Episode Transcript
[00:00:00] Speaker A: Foreign.
In this episode, I sit down with Brian Castle, founder of Builder Methods, to talk about what's going on in his world with AI, with software development, with creating an AI Forward business.
What's up with Clarity Flow and his new podcast? I've known Brian for a very long time, and as you'll find out, he was one of our very first customers at Podcast Motor, kind of the precursor to Kastos. Brian's someone I really, I love respect. And I think you will too. I think you really enjoy this conversation with Brian Castle.
So I want to talk about AI. I'm in the midst of this hundred days of AI challenge for myself. I'm publishing a video on my YouTube channel every day for 100 days about AI. Published number 20 today. It's fucking hard work. Like an actual YouTube video, not like a short every day.
[00:00:56] Speaker B: Wait, I mean, I want to ask you all about that cool 100 days of AI on the YouTube channel. I did see some of those videos come through at least like on the Twitter feed and stuff. So, like, what is your focus in terms of YouTubing about?
Because I'm doing. I'm on YouTube as well. I'm spending most of my time. And you're right, it's literally just before this call, I was finishing editing my next video for next week.
Yeah, it's so time consuming. I can't even imagine how you're doing a video every single day. Like, I can't get one out every. It takes me two weeks to get a video out.
[00:01:35] Speaker A: So. So, like, first they're all like screencasts, so I think it makes it a little easier. So, like, the video I did, I recorded one earlier today and it was. I saw a piece of news about like, bytedance released their latest. Like, there's LLM that's like huge. And it's open source and stuff. And so was like, I saw this piece of news and I look it up in perplexity and I put all this shit into Notebook lm and then I have it condensed that down and I put it into gamma as a presentation. That was the video. It was like 12 minutes.
But I was like, hey. Because I get feedback. Like, hey, I just want to know about, like practical shit that I can do with AI and that's super practical. It's like, I see this thing, I use perplexity for all my news now it's replaced like Google News and BBC.
And this is like an actual workflow I would go through of like, if I was a, you know, fucking marketing person. At a company and competitor X released this product. That's exactly what I would do is like go get all the stuff about it, smush it together and make a presentation out of it for my boss.
But they're all kind of like that. Like I did one the other day is like bolt versus lovable and actually designing a page for our WordPress plugin.
[00:02:43] Speaker B: Nice.
So I'm doing this thing now.
Like I'm working a lot on YouTube, but one of, one of the next pieces of my business that I need to put into place is a weekly newsletter.
So I'm building this builder methods business. It's essentially a creator like content led business.
I've got the YouTube rhythm down now I need to add like an email newsletter piece. I've been growing the list, but I haven't been sending to it. So I need to send a weekly newsletter and I want it to be sort of like what you're describing, where I'm pulling news, like pieces of news from different sources. And, and I'm focusing on like the developer, developers using AI in their workflow. Right. So, you know, news from the big tools, news from the industry about how to, how to, you know, improve your workflows with AI.
So I'm actually building a SaaS tool now.
I'm calling it Newsletter Lab to help me create my weekly newsletter. And it. And the idea, I'm like halfway through building it right now. It's like I'm putting in sources so like rss feeds, podcasts, YouTube channels, blogs and, and it'll just monitor these sources for links and news items that are relevant to my audience, that meet my criteria. And then it'll like automatically score them based on what I want to share in my newsletter and just assemble my weekly newsletter for me. And then, and then I just need to go in every week and like approve and, and edit a few things and then just, you know, send it out. Yeah, that's, that's awesome. It's like I want to send a newsletter, so I better. So instead of just sending the newsletter, I gotta like build a toolkit.
[00:04:22] Speaker A: It's very developer of you.
[00:04:23] Speaker B: Yeah, exactly.
[00:04:24] Speaker C: Yeah.
[00:04:26] Speaker A: How's that going? Like, I know I see you on Twitter like talking about build a method a lot. Like, how's it going?
[00:04:32] Speaker B: Like, I think it's off to a good start.
I'm at the point. So like, you know, builder methods is.
I'm still trying to figure out what it is, at least what the business model is and how monetizing it, because that is my main Open question. At the, at the moment, luckily for me, which is different from my history in business, is like I actually started with the marketing channel first, which is YouTube.
So I've been working all year on just growing my YouTube channel. I've been able to grow it starting from around like 2,000 subscribers at the beginning of the year. Now it's like over 12k.
And ever since I started making these videos about AI, it started to really pick up and I'm putting a lot of effort, like multiple days per week just producing one video, which is really time consuming. But it is a good top of funnel channel and I'm really excited about that because it's kind of like the first time I actually found a solid channel for something where I'm actually exposed to new audit, new leads every single day. They're coming into my email list, they're, they're filling out surveys, it's, they're, they're inquiring about products and courses and coaching.
But I'm not actually offering any of that yet. The only thing I am offering so far is this free open source tool called Agent os, which is a tool for developers to, you know, give their cloud code or their cursor, like an operating system to build specs and build software more efficiently using these tools.
That's, that's been off to a pretty good start as well. It seems to be like some, some pretty good traction as a, as a free tool. So that's also bringing action into the business on the top of the funnel.
[00:06:13] Speaker C: Yep.
[00:06:14] Speaker B: My big question right now that I'm thinking through is like, okay, it's time to get some first dollars into this business.
[00:06:21] Speaker A: Got to make some cheddar.
[00:06:22] Speaker C: Yeah.
[00:06:22] Speaker B: And, and I've been weighing different options. Like I, for a while I thought this was going to be like a courses business where I would just create courses about building with AI.
Now I'm sort of starting to rethink that and I'm leaning more towards like a membership like you, you get a builder Methods Pro membership and you'll pay an annual fee for that and making it more like workshop based. I think I'm going to start to do live workshops on a regular basis.
And like the thing with courses in this type of space is that as you know, like these tools are shifting. They're changing so quickly.
So the, like, I can't create a course because it's going to be out of date like in a month.
So I think these live workshops where I could just do them repeatedly and they evolve along with the Tools seems a little bit more.
Seems like it actually helps people a little bit more. And then, and then the idea is like, members get access to these workshops, non members can buy tickets to these workshops and sort of start to build it out that way. And I'm, you know, I'm weighing out, weighing out different, like, additional revenue options, like maybe sponsorships, maybe cohort programs. I'm also talking to a lot of teams who want some coaching on, like, coaching their developer teams on leveraging AI. So, yeah, a lot of open questions, things I need to start building. But yeah, but I think as a whole, like builder methods.com and the YouTube channel has what to me, like, the hardest part of the business started to figure out which is the marketing channel.
[00:07:59] Speaker C: Yeah, yeah.
[00:08:00] Speaker A: I want to talk about, like, monetizing AI, I guess, a little bit.
But I have a similar approach. I'm just earlier, I think, which is like, I just want to get through these hundred days.
And my hypothesis is, like, if I will not if, like, until my wife, like I told my wife, after these 100 days, like, my life is going to be different because I'm going to have this huge funnel and this channel and this audience and this way to do a thing. I just don't know what the thing is right now. So I, like, I. I was on vacation two weeks ago and I recorded, you know, from my freaking bed, like, one of the videos because, like, I got to get it out.
But I think that's the question for a lot of folks is how do you monetize AI?
Because there's like, we'll just give the handful of options one, like, as just as a corollary. And I want to talk about SaaS in a minute.
I think what is happening is the gap of, like, bootstrapped companies and bootstrap SaaS is widening a lot where, like, Bootstrap says used to fit in this nice little spot. I think what's happening is, like, it's so commoditized that it's really easy or you're freaking anthropic.
[00:09:13] Speaker B: Yes.
[00:09:14] Speaker A: And I think.
I didn't explain that word. That middle ground is eroding, where you can build like a solid million or $5 million. Business is just different. Anyways, we can talk about that in a minute.
[00:09:26] Speaker B: But, like, I do. I think I'm with you on that.
Yeah, it does seem like.
I think it's still dependent on which vertical you're talking about. I think there's.
On one hand, it seems harder.
It's always been harder. It's always getting harder and harder. Like sass is traditionally hard, but it's like, it seems like hard as fuck now, but there's still the play I think to do like the B2B super niched business like this solution for H Vac companies, whatever, you know, but I think that even that is, is very quickly going to start to become eliminated with, with AI.
Like, you know, people like to like to predict like, oh, there's going to be a whole collapse of, of the SaaS industry and I don't believe that.
But starting a new SaaS from the ground up today, it like I've already decided, like I'm just not like that's not gonna be my business. I, I think that that ship has sailed for the bootstrapper starting up just to start.
[00:10:34] Speaker A: I, I still think, I mean Castos is an amazing business. Like it's, it's such a good flow churn. We have customers coming in, we do some marketing stuff. Like it's just kind of sitting there marinating. It's not a rocket ship, but it's like a really steady business and I don't expect it to go anywhere because like SaaS is, it's like a rock. Like you gotta freaking do a ton of shit to move it. But once you get it there, like does pretty good at staying there.
[00:10:57] Speaker B: I think. You know what you're saying about the, the gap with AI to two new challenges bootstrapped SaaS businesses are going to face in the next couple of years, like one, two, three years.
They've always had the challenge of competition. Like there's, there's just hyper competition in every single category. It doesn't matter how niched down you are, there are 50, 100 or more competitors trying to make the same exact tool as you. And then it's a battle of like getting customers to switch from this tool to that tool. Right? Like that's always been there for the, and that's harder, hard, harder than ever. But now you have the additional challenge of number one, the, the, the AI tools themselves. So ChatGPT and Claude especially ChatGPT I think and OpenAI like they're coming for everyone, you know, like they're making these, these agents and like they want to be like the everything app that like any small business can use to accomplish tasks. And it used to be like B2B SaaS companies can replace spreadsheets. Well now ChatGPT can replace a lot of spreadsheets. So there's that challenge and then there's also the, for maybe like selling to larger businesses, there's always been the question of does the business hire a SaaS tool and just purchase like the enterprise plan, or do they build their own custom software? And it used to be easy to just buy the SaaS tool because custom software is way too hard. Now custom software ain't so hard anymore because you can have a smart employee or two in house who knows their way around AI tools and they can whip up the perfect tool for their use case for their department, you know, yeah, we might not be there today, but there's no question that that's coming in the next couple of years.
[00:12:46] Speaker A: So it kind of begs the question then back to how do you monetize AI like SaaS and the wrapper companies? Like, I know Jordan is doing no right, no effing way.
[00:13:00] Speaker B: I think Jordan's taking a good approach though.
Focusing on AI voice is interesting because like, I don't. That's, that's one where like, I don't know that like an open AI would necessarily get into that side of it.
Google might though. Like, Google sort of is already, they already have like an AI voice thing. So.
But he's, he's playing the VC game so that you sort of have to, have to take those swings, you know?
[00:13:27] Speaker A: Yeah, no, for sure. I mean, I think, I think.
I agree. I wouldn't start another SaaS today, especially like a rapper kind of company.
I think the membership concept is really good. That's the most popular that I see of these folks who have knowledge, have an audience.
A lot of them are on school, which. That's cool. I pay for several school communities and they're really valuable.
[00:13:53] Speaker B: Yeah, that's my take on it. For me personally, like the decision to make my next company be this like membership community, education focused.
That's where I see the wave and the opportunity because we're just entering this time where there is just so much rapid change in every single industry. And for me, I'm a developer, I'm a builder, so it's like it's affecting us. First, my, like, people who build software, it. It is like everything is changing. You know, people are losing their jobs, people are redefining how they build. The way that I build is completely different from it was like 12 months ago.
And so there's just a mass demand for like staying ahead of the curve and, and just learning. So, so that's, that's what I'm trying to sort of get the timing right on is, is entering like, that's where for me personally, I see that as the opportunity.
And then marketing wise, I also think that creators and audience led marketing is for me the only one that works.
I know that other businesses are have success in other ways, but like SEO is going to get harder and harder. It already is.
And that's being replaced with AI results.
Cold email is not as easy as it used to be. Ads are insanely expensive.
Everything else is expensive and difficult. But people trust people and that's where video really works. I think.
[00:15:42] Speaker C: Yeah.
[00:15:42] Speaker A: I think about just in general we can go away from AI, which is like SEO. Like you're going away from the risk of AI by going away from SEO. Or you can kind of lean into AI, which is like talking about it. Right. And talking about using it and adopting it on a human platform, which is exactly what you're doing. It's exactly what I'm doing too. Which is like a platform where people trust people and want to engage and learn from people about a topic that is affecting literally everyone on earth right now.
I mean, my kids have chatgpt on their phone. Like yeah, 13.
[00:16:20] Speaker B: We did a road trip. We did a road trip the other day. Now our like, our new favorite thing to do in the car is to run chat GPT and do like a trivia game in the car with the kids, you know.
[00:16:29] Speaker A: That's cool.
[00:16:30] Speaker B: Yeah, yeah, it's crazy.
[00:16:32] Speaker A: Yeah, it's a trip.
[00:16:33] Speaker C: Yeah.
[00:16:34] Speaker A: I mean, I think, you know, actually right behind YouTube, my favorite channel right now is cold email though that kind of tells you the state of the rest of the conventional like white glove, happy marketing channels that I know you and I kind of grew up on. Like, it ain't content, it ain't SEO, it ain't LinkedIn. LinkedIn outreach works pretty well. I use Dripify to like automate a bunch of that.
[00:17:00] Speaker B: I hear people, my whole life I've heard about people talk about LinkedIn. I'm like, one of these days I'll do something on LinkedIn.
[00:17:07] Speaker A: Oh, it's assessable. I mean for organic stuff, it's just terrible. But sure enough, like you send a bunch of. It's just like cold email just on a platform.
But yeah, I mean I, I think like, and maybe that's why I feel like the, the capacity for kind of bootstrap SaaS is narrowing. Is the number of decent marketing channels is really reduced to cold email. Maybe YouTube and, and like founder led marketing.
[00:17:33] Speaker B: Yeah.
[00:17:34] Speaker A: Which I think it's like will definitely have its day soon.
[00:17:37] Speaker B: I think it's like it's always been. If you already have an inroad if you have some sort of advantage, whether it's, you have a large audience or you really know about some painful aspect of some niche industry that nobody else even knows about.
[00:17:53] Speaker A: Which is pretty unlikely now.
[00:17:54] Speaker B: Which is, which is pretty unlikely. But it does happen.
But then it's a question of like, do you even have the capability to build a product?
And then, and then to like it's industry specific. But I, you know, those, those opportunities still exist, but it's harder than ever, you know. And like, and when we're talking about a bootstrapper, like, yes, you, you might be able to find one of those rare opportunities that still exist, but if you're bootstrapping, it's a matter of time. Like how much time can you invest into this risk and how much risk are you willing to take on? Especially at, and at what age are you in your career? Right. Like, for me I decided like, because I've been doing SaaS businesses for years and like I decided like it's not worth investing multiple years into a, into a big question mark of whether this, whether I can actually get it to product market fit.
Whereas something like investing in YouTube and then investing in a membership type business that seems a little bit more straightforward to get to revenue.
[00:19:01] Speaker A: Yeah, I heard, I think John Waro, like the built to sell guy, Talk about like 10 year cycles in your life that like, it's just a natural kind of cadence. Seven to 10 years you have like a career change or a life change or something like that. And that definitely resonates. I mean we're right at 10 years for me with like podcast Motor and then Castos is, you know, a couple years behind and yeah, I, I won't do it forever. Right. I'm not, I'm not going anywhere. Like Cast does not go anywhere. But like 100% of my time on, on it is, is like not, not the case anymore. And actually I think it's healthy. You know, like I've, I've taken a kind of half a step back and the business in some ways operates better because of that.
[00:19:45] Speaker C: Yeah. Yeah.
[00:19:46] Speaker B: I mean that's, that's where I ended up with my last SaaS as well. It's still a current SaaS that I own called Clarity Flow. And I was full time, 100% focused on it. No other shiny objects for, for three plus years, for almost four years of that business.
And then it got to a point where it, it didn't turn into the, into the rocket ship that I hoped it would originally.
And so I decided like I'm still going to hold it, it still operates just, just fine. But I can't invest years into it anymore. I need to start to think about what that next phase is. And that's what I've been focused on for the last two years. Sort of hopping around different ideas until couple months ago when I really landed on this, on this builder methods direction.
But you know, here I know that we're both in our 40s. It's the mental game is so much different now than it was in my 30s and late 20s when I was doing new startup ideas. Like back then I was just like, whatever, I'll just try whatever.
I could throw a couple of years at this or that. I've got plenty of time and I know I'm not, I'm not retiring anytime soon, but my energy, like, I know that I only have a couple more years of like super high energy output to throw at a business.
[00:21:09] Speaker C: Yeah.
[00:21:10] Speaker B: And I already feel it now, so I have to, so. And what that does is it makes, it makes it much more difficult to pull the trigger and commit to, to a business direction without, you know, just, just. It used to be just like, all right, I got this, I think it's a good idea, let's try it. And, and now and, and then commit myself to a year.
[00:21:34] Speaker C: Yeah.
[00:21:34] Speaker B: And now it's like, well, if I'm going to try something, I gotta be really feeling, feeling confident about it, you know?
[00:21:41] Speaker C: Yeah, yeah.
[00:21:42] Speaker A: I think it's, it's the confident and the time aspect of it. It's like I, I want to do something for the next 10 years. I want to do something I can do for the next 10 years. And like, yeah, that when you're 45 or 40, whatever, like that math is different.
The risk tolerance is different, ability to work is different. The place you should work kind of in the stack is different too. Like, you know, shouldn't work in the weeds as much. Should be a thought leader at this point because we've been doing this a while and I think there's definitely like an important part in like the kind of ecosystem for all that.
[00:22:13] Speaker B: Yeah, for me I, I tend to do things a little bit backwards for some reason.
I think that in previous businesses I was much more high level. Like, like when I was running audience ops, I had a whole team including managers and I was removed from that business, like didn't touch it day to day.
[00:22:33] Speaker A: Glad you got out of that one, huh?
[00:22:35] Speaker B: Yeah, especially I sold it in 2021, which is like about two years before I really came around and I had no idea that the AI wave was coming. I did. I, I did hear about like GPT2 at the time, but it was just this novelty that nobody was touching.
But yeah, I am, I am glad I'm out of that business now.
But like back then I was much more high level, like systems team influencer, not in the weeds. And now I actually like the crap. Like I'm really enjoying the craft of building, designing and building software and also creating videos, creating content on YouTube. Like, I, I like spending a lot of hours on that kind of stuff.
Not to a point where I just want to do that like as, as an employee for another company. Like, I still have the drive to like own my own business. So. So that's why I'm trying to figure out how do I turn this craft into a business that I both enjoy to work on every day and can actually grow.
[00:23:48] Speaker A: Tell me about your process for YouTube.
[00:23:53] Speaker B: This is interesting. I want to hear yours too.
I know you shared it a little bit already.
What you're doing with the 100 days, I think is really smart because the key strategy, no matter what kind of YouTube strategy you're doing is like the more videos you do, the better you're going to get.
You're going to improve some. Like the game is just improve something on every single video. And there are so many little things that you can improve.
From choosing the right topic to the script, to the title, the thumbnail, how you edit, how the, the pacing of the video.
There are so many little things that you can just improve.
Like the other, like two videos ago for me, I switched from reading my script to using a teleprompter and that like saved an hour of recording time, you know.
[00:24:47] Speaker C: Yeah.
[00:24:49] Speaker B: And.
Yeah. So right now it's, it's. For me, it's all about.
My big challenge right now is figuring out how I can cut down the time it takes me to produce a video.
Earlier in the year I had an editor that I was working with and it was more of a workflow where I would just record something raw, throw it into Dropbox, let him do all the post production, and that worked fine. He was a good editor.
I didn't really have the topic selection nailed down to a science the way that I do. It's not really a science now, but I just put a lot more care and effort into it.
[00:25:28] Speaker C: Yeah.
[00:25:28] Speaker B: Um, so my process now is mostly thinking a lot about what the next most important topic is that I want to create a video about. I usually just, I'm Already I'm always thinking of something that's like, on my mind, and that's usually the thing that I'm going to do next.
Um, but then I'll spend like a good few days, like shower thoughts and stuff, just, just thinking about, like, okay, I know the general topic, but how am I going to frame it? Like, what, what's the title and the thumbnail going to be generally? And I'll just have a rough concept for that first.
Then I'll spend a day or two doing a script for it.
And I'm using a lot of Claude in this process now. So I have a cloud project called YouTube Planning.
And in, in that project, I've uploaded all the transcripts from my past, like 10 videos.
So it, it knows my tone, it knows my past topic selections, it knows how I like to say things. And so then I'll just, I'll write into Claude, like my raw concept for a video, maybe a couple bullet points on, like, what I want this video to cover. And then Claude will draft me a first draft of the script. And then I'll go back and forth with Claude maybe five times, and then I'll bring it into a text doc and I'll just tweak a few phrasings to make it really sound like me. All right, so now that's like two days of work. I get that dialed in, I've got a full script.
I'd say most of the, Most of the video is scripted. Usually the middle part where I'm like showing my screen and doing some coding stuff, that. That part is not scripted, but especially the intro. I put the most emphasis on the first two minutes to focus on the hook.
Then I do recording. Usually, like on a fresh day, like in the morning, I throw it into. I just started using the teleprompter again. So I put my iPhone on the teleprompter and read it off of that.
And then I get into editing, which is another. Usually, like two days of editing. Yeah, I've jumped around from different tools. Right now I'm using Adobe Premiere and I've started to build up a library of B roll. That's something that's been really helpful. So I do a lot of B roll. So there's a lot of, like, me talking on camera.
That's like the whole video.
There's a middle section where it's me plus my screen showing something, but there's a lot of time where it's just me on camera. And for that I really try to show some B roll. At least every 10 or 20 seconds to. And it's usually B roll that I created showing clips of things that I'm talking about.
And, and a lot of my videos are the same, so I'm doing videos about software development with AI. So I've started to build up a library of B roll clips that I talk about all the time.
So I just reuse a lot of the same clips. If you look at a few of my videos, you'll start to see like, these B roll clips are the same.
[00:28:40] Speaker C: Yeah, yeah, yeah.
[00:28:42] Speaker B: And, and so that, that I think, helps with the rhythm and it keeps viewers engaged.
And, and it, especially in that, in those first two minutes, like a lot of B roll in the first two minutes just to keep that action going.
[00:28:54] Speaker A: Mm.
[00:28:57] Speaker B: What else after that? I, I, I also use Claude to like draft 10 different title ideas and thumbnail ideas.
I, and I usually just settle on one and then I, and then I'll use YouTube's like a B tester for the thumbnail. So, and, and Claude gave me like two, two things to test.
[00:29:18] Speaker C: Yeah.
[00:29:20] Speaker B: In terms of structure, I, Right now I've sort of landed on.
First minute is like an intro to hook them and sort of convince you to watch the whole video. At the end of the first minute, I'm going to make a pitch for my newsletter.
Every video now has like a dedicated pitch with some B roll that says, like, if you want to really stay ahead of the curve with AI, join my Builder Briefing weekly newsletter, you know, so that points them to my, to my website. And people are going over there and signing up every day.
[00:29:52] Speaker A: So that's interesting.
[00:29:53] Speaker B: That's kind of good. And then I get into the meat of the, of the, usually showing my screen.
Then we get into the ending 20 seconds. So if you liked this video.
So if you like this video all about Claude code. Well, there's this other part of Claude code that you really need to learn about and I've got another video that talks all about that. So when you're so, so now, like, you know, watch this one over there and point.
[00:30:16] Speaker C: Yeah.
[00:30:16] Speaker B: You know, get them into the, into the rabbit hole. So.
[00:30:20] Speaker C: Yeah. Yeah.
[00:30:22] Speaker A: How do you, how do you script the whole video while sharing your screen? Because, like, presumably, like you're clicking around buttons and doing stuff and writing code, are you like, talking about that code you're writing or is it just like they're disconnected and so what you show on the screen isn't really important When.
[00:30:40] Speaker B: I'm like, working on something on screen or like actually demonstrating, like Me doing it live.
Not live, but like on the video.
[00:30:49] Speaker C: Yeah.
[00:30:51] Speaker B: It's not scripted word for word, but I. I do spend a lot of time pre planning what I'm going to show and sometimes that involves like building a demo app or something like that. So I. That's another prep time. Figure out.
But the.
But like, I'm. I'm pretty good now at like, because if I'm going to show it on a video, I'm really familiar with what it is already. Like, I've already done it a hundred times, like, just in general. So, like, I'm pretty good at just like doing it. And my camera is right here to my left. So I just pretend like I'm talking to a co worker or like an intern and I'm just. And they're looking over my shoulder and I'm clicking around and I'm explaining what I'm doing, you know.
[00:31:38] Speaker C: Yeah.
[00:31:39] Speaker B: But, you know, that's also the hardest part to edit. So like, I just edited one today.
The final video is going to be like 20 minutes. The recording was like an hour.
[00:31:50] Speaker C: Yeah.
[00:31:51] Speaker B: Of footage. So there's a lot that I just cut. You know, like if and. And especially if I'm not working off of a script, like for those sections, I'm going to ramble, like way too much. So when I get into the edit, I'm like, okay, I just said the same thing like four different times. I could cut three.
[00:32:07] Speaker C: Yeah.
[00:32:09] Speaker A: So that's cool. That's cool. I. So a couple of things that I've kind of played around with. So I was doing on my YouTube channel, like a. A fair bit of like founder stuff and sales stuff before, and I was. That was all entirely scripted.
I mean, I'll just say what I did, because it's really shameful is like I would just take. It would basically the skyscraper method for YouTube. Right. Just take like this video and this video and this video, stick them into Notebook LM and have it just give me the best parts of all of them and then either stick that in Claude or I used subscriber AI a little bit. It's like, I don't know that YouTube script. It's a YouTube script writing tool.
It gives pretty good scripts. Like, they're really thorough. It has all these like, here's the payoff and here's the hook and all this kind of stuff.
[00:32:58] Speaker B: Nice.
[00:32:59] Speaker A: Yeah, so that's fine. Like, I think if you're doing talking head stuff like that, something like that is fine.
[00:33:07] Speaker B: Have you tried. What is it? Vidiq?
[00:33:10] Speaker C: Yeah.
[00:33:11] Speaker A: It's. I don't use it. I have it. I use it just for, like, in my YouTube dashboard to give me.
[00:33:17] Speaker B: Yeah.
I think I had an account like a year ago and I looked at it like three times and it just.
[00:33:23] Speaker A: Right.
[00:33:23] Speaker B: It's just never got. Got into my workflow, you know?
[00:33:26] Speaker C: Yeah. Yeah.
[00:33:29] Speaker A: So. So for my. So, yeah, I think scripting is really powerful. Like, I think everyone should script as much as they can. It just makes you sound better because the words are planned out and you're not just winging it.
Like, that's. It's so important.
[00:33:42] Speaker B: That was a big difference. So earlier in the year, like around January, February, I was doing these no script videos.
Uh, I would just record Raw and give that to my editor and be like, you figure it out. Find the interesting parts, you know, um, and no surprise, like, those videos got like a hundred views and the ones that I do now do thousands. Because I actually care. I actually write.
Especially the hook, you know, like, that is completely scripted.
[00:34:15] Speaker C: Yeah.
[00:34:16] Speaker B: I used to think, like, oh, I can't read a script because it won't sound natural.
I think it's still pretty obvious that I'm reading off of a script, but I've improved at making it sound natural.
But either way, the content is just more compelling because Claude helped me write it and I stressed over every word and the ang. Like the positioning and the pacing.
[00:34:40] Speaker A: Yeah.
So that's. The next thing I need to do is like, script at least the intro and the outro. Like the hook. The hook in the first minute and then the. The bit at the end.
Because what I'm doing is like. I mean, it's every day, so it's like. And some days I'm doing two a day. If I want to, like, get ahead.
Like, I just turn the camera on and I say, like, I have a pretty good. Like, I've done it enough to where I say, like, in the first 10 seconds, you have to, like, tell them exactly what's about to happen. Like, it is a lot more like a short than anything these days. So, like, the one today was like, ByteDance, just released the biggest LLM ever and it's entirely free and open source, blah, blah, blah, blah. Right. Like, you got to hook them.
[00:35:25] Speaker B: Yeah, for sure.
[00:35:27] Speaker A: And I'm just like, I'm sharing my screen the entire time, basically. Like, sometimes I'll be full size and then I'll go down to like, the thumbnail thing or like, whatever, so you can see my screen, but I'm just like, playing around with the Tools and stuff.
[00:35:39] Speaker B: I think, I think you're right that it's all about the, the intro script, first 10, 20 seconds for sure. But I, I think it's even more important to have the topic, like the topic selection.
[00:35:52] Speaker A: Right.
[00:35:53] Speaker C: Yeah.
[00:35:54] Speaker B: Which then is. Is like the title and the thumbnail.
But it's not just like any topic, as long as you have a strong thumbnail will work. It's more about like just choosing something that you're pretty sure that your audience will be like, ooh, what's that? You know, that's.
[00:36:10] Speaker C: Yeah.
[00:36:11] Speaker B: And that's the hardest thing. You'll never, you'll never really know. I mean.
[00:36:14] Speaker C: Yeah.
[00:36:15] Speaker A: But yeah, yeah, that's wild as well.
[00:36:19] Speaker C: Yeah.
[00:36:19] Speaker A: I mean, I'm editing about half of my videos just because I'm so far behind. Like, I can't. It's not really.
[00:36:23] Speaker B: I wanted to ask you, are you editing or you have an editor?
[00:36:27] Speaker A: Yeah, about half the ones I'm editing. And I just use screen. What is it that I use? Screen flow.
[00:36:33] Speaker B: Yeah, it's pretty basic.
[00:36:35] Speaker A: Yeah, it's pretty basic. Like just cut, you know, different tracks, mute zoom, all that kind of stuff.
Just because they're every day, like it can't be, you know, freaking Mr. Beast style.
[00:36:46] Speaker B: I use Screenflow. I still use it sometimes, but that was my go to for like 20, like 15 years.
[00:36:54] Speaker C: Yeah.
[00:36:55] Speaker B: Only this year I started to play around with the pro tools.
I was using descript for a little while, but it's just too buggy and slow.
And then I was using DaVinci Resolve for a couple of months, which was pretty good.
And I just recently switched to Adobe Premiere. And the only reason I'm switching between these is I'm just trying to find like, what is the absolute fastest workflow for me with keyboard shortcuts and everything and tracking my B roll. And right now that's Adobe Premiere.
[00:37:30] Speaker C: Yeah, yeah, yeah.
[00:37:32] Speaker B: I mean like I do. This is the. I'm, I'm like saying it today. I was like, I gotta hire an editor. Cause this is too much, you know?
[00:37:40] Speaker A: Yeah, yeah, they're, I mean they're definitely, they're, they're a lot more expensive these days. Like you go hire an editor, like even in the Philippines, they're like 2,500 bucks.
[00:37:51] Speaker B: My concern also with it is like, my thought was like, all right, maybe I can find a smart but low cost editor and I can just really train them on my method of using Adobe Premiere.
And I probably could do something like that. But my concern is that my content is technical.
And, and the most time consuming part is choosing the B roll and putting the B roll clips in place.
And you can't even do that unless you know what I'm talking about.
[00:38:24] Speaker A: Be tough.
[00:38:25] Speaker C: Yeah.
[00:38:25] Speaker B: You know, like when I talk about a piece of code, you got to show like that code, you know, Right. And right.
I don't, I can't find a developer, a video editor who would know that kind of stuff.
[00:38:37] Speaker C: Yeah, yeah.
[00:38:42] Speaker A: I want to, I want to like wrap by like talking about kind of where you think AI is going. So I see like in this week in the news, like Sam Alan saying like, oh, it's probably overhyped and everybody's talking about a bubble and all this kind of stuff. Like, I'll tell you my, like, I.
Two months ago I was like, kind of depressed and like we're super worried about it. Like, holy shit, it's going to ruin my life and my business. We're all going to be living on a farm, like, you know, in a commune.
[00:39:11] Speaker B: There's a lot of like, doom scenarios and like, many to choose from.
Are you thinking about like, it's going to crush your business? Are you thinking about it's going to like, kill humanity? Are you thinking about like the. It's going to cause an economic depression.
[00:39:24] Speaker A: Like, like the first and the last? Yeah, I think that, I think that probably, like what's pretty likely is it will cause mass unemployment which will cause a huge depression.
It's happening. Like you're dumb and blind if you don't see it in the numbers from the unemployment numbers and the new job numbers. It's 100% happening. Nobody's hiring other than these big AI companies.
[00:39:50] Speaker B: I agree with you. That's the thing that I'm most concerned about, I would say is at some point in the very near future, it doesn't feel like we're there yet, but I think sometime in the next two to five years it's going to start to come for all the jobs and that's going to be a massive disruption and I just don't know how bad that's going to get, whether it's going to be a depression, whether it's just going to be a recession, whether there's going to be some new. I don't know what it would be, but some rotation into.
Yeah, there could be a positive that comes out of that maybe quicker than we think it will, but that's my op that's being optimistic. I, I think there's going to be a lot of pain out there.
Yeah, I I agree that like on that front, like you're, you're blind if you just don't see that coming. But at the same time it's like the everyday person today does not see that coming. No matter. Like, yes, they're talking about it on the news, but it's still just like, ah, that's funny.
But like the, the school teachers in my, in my, in my kids school or my parents who are still working, like it's still, still every day, like nothing, nothing's really changing. They're not. Yeah, they've maybe tried chatgpt once or twice, but it's not in their life yet. Yeah, for me, I'm a developer, so my whole industry is all, it's already undergoing massive changes, but there's no getting around it. Especially as the older generation is, is going to retire out.
The newer generation, there's going to be fewer, fewer of those jobs to replace.
You know, I know that younger people coming out of college, it's, it's bleak because of AI right now.
[00:41:46] Speaker C: Yeah.
[00:41:49] Speaker A: What are you most excited about?
I think that I'll give you a second to think and I'll just say how my perspective has shifted a lot since I kind of was scared and leaning away from AI to leaning as hard into it as I possibly can.
Because now AI is the rising tide and either you get like smashed by the tsunami or you're in the boat on the top of the wave. Right. And like I'm going to be in.
[00:42:20] Speaker B: The top of the boat too. That's how I feel like, yeah, I'm most excited about it based on what I do. I build products.
So I, I'm super excited about like I could build products 100 times faster, you know, And I get a lot of comments on my, on my stuff from developers who are like, oh, you're, you're part of the problem. You're going to replace all developers with this AI stuff, right? Like there's some negative negativity out there about it. And I'm, I just don't agree with that mindset. It's like, okay, if you, for me, I'm just talking about like programmers, coders, right?
If you, if you just really like writing code, if you are a programmer and you like the art of programming, then I hate to tell you that's going away. You're not like, you could do it as a hobby, you could do it as a craft, but businesses and production level software is not going to be written by humans in the very near Future. It already 90% of my code is not. Yeah, but if you are more like, like, the only reason I ever got into software development was to build products. Like, it's a, it's a way for me to build and ship my own products.
So from that perspective, I love it. And like, and the side of it. And this is. I've done a lot of content on this recently where it's like the, to me, the, the best, most creative, most interesting parts of creating software are the parts that are not handled by AI, and they probably won't be anytime soon. So figuring out the right problem to solve for users, architecting the right UI and the user experience, positioning it, communicating the problem, marketing the problem, that's the most interesting part to me.
And it's still very technical. It's still very, you know, really focusing on the architecture and solving, like, system problems.
[00:44:15] Speaker C: Yeah.
[00:44:15] Speaker B: But I just don't need to, I don't need to remember how to write a line of ruby code or JavaScript code anymore in order to do that. That's the AI's job now.
And I think the sooner we as an industry can start to embrace that mindset. But then I'm also really excited about, like, I think that it's, I just really think it's a multiplier. Like, when we talk about YouTube content and just content in general, any kind of writing at this point, I'm becoming much more comfortable leaning heavily on AI.
[00:44:48] Speaker A: Kind of like a director. Right. Like, this is generally the plan. Go this way.
[00:44:54] Speaker B: Yes. It's not like, it's not a replacement for me as the person. Like, it's not going to create the content, for the content won't exist without me in the picture. Cause I'm the one with the idea.
[00:45:07] Speaker C: Yeah.
[00:45:09] Speaker B: And I come with the idea because I see some need out there, and then I come with like a raw version of it. Like, here's kind of what I want to say about this, a couple bullet points.
Now give me a draft and make it sound interesting.
And then all of a sudden, like, their draft is like, way better than something I would have written.
And then I just need to tweak it a few times and it's like.
And now I've published something that is ten times better than what I would have written myself on a blank page. And it's still, it's still getting my idea out there, you know, and, and that, that is just a multiplier, you know, so when people say it, it replaces creativity. Like, no, it doesn't. It. To me, it multiplies. It you know.
[00:45:55] Speaker C: Yeah, yeah.
[00:45:58] Speaker B: One thing, I haven't really talked about this, but a lot of people are like artists, visual artists, music artists. A lot of them talk about like AI replacing that whole part of the world. And I don't agree with that at all. I think that there's always going to be.
I think that's a big prediction that is very wrong if people think that like there's not going to be human rock stars and bands and musical artists and painters out there in the world.
I think there always will be that.
And writers too, you know, authors of books and stuff.
Yes. I can like technically create stuff. Like it can, it can generate a pop song, it can generate a book.
But I think that there's always going to be people creating art because it comes from their, their, their personal experience, their, their inspiration. And then there's always going to be people who want to enjoy the art of a human. There's always going to be that demand, I think.
[00:47:04] Speaker A: Yeah, I don't agree.
No, I hardcore don't agree. No, I just think that we're not there. Right. In the context of today, you as the artist, in your code context, you have to guide the AI to this place. And that's what AI is today. It's a really helpful assistant. It's really great autocomplete. Right. Which, which like not to minimize it, it's an amazing tool. I literally spend like half of my waking hours in some kind of AI tool which is allowing me to run a, you know, freaking seven figure SaaS business like with half the team that we used to have or a quarter.
I think let's call it end of next year there will be such an advance to where it's just like it's going to take a couple more levels of cognition to where it's going to say, oh, this is what a good song is. This is what makes like AC DC and like Pearl Jam and all that. This is what makes them good. I've analyzed a bajillion bytes of code and I understand like what makes good rock music.
[00:48:11] Speaker B: I think I could do that already today there was that I'm wasn't there on Spotify. That was like completely AI generated. Did like a million downloads or something like that.
[00:48:21] Speaker C: Yeah.
[00:48:21] Speaker A: And it's going to. And I think the big thing is it's going to go create a thousand versions of different songs and one of them will be a winner.
[00:48:29] Speaker B: I think that's true. Like I think that's always. That's gonna happen. It's already happening.
But I Also think that that is going to create more demand for, like, the pure stuff, the human stuff, you know, like, people are going to know that that's AI, and then they're going to seek out real art.
It's going to create a hunger for it. Um, yeah, you know, like, I was listening to. I listen. I like to listen to Bill Simmons podcast, which is mostly about sports, but sometimes about. He. He was talking to Chuck Klosterman about AI, and they were talking about Nirvana, right? And.
And they made the. The point, like, if you played. If you play a Nirvana song to two people, maybe one person is a bit younger, and they don't really know the story of Kurt Cobain and the fact that he committed suicide. And. And. And, like, the. They're. They're just hearing the song smells like team spirit. And the other person knows the whole backstory.
Like, the person who knows the whole backstory is going to receive that completely. Is going to hear the music completely differently than the person who knows nothing about Kurt Cobain, you know, because they know there's a person behind it, and he was going through a lot of. So it's like. Like, I think that that kind of thing is always going to be there, you know? I don't know.
[00:49:51] Speaker A: We'll see.
[00:49:52] Speaker B: We'll see.
[00:49:54] Speaker A: I'm happy. I wanted to talk because I'm glad we're talking because we've known each other for a long time. I'm glad to see you're kind of leaning heavy into this. I'm super excited about leaning heavy into AI. I think it's definitely the most exciting thing going on in our kind of, like, digital business world right now. Whatever kind of flavor of that you want to call it, the whole world.
[00:50:16] Speaker B: So I talked to my family about it. They're like, oh, here. Here he goes again, talking about AI.
[00:50:21] Speaker A: You know, I was done visiting family last week, and I was like, yeah, I did this thing in. In Chatgpt, and then I did this thing in Claude, and they go, what's Claude?
[00:50:28] Speaker B: Yeah, Nobody knows.
[00:50:29] Speaker A: I was like, but you're cooked. You're totally cooked. Yeah. All right, buddy. So what. What's the best. So cast jam on Twitter. What's the best place? Where's the newsletter? What do you want, folks?
[00:50:41] Speaker B: Yeah, I'm working on buildermethods.com that's. That's the thing.
[00:50:44] Speaker A: Awesome. Awesome, buddy. Thanks, Brian. It's good to catch up, man.
[00:50:47] Speaker B: All right, Craig.
