can make it basically becomes an API. And so you set it like in the UI, and then you basically make an API. You can build an app on it or use it internally for your work process. you're running three different models. What step one, step two, step three. But instead of doing it manually or stitching it together yourself by calling the API, you make it in fall and that's one single API.

Pierson Marks (44:05.393)
Really.

Bilal Tahir (44:21.776)
underutilized that workflow there, but I like the pipeline flow. But to your point about this, know, like when I saw Ray Du, honestly, I was on a walk and I was so excited because I had all these ideas. What I wanted to do was I wanted to take an old anime shows like the Iron Man intro and I wanted to make it cinematic or I wanted to take the Avengers, like, know, the porous even they all come back, you know, in the final moment. I was like, let's make an anime version. And I was like, yeah, first is super expensive. Second is

Pierson Marks (44:44.39)
Right.

Bilal Tahir (44:51.168)
you know, if a scene changes, you need to give the frame again, because it totally dissolves. So actually what you need to do is you anything one to two minute long video, you actually need to divide it into three second chunks, then make sure those three second chunks have a single scene and it doesn't switch, then get the frame. And I'm like, my God, this is insane. But

Pierson Marks (45:10.438)
Right.

Bilal Tahir (45:11.566)
Somebody who figures that out, mean, and you can like, you can charge a premium for that shit and people will pay, you know, imagine like taking the Avengers scene or whatever and making an anime version of that. You would go viral for sure, right? I mean, it might cost you 30, 50, a hundred dollars, but if you get millions of views, you know, it's insane. So.

Pierson Marks (45:21.82)
Right.

Pierson Marks (45:26.554)
Right. Totally right. It's kind of nuts because this I remember like five five years ago say you know you're looking at I want to build something super cool like what's super cool and all the super cool stuff was hard. You wanted to build an API like let's say you had this idea like I want to build an API as a service. What am I going to do.

I don't know, everything's been built. the reality was like five, 10 years ago, most businesses, I'm not saying all, but like a lot of businesses and the ideas were kind of taken. could, was YC started for still being created for sure. Absolutely. But like there were incumbents, the internet was mature and you you needed a lot of money, VC money, and you need to hire a lot of people, put a lot of smart people in a room and you finally solve some really technically hard problem that you can charge some money for.

and you build a business off of that. Now, when the paradigm shifts, you have like really cool technology that just isn't user friendly. And so an individual developer, like somebody that's just a recent college grad, can go in, find this one thing. Hey, I'm gonna spend a day, like literally you could spend a day, write a Python script, make this workflow, put it behind an API, integrate Stripe in a week, a week max. You have this cool tech,

Bilal Tahir (46:29.582)
Mm.

Pierson Marks (46:50.452)
You wrapped it around with some really convenient features and everything. You charge for it, boom. There's your micro business.

Bilal Tahir (46:56.482)
Yeah, yeah. No, absolutely. There's so much going through. And speaking of that, because I know you're super into product tree and stuff. Whatever you want to talk about, because I know you're a raycast nerd and you use other tools and stuff. So you want to talk about your workflow and stuff, because I feel like you are really good at using the tooling around you, just like 10x your productivity and stuff.

Pierson Marks (47:17.712)
Yeah, totally. I'll talk about this real quick. yeah, I really, think tools are a cool force multiplier and it's interesting and it's important to recognize, know, hey, what's this process that I'm doing over like multiple times over the course of some period, like, your job required you to do this. Like, how can you automate this? How can you think about that differently? And so I won't get into MCP much and

but it's related and we'll dig into it. Maybe next episode we'll talk a little bit more about MCP and how we're using it. But just like this ChatGPT, it's a chat bot. Everybody knows ChatGPT. There's Claude. They have the desktop apps that you can download to your desktop. And a lot of software engineers, they've liked this app called Raycast.

And essentially what Raycast is, it's a spotlight replacement. So spotlight on your Mac is kind of just like this universal search that you open up and you can search your files. it's, you know, it was great 10 years ago. It was cool because Windows didn't really have this, but now it's just like super bad and it, it should have been better. And Raycast came in essentially, they're like, replace spotlight with Raycast and has all your search functionality is everything.

but it automates and it creates extensions for like all these certain things that you can do on your computer, such as managing windows. So I have, you know, like two windows left and right, and I could just easily with a keyboard command move, you know, my Gmail to my left and my code editor to the right, like boom, boom, done, easy. And then I can have shortcuts and I learned some of these keyboard commands and they help improve, you know, the things that I do every single day that take me 30 seconds, but now it takes me one.

But then they also recently added AI chat. And that's really kind of when I downloaded Raycast was they had a desktop chat. So it allowed me to access every AI providers models, chat GPT, or GPT-4, Gemini, Claude.

Bilal Tahir (49:16.398)
you

Pierson Marks (49:27.474)
The other ones like Grok and DeepSeek and whatever, they have all of the models provided. you know, anyone comes out, you can switch between them without downloading a new app, without having to create a new account, you use it. Very cool. But they're really developer forward.

And they added in essentially extensions into your chat. just like how chat GPT now has a web search and some like deep research and image gen, they have that like on steroids and users can create extensions. And so essentially what I did was we have a blog for Jellypod and that blog has our new product announcements. It has just random articles. And in Raycast, I just hooked up an extension and MPs,

And I said, hey, create a blog about this topic. Raycast will go, it'll look at our blog.

the back end of our blog, it'll see all the articles that we've written, it'll see the structure of those articles, it'll write a blog article for me, and it'll post it into our content management system, and it'll set it as a draft. And I'm essentially just living inside that chat window, like, like, make a change to that blog article, make it longer, or, you know, generate the image and put the image in that blog there. And it knows how to do that with the structure that I've already configured.

Bilal Tahir (50:42.542)
Hmm.

Pierson Marks (50:52.212)
And it will just go and live in our blog and so that kind of blew my mind because previously you know how we're working It's like we go to chat GPT. We we put some articles and we connect like we like here Here's a previous article Use this as a reference and you copy and paste and you do this stuff and you have to copy the final output put it into the system so you have to the title you have copy paste the title over Description the blog content and all that stuff, but now it's like why do you need a copy and paste?

everything. Why do you have to like do all this process around doing work to do the work? So now I'm just living in Raycast doing that saying hey no it looks good publish it and you know it just published the article for me.

Bilal Tahir (51:28.803)
Right.

Bilal Tahir (51:35.054)
Love it. No, that's so cool. That's so powerful. And because Recast has integrated to your OS, it has access to all your files and folders on your local machine. So instead of like copy, paste, and so on, you can just say, go to this folder, get my notes for whatever, and make a blog about it or whatever, and then post it there. So instantly connected to all these things. So that's awesome.

Pierson Marks (51:43.099)
Exactly.

Pierson Marks (51:49.639)
Bye.

Right. And they have presets. So I have a few presets. like one is a Jellypod blog writer. And so it has a system prompt that I've wrote. It's like, you're Jellypod. This is what Jellypod is. You have access to these tools. And I can say, hey, you have web search. have the CMS, the content manager system tool, and so all that stuff. And.

Yeah, no, it's sweet. Like, and the next thing is like GitHub. Like when we have product releases, I just connected GitHub into Raycast. So I'm like, Hey, look at the last PR that I just merged. at that code, read the description of that code and write a product update announcement. And it'll do a draft. It'll put it into our product updates in the blog. And then I'll just go in and I'll review it. I'll make tweaks, but at least I have like an 80 % solution. So that's sick.

Bilal Tahir (52:40.834)
that's pretty sick actually. I wonder now I'm like, maybe you should make a GitHub action with like a product up to his flag that just does that or something. That's pretty cool.

Pierson Marks (52:47.814)
I know I've been thinking about that too because yeah, I was like, what's the right way to do it? Is it like a product of this flag? Probably. Is it like runs on every merge and it was just like reviews the description and it's like, this wasn't a product update. That's, that's right. A draft and then ping us like, Hey, I wrote a draft product update for this. Is this right? And you're like, wait, no, don't need that. I'll delete that. That'd be cool.

Bilal Tahir (53:03.31)
right.

Bilal Tahir (53:08.598)
Yeah, yeah, no, so many cool water, mate. I love it. love it. Life is just a series of pipelines. just, you know, gotta polish them.

Pierson Marks (53:15.602)
It's the biggest leverage thing. If you're listening to this right now, it's like I told some of my friends this last week. If you just start to think about problems like this and know that there is intelligence. If you could do it as a human, and it's more of a copy and pasting and clicking and dragging and dropping, there's now a better way to do that stuff.

And if you can learn those tools and you can kind of be adequate in explaining your, you know, and like what you're doing, there are people willing to hire chief automation officers, like audit, like efficiency engineers, like these are going to be the jobs. So. 100 % 100%.

Bilal Tahir (53:38.478)
Hmm.

Bilal Tahir (53:55.854)
Yeah, get him before Soham takes him. This is like, Yeah.

Pierson Marks (54:03.078)
Well, and I know we're like almost an hour now, but we talked about a lot of stuff, not as much media today, but we have all these things on this agenda and we'll toss them to next week. So.

Bilal Tahir (54:10.638)
Yeah, thanks.

Yeah, yeah, and hopefully you guys enjoy, you know, riffing on the meta stuff as well. know, obviously real technology is the core to this podcast, but I feel like a lot of the things we talk about.

know, inform the technology and stuff. So it's important to kind of be on top of it and, you know, talk through it and try to wrap your head around it because the world is changing so fast and there's so many, so many opportunities and it's exciting and it's scary at the same time. you know, I personally, just, I almost live in a perpetual state of film. I'm like, my God, I got to do X, Y, Z, blah, blah, blah.

Pierson Marks (54:51.002)
Right, totally. Sweet. Well, okay. Well, we'll talk next week and have a happy July 4th.

Bilal Tahir (54:56.514)
Yeah? Yeah, happy July 4th. Take care, guys. Bye.