YT YouTube2Text

YouTube2Text → Transcripts → Taylor Otwell on Why Laravel + AI Pair so Well, and the Future of Programming

Taylor Otwell on Why Laravel + AI Pair so Well, and the Future of Programming — Transcript

by Laravel Podcast · 11,141 words · 1,585 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:09[music]
0:10All right, welcome back to Laraveville
0:11podcast season 7. I'm your host Matt
0:13Staer, CEO of Titan. And in this season,
0:15I'm joined every episode by a member of
0:17the Laravel team. And it's kind of weird
0:20to call Taylor a member of the Laravel
0:22team, but whatever. Today, I'm joined by
0:25Taylor. What is your official title? Are
0:27you CEO? I forget every time. Yeah. CEO.
0:30>> Okay. CEO of Laravel, OG, originator,
0:32creator, uh, Papa Bear, whatever you
0:34want it to be. Uh, Taylor, would you say
0:36hi to the people?
0:37>> Hey, what's up everybody? Good to be
0:39back.
0:40>> Dude, it's been so long. Um, you you
0:43know this, the listeners don't know
0:44this, but I am uh wrapping up the season
0:46and planning next season. And as I'm
0:49doing that, one of the things I've done
0:50is like kind of gone back to the
0:51previous seasons. And I have so much
0:53nostalgia for when, you know, both when
0:55it was just you and me, but also when it
0:57was uh you and me and Jeffrey and when
0:59we had Adam and I'm just like, ah, we've
1:01had so many good times on the podcast.
1:02So anyway, thanks for hanging out.
1:04>> Yeah. Yeah. It's good to be back. It's
1:05good to be back. It's been a minute.
1:06>> Okay, so there's so many things I want
1:08to talk to you about, but Laron is next
1:11week and I think I remember Kayla
1:13telling me that this episode that we're
1:14recording on July 20th, I think this is
1:16coming out on like the 21st or 22nd. So
1:19this will come out before Laron. So, I
1:22know that you're in Larcon prep mode.
1:23So, what's going on? What's interesting?
1:26What's new? What's different? If people,
1:28you know, aren't going or are going,
1:29what should they look forward to? Kind
1:31of what's what's the situation with Lar
1:32right now?
1:33>> Yeah, I mean, it's it's we're in the
1:35final stretch here the last week. Um,
1:37I'm really excited about this year's
1:38Larcon. I think the venue is really
1:40cool. I really like Boston as a city.
1:41Honestly, we went there as a family. Um,
1:44>> I guess, yeah, last fall. Yeah, last
1:47fall. Really enjoyed it. Cool place. Did
1:49you guys know it was a potential venue
1:51when you went?
1:52>> Oh, I don't think we did actually, but
1:54it was soon after that. It was soon
1:55after that that we got that figured out.
1:57>> Um, no. But I'm excited about it. Um, I
2:01mean, we have a lot of great talks. I
2:02can speak to my talk a little bit. You
2:04know, I'm going to be doing kind of the
2:06usual um demo of some like open source
2:09stuff just like in the code like I
2:11usually do. If you've ever been to a
2:12Laron, I kind of have like a routes
2:14file. I show some new features. We kind
2:16of walk through it together. So there's
2:18a chunk of that. Um, and then I'll dive
2:20into uh some of the cloud stuff we've
2:22been doing. And I should say on the open
2:24source side, you know, there's a mix of
2:26kind of like some framework stuff, um,
2:29some inertia stuff, uh, some AI SDK and
2:32boost stuff. Actually, some pretty cool
2:34stuff coming there. Um, and then I'll
2:36dive into some cloud stuff and kind of
2:38um a recap of stuff we've recently
2:40launched maybe in the last like month or
2:42so, as well as um some things that are
2:45launching sort of like now at Laracon
2:48and then um a preview of some things
2:50that are coming in the next couple of
2:52months, let's say. So,
2:53>> yeah, you've been teasing some things on
2:55Twitter that I'm I'm I'm very excited to
2:57see what you're doing there.
2:58>> Yeah, I mean, let me look at I've got
2:59the list of my Laracon topics pulled up.
3:02One, four, five. There's at least
3:04probably like 18 or 19 things. I'm
3:06probably going to have to prune it a
3:07little bit.
3:09>> Um,
3:10>> nobody's going to complain if you go
3:11long. I'm going to tell you that right
3:13now. I don't know. I assume. Are you
3:14like at the end of the first day like
3:15usual?
3:16>> Yeah.
3:16>> End of the first day.
3:18>> Yeah. If we have to go late because
3:19we're listening to Taylor talk about
3:21amazing new stuff.
3:22>> Yeah. Yeah.
3:23>> Maybe the venue will complain or
3:24something or whoever's running ops, but
3:26we will.
3:27>> I feel like last year's keynote was a
3:28little long, so I'm not going to try to
3:30go that long, but
3:31>> All right. Well, I I didn't notice it,
3:33but I know that when you you have a
3:35space book, they're very sticklery about
3:37kind of when you wrap things up, but
3:38from our perspective, give us the
3:40tailor, man. Like, okay. So, there is
3:43new and and I have actually tried to
3:45have multiple people from the team on
3:48the podcast and they've literally told
3:50me there's nothing I can share that will
3:52be even remotely interesting cuz
3:53everything I've worked on is top secret,
3:55but you can have me on after Laron if
3:57you want. So, I'm like, well, well,
3:59[laughter] damn.
4:00So, I'm very excited.
4:02>> Awesome. Okay. So, I've never been to
4:05Boston, but I know it's at the SOA power
4:07station. So, it's this like big kind of
4:08brick building that's historic and
4:10beautiful and everything, but I assume
4:12you've because you guys did your
4:13vacation then. Did you fly back and
4:15actually see the SOA power station in
4:16person once you guys picked it or have
4:18you never So, you never been to the
4:19>> seen it? No, I've never seen it. I think
4:21this is the first actually that uh I
4:24have not seen. I saw last year's in
4:26Denver. Um
4:27>> Uhhuh.
4:28>> Well, did I? I don't even know if I saw
4:29that one beforehand, but I definitely
4:31have not seen this one in Boston. Um, so
4:33I will be uh, you know, surprised
4:35myself.
4:36>> That's awesome.
4:36>> I actually probably also know the least
4:38about sort of just like what the stage
4:41decorations look like and things like
4:43that than any previous year. So, I've
4:45kind of looked
4:45>> How do you feel about that? Is that a
4:46piece? Is that a peaceful thing for you?
4:49>> Yeah, I'm fine with it. It feels good. I
4:51mean, it'll be fun to be surprised a
4:52little bit. Plus, um, the Laracon team,
4:55they've done a Laracon or two before, so
4:57they know like, you know, what to do.
5:00>> Yeah.
5:01>> Um, so
5:03>> this is not live as of the recording,
5:05but it will be live as of publishing.
5:08Uh, and you already know about this, but
5:09I built a thing called Voxicon. Anyone
5:11who hasn't seen it, just go check out
5:12Voxicon. V Ox A.Lar.cloud.
5:16Um, and one of the things that it is is
5:18a as accurate as I possibly could get in
5:20a Minecraft style uh thing
5:22representation of the the venue. So, I
5:25have been pouring over every blueprint,
5:29every picture, every video I could find
5:31of the SOA power station. So, it's crazy
5:33cuz one of the things that I'm very
5:35curious to see is when I get there, like
5:36what did I miss? You know, like what was
5:38not access? There's one thing I know for
5:40sure. There's no pictures of the kitchen
5:41in the entire internet. So, I have no
5:43idea what the kitchen looks like. Um,
5:44but outside of that, I feel like I have
5:46a more intimate understanding of the
5:48layout of the solar power station than
5:50any other person in the history of the
5:51world who has not been there in person
5:53because I've been obsessing over it.
5:54>> Yeah, that's funny.
5:56>> Okay, so you mentioned one of the things
5:57you're going to be introducing is
5:58Laravel and AI and I wanted to hear a
6:00little bit about um, you know, like we
6:02have been talking a lot publicly about
6:04how Laravel is an ideal uh, environment
6:07for somebody who wants to to work in AI
6:09and I think that's true for those of us
6:11who are already, you know, writing
6:13Laravel. you've got the AI SDK and stuff
6:15like that, but I think part of the kind
6:16of the case that we're making here is
6:18also that it's valuable for people who
6:20don't currently use Laravel. So, I live
6:23in Atlanta. There's a lot of people in
6:24Atlanta Tech Village and and you know,
6:26the surrounding tech world who are doing
6:28the startup world and they want to find
6:30a productive ecosystem for developing
6:33applications using AI ready to ship
6:36securely, stably as quickly as possible.
6:38Can you give the pitch for like what
6:41have you been working on with Laravel to
6:43make it the ideal environment for those
6:45types of folks?
6:46>> Yeah, I mean I think first of all I
6:50agree that like Laravel works really
6:52well paired with coding agents and I
6:54pretty much see that every day you know
6:56in my own work with Laravel.
6:58>> Um I think the foundation for a lot of
7:00the work we do around making Laravel
7:03great for coding agents is with Laravel
7:06Boost. um as far as like new package
7:09development goes. We've been working on
7:10some new boost features which I'll show
7:11at Laracon that make uh you know
7:14building Laravel apps even better. Uh
7:17but you know boost basically feeds a
7:18bunch of context about Laravel into the
7:20coding agent gives it access to the docs
7:22things like that.
7:23>> Yeah. Um I think we also just benefit a
7:26lot from Laravel's opinionated structure
7:28as we've said you know many times before
7:30where the agent sort of knows what to
7:32do. It knows how to access the database.
7:35it doesn't have to like oh you need to
7:36go sign up for superbase you know and
7:38link that into your project not that
7:41superbase isn't great but you know like
7:42the database is just available you know
7:44right there
7:45>> um and just all of the conventions you
7:48know around Laravel I think make it
7:49really
7:50>> just a productive stack for like hitting
7:52the ground running with minimal sort of
7:55like setup and decisions you have to
7:58make you can kind of just like start
7:59almost vibe coding right out of the box
8:01you know with like a complete package
8:04>> yeah um you know so I think that's like
8:06the main thing that makes it really
8:08great. We are working on a couple of
8:09other things um which I'll show at
8:11Laracon that I I think will make it even
8:14better. Some booster related, some not
8:16booster related. Um, but yeah, I mean I
8:19I'm actually continually surprised and
8:21during my prep for Laracon, it's been
8:23interesting because like historically
8:24one of the most timeconuming parts of
8:27prepping for Laracon was like prepping
8:29the demos themselves, especially if it
8:31had like a graphical component or was
8:33like a dashboard where I I needed to
8:35show off some feature. And now with like
8:38I was using boost in a fresh Laravel app
8:40and I was like, "Hey, I want to demo
8:41this feature at Laracon. Here's kind of
8:43like the docs for it. I need to build a
8:45dashboard that has these buttons and
8:46lets me interact with it in this way.
8:49>> And it's just like done, you know, in 5
8:51minutes. And it's basically oneshot it.
8:53And it actually used the um inertia
8:55react like boost skill to like know what
8:59>> and that just that would have taken me
9:01like all day pretty much, you know, to
9:03to build that kind of thing. And I I
9:05would have made mistakes and like had to
9:07fix stuff and hunt bugs down and like it
9:09just so much time saved.
9:12>> Wow. I mean that's you you've often said
9:14in the past that one of the things that
9:16makes you able to build a framework that
9:18is easy for newcomers is you constantly
9:20put yourself back in kind of this
9:22newcomer state and I mean I don't know
9:24about you but for me every time I you
9:26every time I hit Laravel new when it's
9:28been a few months I feel like I'm able
9:30to enter that newcomer state I'm like
9:31what what what where do I do that where
9:34is that and the easier that process is
9:36and to your point it's so much easier
9:37now than it has been in the past the
9:39more I'm just like okay like I can I can
9:41walk a newcomer in if I if my brain who
9:43has been maintaining legacy apps or
9:45doing marketing or whatever for the last
9:47year all of a sudden has to start from
9:48scratch. I'm getting some of that
9:50newcomer energy and I imagine these kind
9:51of like these prep experiences are like
9:53that for you.
9:54>> Yeah, definitely. And it's crazy how
9:57easy it is to get started, you know,
9:58these days just in programming in
9:59general, but especially with Laravel to
10:01where,
10:02>> you know, a couple years ago someone
10:03told me they wanted to like learn
10:05Laravel and start building apps, I would
10:07have been like, "Oh my gosh, like
10:08[laughter]
10:09>> this is going to be like, yeah, this is
10:11going to take years, you know, to like
10:12get up to speed." Uh, but now it's just
10:15like chat with a coding agent about it.
10:17If you get confused, just ask it a
10:19question
10:20>> and keep moving along. Well, I mean, you
10:22you and I both know uh Kayla Helmick,
10:24and she was featured on the the Laravel
10:26blog for anybody who's interested. We'll
10:28make sure that she's linked in the show
10:29notes. And she is has no history as a
10:31programmer, um but understands
10:33programming between being friends with
10:34you and and being my VA and stuff like
10:36that. And uh you know, she's built
10:39multiple functioning Laravel
10:41applications from scratch just using an
10:43agent. And she knows her bounds. I mean,
10:45not everybody does. So, one of the
10:46things she knows, she's like, I'm not
10:47going to do anything that takes credit
10:48cards. I'm not I'm not even doing
10:49anything that, you know, takes usernames
10:50and passwords, but still super
10:52functional tools for her for her um for
10:54her clients. And that's been really
10:57transformative to me to watch her
10:58experience cuz this is not theory. We're
11:00not just throwing out phrases, oh,
11:01batteries included this or whatever.
11:03It's like, no, we have a real person we
11:05know who doesn't understand a line of
11:06programming who has built functional
11:08applications that clients have paid her
11:10for. And I'm like, that's different.
11:13>> Yeah,
11:13>> that's really special.
11:15One thing I also notice, you know, with
11:16Laravel and AI is I'm starting to see
11:20or or maybe this is one reason I would
11:22say that Laravel is great for even new
11:25people that are interested in building
11:27software. I've been getting more and
11:29more sort of like a lot of it is just
11:31through real world friends, so-called
11:33real world outside of like Twitter, you
11:35know what I mean? Like people I know in
11:37Arkansas.
11:38>> Yeah. that are either referring to me,
11:40people they know because they know I do
11:42web development. And I'm seeing more and
11:44more people that are using sort of like
11:46vibe coding platforms that are getting
11:49like call it 50 or 60% maybe even higher
11:54of like on the way to their idea.
11:56>> Yeah.
11:57and they eventually hit sort of like a
12:00wall or a limitation in the platform or
12:03something or they run out of tokens or
12:05whatever
12:06>> and they're like I'm stuck and I think
12:09Laravel is really a great alternative to
12:11that because
12:12>> you don't really have the same ceilings
12:14you know like
12:16when you're using Laravel and Boost you
12:18can build even like multi-tenant
12:20multiple users um using the app you can
12:23just get a lot farther a lot cheaper
12:25usually honestly um by just by linking
12:28it to your like existing chat GBT
12:30account and using a coding agent and you
12:31just don't have a lot of the same
12:32barriers um
12:34>> that just people seem to inevitably hit
12:36you know on sort of like the prepackaged
12:39vibe coding platforms.
12:40>> Yeah. Well, and there's something fun
12:42about that too though is we have taken
12:44now several clients from a lovable or
12:47replet or whatever else um uh vision to
12:52a production scalable secure Laraveville
12:54application and we're able to do the
12:57final build for them faster than we
12:59would if they had not use those vibe
13:00coding platforms. So there's actually a
13:03way where I'm like regardless of whether
13:05you started with a lovable or not
13:07launching on Laravel still makes sense,
13:09right? Right? So you either start from
13:10Laravel on day one, that's totally fine.
13:12But like for example, we had a client
13:13who's just like he's like, I don't have
13:14the budget it would take for you guys to
13:16build this whole thing in in Laravel.
13:18>> Part of which is because I have not
13:20fully defined the thing. And I know that
13:21there's going to be a ton of back and
13:22forth and I'm not going to be a great
13:23client. But if I'm going back and forth
13:25with lovable for 6 months, then when I
13:27get the lovable to the point where it
13:28looks good, then I hand you the Lovable
13:30and you actually make it functional
13:32Laravel base. And we did and he launched
13:33and he's got paying subscribers and
13:35everything went great. I'm just like,
13:36and this has happened multiple times. So
13:37I'm just like Laravel should be the end
13:39result regardless of whether you started
13:41Laravel or started something else
13:43because there's no way that he could
13:44have done he could have hosted on
13:46lovable or replet or whatever else the
13:48level of complexity that he needed but
13:50we could build it in Laravel faster
13:51because lovable built the whole front
13:53end for him. So it's it's really fun to
13:55be able to say regardless of where you
13:57start this is where you should land.
13:58>> Yeah. Yep. And it's it's crazy how like
14:03using AI a big mindset from shift for me
14:06has been that like code is sort of like
14:07cheap in a way
14:09>> where like even if you build something
14:11unlovable or you start somewhere else
14:13and you're like
14:15>> oh I kind of wish I would have done this
14:16in Laravel.
14:18>> Yeah.
14:18>> You know previously that would have been
14:20like uh like this is hell. [laughter]
14:23>> Yeah.
14:24But with with with AI that kind of stuff
14:26like the code is not so
14:29>> intimidating or like you know those
14:30kinds of migrations are maybe a little
14:32bit more palatable because like it is so
14:34much faster
14:36>> to do with AI and like you know I've had
14:38to like train my mind to think that way
14:41where it's like oh I want to try some
14:43kind of quirky big idea.
14:46Previously that was a lot of commitment
14:48you know like
14:49>> do I really know that it's worth
14:50>> eight hours to exploring this
14:53>> but now if I can kick something off in
14:54five minutes and just get a sense of
14:56like if it's a good idea or not that's
14:58so much time saved.
14:59>> Yeah. Um have you found yourself in
15:02positions where you generate stuff that
15:05you're like feeling exhausted from
15:06having to maintain it or you're like no
15:08I just feel comfortable throwing it away
15:09it's no big deal
15:11>> in regards to like Laravel framework or
15:13what do you mean? Well, just any. So,
15:15what I found the same thing as you where
15:17it's I'm able to generate more
15:19experiments than I have in the past, but
15:21sometimes I'm like, ugh, I spent all
15:23this time all this time cuz really like
15:24I spent this time prompting this thing.
15:26I told people about it. I launched it
15:28and I'm like, you know what? This would
15:29not have stood the test of is this worth
15:32my time in the past and it wasn't worth
15:34my time. And I've just had to give
15:35myself permission to say that's fine. It
15:37was an experiment. I launched it and now
15:38I'm deleting it. You know, like, and I
15:39just tell people internally, I'm like,
15:40"Hey, you know, that internal tool I
15:42built, I'm going to get rid of it cuz it
15:43wasn't worth it." You know, have you
15:44ever had things where you're like, I've
15:46gone further down a road than I would
15:47have before cuz it is so cheap to do so.
15:50>> Oh, yeah. Totally. And I think it's part
15:52of the training my brain is that it's
15:53okay to delete it, you know, like
15:55>> I would have held on to it a lot harder.
15:57And I think my
15:58>> I want to sometimes because of, you
16:00know, it used to be that it required so
16:02much time investment in code, but now
16:04like
16:05>> with the code kind of being disposable
16:07in a sense,
16:08>> Yeah. like I'm more comfortable or I'm
16:10trying to become more comfortable saying
16:12h this didn't work out we can just trash
16:13it entirely you know
16:15>> yes which is a weird thing to learn
16:17because it would have been so wasteful
16:19in the past right and now you're like
16:20look it was a it was a thought
16:21experiment I it was pos valuable for me
16:24in this way I learned these things
16:26>> and that doesn't mean it has to be in
16:28production
16:29>> that's okay
16:30>> totally
16:30>> yeah um okay so I before we step off of
16:34AI there is one thing we use the phrase
16:37batteries included all the and and
16:38people a lot of people marketing Laravel
16:41out to the outside world talks about
16:42Laravel's value as a batteries included
16:44framework even before AI. Can you talk a
16:47little bit to somebody who's been
16:48hearing that phrase and isn't actually
16:49familiar with what you're talking about?
16:50What does that actually mean and how
16:52does that set Laravel apart from other
16:53frameworks?
16:54>> Yeah, it's a good question. Um when it
16:58comes to sort of like popular web
17:00frameworks in today's age there's kind
17:03of um
17:06I I mean the most popular framework you
17:09know in the JavaScript space which is
17:10Nex.js doesn't have a lot of opinions on
17:14how you do various things which is can
17:16be good or bad.
17:18>> So like it doesn't necessarily have
17:19strong opinions on how or where or with
17:22who you run a database. uh
17:25>> it doesn't necessarily have super strong
17:27opinions out of the box on how you do
17:30authentication and things like that or
17:32how you do ooth or social or things like
17:34that.
17:35So what it means is like if you were to
17:38start building an application like that
17:41or using other popular like frameworks
17:43in that space there's a lot of sort of
17:45like preddecisions you need to make that
17:47maybe you're not even really equipped to
17:49make
17:50>> u around like where what kind of
17:52database where's your off going to be
17:54how's it going to work how are you going
17:56to do like scheduled task how are you
17:59going to send emails
18:01um you know all of that stuff you're
18:02going to have to make decisions on or
18:04the agent's just going to maybe make
18:06decisions for you, which is probably
18:07likely the case.
18:09>> Um, and it can lead to a little bit of
18:12just like vendor sprawl where you're
18:15signing up for a lot of different
18:17services, you know, for across database,
18:20off email, um, maybe even cued jobs, you
18:26know, it's just spreading out quite a
18:27bit. Um, it's also going to be like
18:31there there's not necessarily a
18:33structure or an opinion on how the code
18:35is actually architected to do those
18:36things in many cases. So,
18:38>> it can lead to a little bit more of
18:40like,
18:41>> you know, we know AI is great, but it
18:43doesn't always really like architect the
18:45most like firm foundations for a
18:47longived project that's you're going to
18:49want to maintain for years and you need
18:51to be stable. um it it can build a demo
18:53really quickly, but as far as like
18:55longevity and maintaining over long time
18:57horizons still requires like good
19:00architecture,
19:01>> some a lot of times human input and kind
19:03of human review to make sure it's not
19:05going off the rails.
19:07>> Yeah.
19:07>> Um so with Laravel, a lot of those
19:09opinions are baked in. A lot of the even
19:11the structure is baked in of where
19:12certain files should go and kind of what
19:14should be in them and how they should
19:16all work together. And those opinions
19:17have sort of been formulated over many
19:20years of building web applications
19:22ourselves, but also just kind of
19:23learning from what came before Laravel,
19:26whether that was Ruby on Rails or other
19:28popular frameworks at the time, and sort
19:30of incorporating the best practices that
19:32generations of web developers have sort
19:34of distilled.
19:36>> Um, so that leads to sort of just like
19:39out of the box kind of more maintainable
19:41code that's going to just like survive
19:43longer and hold up better. That's not to
19:45say you're never going to need like
19:46human eyes on AI generated Laravel code
19:49because you probably will.
19:50>> Um but I think it's just going to get
19:52you further faster and without
19:54necessarily needing to result to signing
19:57up for a bunch of different vendors, you
19:58know.
19:59>> Yeah. Yeah. I mean with our clients
20:01these days, you know, the biggest
20:02complaint often was, "Oh, I have to sign
20:04up for Forge and Digital Ocean." So
20:05like, oh, well, those are pretty easy
20:07infrastructure changes. So again, I just
20:09sign up for Forge VPS or just sign up
20:11for um you know, cloud or whatever. Um,
20:14but it is definitely not a point where
20:15you don't have a managed cube provider
20:16and a separate database provider and all
20:18these other things. Um, but I also
20:20remember onboarding some JavaScript
20:22developers into Laravel.
20:24>> I don't think it was 10 years ago, but
20:26quite quite a while ago and they were
20:28they were a little bit overwhelmed
20:29because they said there's so many files
20:30and directories. And in some ways, you
20:31know, you've worked on that to say
20:32there's less off the bat.
20:34>> But part of what I had to tell them was
20:36part of the reason there's more here is
20:37because it does more things, right? Like
20:39what you're seeing is stuff that's
20:40actually functional out of the box. And
20:42yeah, there might be more things here,
20:43but it's because you don't have to build
20:44those things. And bike shedding is a
20:46real problem, man. Just like, oh, we
20:48have to start something now. Let's spend
20:49the next six months deciding and and on
20:52every app you've ever built, it's
20:53different. So, good luck remembering how
20:55to fix it and having, you know, and now
20:56have good luck having the AI know how it
20:58works because first thing it has to do
20:59is research which of 17 different
21:01situations you set up with versus with
21:03Larl, it's like, oh, I know I know
21:05you're using eloquent, so it's going to
21:06be eloquent, so that's it.
21:08>> Yeah.
21:08>> Yeah. For many developers, the bike
21:10shutting part is the part they enjoy the
21:11most, you know. [laughter]
21:13>> Yeah. So, let's talk about that because
21:15I I've you just mentioned something
21:17about code being kind of uh throwaway,
21:20but you're also the the eloquent artisan
21:23guy, right? Like you you know, origin
21:25story of Laravel is, you know, what does
21:28it make sense if I pay extra attention
21:30to my you know, my lines and everything
21:31like that? So, how do you how do you
21:34feel about saying that? Is it some are
21:36you disappointed in any way? because I
21:37know we've had a lot of conversations
21:38online about like some people it was
21:40always about writing the code and some
21:41people it's about building the product
21:42but you've always been both. So what
21:44does this new world feel like for you?
21:46>> I think it is it's a great question. I
21:49think it is a very like interesting time
21:51where like you know historically Laravel
21:55and myself have taken a lot of pride in
21:57sort of crafting these sort of really
22:00elegant APIs and interfaces for doing
22:03different things whether that's just
22:05being able to say like
22:06>> Q colon later you know to delay a job or
22:10something like very concise very
22:12memorable syntax that sort of is a
22:14pleasure to read. Um, I still do that
22:18for what it's worth on all of the open-
22:20source stuff. So,
22:22>> we just worked on a really big feature
22:24for the AI SDK, which I'll show at
22:26Laracon
22:27>> and um, you know, a lot of the code is
22:29like AI assisted and generated, but I
22:33still reviewed every file. I still sort
22:36of like sweat the details on how the
22:38user is going to interact with this API,
22:42even if
22:43>> no user ever does. Maybe only AI
22:46generates this code, but like I still
22:48spend the time thinking about like okay
22:49let's assume that users are going to be
22:51interacting with this like what is the
22:53most like intuitive
22:55>> joyful to write API. Maybe that's like a
22:58lost cause,
22:59>> but I still kind of derive like personal
23:02satisfaction from that. I think a lot of
23:04developers actually probably still do
23:06like even though the AI is assisting
23:08you, you still find joy in sort of
23:11finding just the right abstraction or
23:14just the right name for something. That
23:16is still like a really fun for me. And
23:19again, I still just derive personal
23:20satisfaction from knowing that sort of
23:22like you're building like a dresser,
23:24right? And the backside of it is still
23:26like finished perfectly even if no one
23:28ever sees it because you know and you
23:30know that the tool is like good and well
23:33built. Yeah. And for me, like I still
23:35don't want to use tools that or I I
23:38wouldn't really be comfortable using
23:40Laravel if I didn't know that the whole
23:42thing was thought through, you know, all
23:44the way through.
23:46>> Um, so I I still do that and maybe that
23:49will become sort of an outdated mindset
23:51over time. It's possible, right, as AI
23:54keeps getting better. But for now, I
23:56still find it personally satisfying
23:58>> and I I it's one that I hope that you
24:00continue to take the luxury to do. uh
24:02because a I think it's just good and
24:04beautiful to do that, but b um I'm still
24:07one of those sticklers who says even if
24:08I'm using AI, I'm going to read every
24:10single line of code and understand every
24:11single line of code. And I've found a
24:14draft drastic difference as as I have
24:16been using AI to delve into worlds
24:19outside of Laravel that I wouldn't have
24:21done as much in the past. I find a
24:22drastic difference in the ease with
24:24which I can comprehend and review
24:27Laravel code and code outside the
24:29Laravel ecosystem. So like even if the
24:31AI is writing it, if I'm reading it,
24:33especially if I want to have to skim it
24:35because whatever, like I can it's so
24:37expressive that it's just very clear
24:39what the Laravel code should be doing.
24:40And that's not the case everywhere else.
24:42So if there's a human in the loop, even
24:46if the human
24:49friendly coding is going to if I were
24:52look at two two sets of code that were
24:54spit out by, you know, an AI, uh I'm
24:57much more likely to comprehend and
24:58therefore continuing to choose to read
25:01and review and comprehend and take
25:02ownership over the Laravel code than
25:04something else where I'm like, I don't
25:05understand this crap, whatever. And once
25:07you hit that whatever point, you've lost
25:10your ability to you're never going to go
25:11back and learn it more once that hit you
25:13first hit your whatever. Right. Like
25:14it's gone.
25:15>> It's worse. Yeah.
25:15>> Yes. Exactly. So
25:17>> yeah, I think it's beautiful.
25:18>> Every line that has gone into the Laval
25:21framework, you know, in the last what is
25:23it maybe seven or eight months since
25:24sort of agentic coding really exploded.
25:27I've still reviewed every line that has
25:29gone into the framework by hand
25:31essentially.
25:32>> Yeah. Yeah. I love it. So I want to
25:35thank some of our sponsors. First, I
25:36want to thank our sponsor, Laravel, the
25:38wonderful folks behind Laravel Cloud and
25:40Larcon US. Laravel Cloud is the fastest
25:43way to deploy Laravel apps. And it also
25:45solves all sorts of common problems that
25:47keep web artisans from shipping. And
25:48Laron US is their premier in-person
25:50event for all web artisans to come be
25:52with your people, level up, and leave
25:55ready to ship better and faster than
25:57ever. Check these out at cloud.lar.com.
25:59Larcon us. Taylor and I of course will
26:02both be speaking at Larcon. We'll both
26:04be there in person and we are talking uh
26:06both in our talks about Larbo Cloud. So
26:09definitely check it out. And also
26:10Laracon Australia where I also will be I
26:12don't know about Taylor Australia
26:14returns November 4th to 6th in Brisbane.
26:17I love Brisbane with deeper learning,
26:20practical insights and ideas that will
26:21stretch how you think about building
26:22with Laravel. So get your tickets now at
26:24larcon.auu.
26:27Um so you know one of the questions I
26:29wanted to ask was kind of what's the
26:30future of programming? So let's let's go
26:32there cuz I think it's a pretty natural
26:34transition as you think about the future
26:36of programming. You have built tools for
26:39programmers your the majority of your
26:40adult life. You built you know open
26:42source tools. You've built paid tools.
26:44You run a company uh building you know
26:47dev tooling and also providing open
26:49source tooling for developers. Uh what
26:52do you think the future is of
26:54programming? And I mean that in two ways
26:56and you can add more ways but the two I
26:58mean for sure are the the future of the
27:01the role that we as programmers have. Um
27:04and also the future of the way that
27:06other people outside of us are doing
27:07programming you know so you know for us
27:09is are we going to be writing code? Are
27:11we all becoming product owners? Are you
27:13know we becoming solutions architects
27:14instead of hand hand coders? Are other
27:17people outside of us now going to be
27:18doing a lot more code? like as you look
27:20to, you know, a couple years from now,
27:22what is your sense of where it's heading
27:23or do you know any more than anybody
27:25else?
27:27I sort of suspect that we're living in
27:29it now um to a large extent and what I
27:34mean is that like
27:36I think that the main shift has actually
27:38already happened um in the sense that
27:42many of us are actually not writing the
27:44majority of the code by hand, maybe none
27:46of the code by hand. Um, I think that
27:49was like the big thing. I think like
27:52what remains maybe like to come or less
27:55clear is how much of the review will we
27:58be doing versus like trusting the AI.
28:01And I think people are already at sort
28:03of like various levels with this, you
28:05know, in different places. So like I
28:07actually just went to Anthropic's office
28:09for a founder day um that they did and
28:14got to hear from a few of the people
28:15working um particularly on Claude code
28:18uh but someone kind of on the research
28:20side of anthropics model stuff as well
28:23and I think they are very like high
28:25trust of the model. they're obviously
28:28using like the model um before we get it
28:31and at a faster speed than we get it and
28:33maybe with higher effort, you know, and
28:34less token limits,
28:36>> but they seem to have like very high
28:38trust in the model. And I think their
28:40their view that I gathered out there was
28:42that,
28:43>> you know, the future is not really
28:46reviewing the code. It's more of a
28:49um agent manager. you're sort of like
28:53guiding the agent or steering it in
28:55terms of the overall solution, but
28:57you're not reviewing every line of
28:58syntax. Um, they were also very like
29:01goal or loop focused of, you know,
29:04kicking off longunning goals with the
29:06agent that are maybe running for hours
29:08at a time to do big projects.
29:12But um, yeah, I mean I mean to come back
29:14to the question, I think the the main
29:15shift has happened in the sense that
29:17we're not code writers. Um, I think that
29:20is like already done.
29:22It's just a matter of like how much
29:24trust are we going to place in the agent
29:26as far as the review and overall
29:27architecture.
29:29>> And I think it's hard to say. I mean, if
29:30we if we keep going projecting out
29:32years,
29:34you could imagine with model
29:35improvements. I mean, you could imagine
29:37a world where the code is trustworthy,
29:39you know, like it's not science fiction,
29:42I don't think, to to think about it a
29:44few years out to think that, well,
29:46>> maybe it does become sort of like a
29:48waste of time to review all of the code.
29:51Um,
29:52>> but I, you know, I I don't think we're
29:54there yet, but again, you could imagine
29:56a world
29:57>> and at that point um
29:59>> I don't think it actually looks all that
30:01fundamentally different from what we do
30:03now. You know, I I think we're still
30:05going to have to have a basic grasp of
30:07like the technologies available, how
30:10they work, what the upsides and
30:11downsides are maybe to different
30:14databases or to different strategies for
30:16building things or different overall
30:17architectures, especially at scale. Um,
30:22but yeah, I, you know, I think, um, I
30:25don't know that that's the best I can
30:27like kind of state it, I guess, from
30:28where we are now.
30:29>> Yeah. Well, I don't I don't know if I've
30:31heard anybody say that before that that
30:33that we're there already. Uh,
30:35everybody's got visions whether positive
30:37or negative, but uh I I'm actually
30:39really heartened if that's the case
30:42because so everybody at Titan still
30:44reads every single line of code and a
30:46large portion of the code written at
30:47Titan is handwritten. Uh I have told
30:50everybody you need to be an expert in
30:51AI. Like that's baseline expectation. So
30:53they're all experts in AI. Uh but they
30:55use it to differing degrees. they use,
30:57you know, different models, although
30:58cloud code is the most common. Um, but I
31:02have noticed that in the agency world at
31:05least, um, the people who are doing the
31:07best are people like us where we are
31:10here for our expertise. Like we're here
31:11because we understand the architectures,
31:13the systems, the decisions to make.
31:14We're here because we've seen thousands
31:16of code bases and because we know what's
31:17going to scale and what's going to be
31:18secure. And those things are implemented
31:21through us actually typing characters
31:23into an IDE. But the characters in the
31:26IDE is not really what Titan brings
31:29uniquely compared to somebody else. It's
31:30just an implementation detail of our
31:32expertise. So the implementation
31:35shifting doesn't change the fact that
31:36you hire Titan for expertise, right?
31:38Whereas with people where your job is to
31:41have an expert tell you now go type
31:44characters into an IDE. Those are the
31:46folks that are having the hardest time.
31:48And my hope is we're able to keep a
31:51pipeline where people can be in that
31:53stage and still be able to move up.
31:54Otherwise, we're never going to have any
31:56a junior pipeline anymore, right? So,
31:57there's definitely things to consider
31:59about there. But my hope is that we can
32:01find spaces for people as they are going
32:03through a junior pipeline to be learning
32:05maybe earlier how to be expert thinkers
32:08rather than spending their first 5 years
32:10just learning which characters go into
32:11what state in the IDE. So,
32:13>> yeah, I think that makes a lot of sense.
32:16>> Okay. I think a lot of people I mean I
32:18think a lot of people are also like
32:21>> my AI coding setup is I think much more
32:23vanilla than a lot of people
32:26>> are thinking the future of AI coding
32:27might look like.
32:29>> I think I think Dax I think Dax at Open
32:31Code had a great tweet about this a few
32:33days ago.
32:34>> I mean my my um setup is just like
32:37either cloud code or open code and I
32:39type into it and code comes out the
32:41other end.
32:42>> Yeah.
32:42>> And there's no like advanced
32:44orchestration layer. Yeah,
32:46>> beyond that rather just whatever is
32:49built into the vanilla setup is sort of
32:51what I use. Yeah,
32:53>> I think a lot of time is being spent
32:55sort of like trying to I think this is
32:57very like developerrained by the way
32:59like trying to like come up with these
33:01more like fancy and efficient ways to
33:05like have tons of things going at once
33:07and like it's a big software factory
33:09black box and in a lot of ways I feel
33:11like anthropic actually does operate
33:13this way which is maybe like a
33:14counterpoint to how I see it but I just
33:17see I just like think a lot of that is
33:20sort of
33:21another flavor of I'm going to build my
33:24own framework.
33:26>> Yeah.
33:26>> It's that classic like developer brain
33:29of like I'm the special snowflake that
33:31wants this special setup and I need all
33:33of this abstraction around it to make it
33:36[gasps]
33:36>> work well instead of just like using the
33:39tools like to do something useful, you
33:41know. [laughter]
33:42>> Yes. Like I think that that is what I
33:45mean by I think the fundamental shift in
33:47software is like with AI sort of already
33:49we're living in the future of it
33:52>> of just that was the fundamental shift
33:54is having like this terminal or text box
33:57that I can type into a plain English
34:00thought
34:00>> and code comes out the other end and
34:02multiple files are changed and things
34:04are done for me on my computer. that
34:07that was like the futuristic shift and
34:10um you know I think we're already there
34:12in that sense.
34:13>> Yeah. And I I love that my other
34:15podcast, Pragmatic AI, I allow myself to
34:17get a lot more into like what's
34:18happening this week and we just had John
34:20Oolan on and John was talking and John
34:23has always been one of my friends who is
34:25a building real things and b at the
34:28forefront of AI. I feel like there's
34:30very few people who are both of those.
34:31you're either at the forefront of AI and
34:32you spend all your spare time building
34:34yet another orchestrator, you know, or
34:36you're building real things and you're
34:37like, "Yeah, AI is helpful in some ways,
34:39but you're not like
34:40>> again trying to bleeding edge spend all
34:42your time and energy on orchestrators."
34:44Um, so but it was he and he mentioned
34:46goals, you know, he mentioned goals as a
34:48thing that he's using and I hadn't even
34:49been switching to goals. So I I I
34:51appreciate the fact that I don't have to
34:53be in the latest greatest to feel really
34:56productive. um
34:59in part because the people who are and
35:01that's one of the reasons why I'm just
35:02like I'm perfectly fine if I'm not doing
35:04the same thing as anthropic because the
35:06people who are at the absolute forefront
35:08they will spin through things in two
35:12weeks and then I would be like yep you
35:14you you were hype about it you talked
35:15about it non-stop for two weeks and then
35:17it never ever came up again. Let me see
35:19something that surfaces and actually
35:20sticks around for three months and then
35:22maybe we'll talk about it.
35:24>> It remind it's a little bit reminiscent
35:25also of like
35:27the you know like when Facebook was
35:29popping off in that era of um startups
35:32and there were microervices and things
35:34like that where like yeah
35:35>> the individual devs or maybe the two or
35:37three person teams were like splitting
35:39their app with a 100 users into five
35:41microervices
35:43>> and it's Facebook do it
35:45>> exactly because these sort of like
35:47hyperscaled startups were doing it there
35:49is maybe a little bit of a similar thing
35:50in AI where
35:52>> you know you would be better served by
35:54just popping open the agent and typing
35:57something into it and not spending so
35:59much time sort of orchestrating around
36:01it.
36:02>> Yeah. Or maybe open up a second tab with
36:04second agent or use solo. So you get,
36:06you know, like
36:07>> shout out solo.
36:08>> Yeah, I love solo. The the most
36:10complicated thing I do is if I want to
36:13have distinct sub agents, I've found
36:16using Solo to do it is a little bit
36:18easier than having the agent manage it
36:19itself. And that's it. Yeah. Outside of
36:21that, I'm just sort of like, yeah, I've
36:22got an agent doing this, an agent doing
36:24that, I'm doing this. That's enough.
36:27>> Yeah, totally.
36:28>> And to thank some more sponsors. Move
36:29fast and fix things with Honeybadger's
36:31intelligent error tracking and
36:32application monitoring for Laravel. It's
36:35everything you need to keep production
36:36healthy, users happy. Get your free
36:38developer account at honeybadger.io.
36:41And also, this is not an official
36:42sponsorship, but I do want to let you
36:44all know that at Laracon US next week,
36:46our friends at Geocodio and I or and we
36:49we Titan and Geocodio have partnered
36:52together to give away 500 free ice cream
36:55sandwiches from a local Boston shop. You
36:57just need to come by both of our booths
37:00and get a little thing stamped, I think
37:02it's going to be. And if you take the
37:04double stamp thing over, you will get a
37:05free ice cream sandwich if you're in the
37:07first 500 people. So, if you are coming
37:09to Lacon US, please make sure to come by
37:11both of our booths and get those stamps
37:13and get some delightful free ice cream.
37:16So, one thing that has been really fun
37:18over the last maybe two weeks is your
37:21Twitter account. Um, and uh Spicy Taylor
37:25uh has lots of opinions and it's been
37:27really cool because I feel like uh when
37:31everything happened with uh the
37:33investment and Excel and stuff like
37:34that, you you just had so much to
37:36manage, right? you went from having a
37:38team of like 10 to a team of nearly 100.
37:40And obviously you don't manage every
37:41single person, but as somebody who was
37:43really active on the internet and then I
37:45took over as the CEO at Titan and then
37:47it was really hard to be active the
37:48internet because I'm suddenly doing all
37:50what I was doing before and now fully
37:51running a company. It's I've only been
37:54I've been pretty slow coming back and so
37:55I've been watching you come back and I'm
37:56just like oh like we've got Taylor
37:58again. If somebody's not on Twitter,
38:02I could describe it but I'm curious for
38:03you like did you just feel like was
38:05there just a moment where you felt free?
38:06Did something change? Like what what is
38:09leading us to the point where you're all
38:10you're thought leading, you're sharing
38:12funny stuff and I just feel like we have
38:14old Taylor back.
38:15>> Yeah. I don't I don't know. I I think
38:17like I'm still trying to distill it
38:19myself. I think like the last couple
38:21years have been a lot of like learning
38:24and growth and lessons and kind of like
38:26reflections. I do think, you know,
38:31maybe I'm coming to a place where like
38:33some things are starting to click for
38:34me. um or I'm getting comfortable with
38:36some things.
38:37>> Yeah.
38:37>> You know, but I mean just to go back.
38:38Yeah. Like Laravel has grown a lot. Um
38:41when we first raced a series A in 2024,
38:45like you said, we were like eight or
38:46nine people. Uh mostly devs.
38:51We brought on Tom and Andre as kind of
38:54our first like couple of hires. Um and
38:58we started growing the company. And you
39:01know those foundational people had a lot
39:04of great experience that I didn't have.
39:07Um and to some extent still don't have.
39:10Um I
39:14you know I am not I actually I don't
39:17know that I really manage anyone at
39:18Laravel even today. I don't think I'm
39:21actually a good like manager in the
39:23traditional sense. I collaborate a lot
39:26with people um
39:28>> almost a little bit as like a peer even
39:30though I'm I am like their boss you know
39:33to a certain extent and we're not
39:35totally peers at the same level in that
39:37way but I prefer to work in this like
39:38very collaborative way.
39:40>> Yeah.
39:40>> Um so like I work a lot with like Joe
39:44Dixon in in product. Um where we're just
39:48kind of thinking about like what are the
39:49projects we want to work on next? where
39:51where are we kind of heading like long
39:53term? Um and even that I'm I don't I
39:55prefer not to think like in terms of
39:57multiple years. I think that's really
39:58silly in tech. Um if we would have tried
40:01to do that a couple years ago, it would
40:02have all been useless.
40:03>> Yeah.
40:04>> Um but I try to at least think of a few
40:06quarters, you know, of like where where
40:08do we want to go and kind of what are
40:09the like
40:10>> vague uh goals we want to hit, you know,
40:13in the next year or two.
40:15Um, I spent so I spent a lot of time
40:17working on that in open source and then
40:19just sort of like collaborating across
40:20the business. You know, it could be like
40:22in a marketing thing for Laracon. It
40:24could be on a website thing.
40:26>> It could be even on an email we're
40:27sending and like what the copy should
40:29be.
40:30>> Um, it could be looking at something for
40:33one of our private cloud customers. Um,
40:36something's not working quite right in
40:37the framework itself maybe.
40:39>> Yeah. Um, but you know, like regarding
40:43sort of like the the the the reversion
40:46to sort of like old Taylor on Twitter, I
40:48think that like for me
40:51maybe one of the biggest regrets I have
40:55over the last couple years, and not to
40:56say it was like a huge mistake or it was
40:58a big regret, was like not leaning into
41:02myself enough or like not like really
41:04like
41:06>> I it's hard to say it without sounding
41:08like proud are sort of like conceited
41:11but not sort of like just
41:13>> owning that I have actually been
41:14successful at at a software business um
41:18in ways that other people at the company
41:20maybe haven't.
41:21>> Um
41:22>> yeah,
41:23>> I've actually, you know, opened up an
41:24editor totally blank
41:26>> and convinced hundreds of thousands of
41:28people to use this tool and the business
41:31millions of dollars, right?
41:32>> Yeah.
41:33>> Um which is just like rare. Um, so I do
41:36think that um I have a very like
41:39particular opinionated way of doing
41:41things, right? The spicy takes like you
41:42can do a lot of things in one day or
41:44like why would you ever start with like
41:46a longer time frame or you could run the
41:48whole company out of 10 notion docs and
41:5010 Slack channels.
41:51>> Yeah,
41:52>> I think that um I am very that is very
41:56true to me as a person like I am very
41:57radical that way. So there is like so
42:00much I feel what I'm saying is like
42:02actually true. Yeah. [laughter] Um, and
42:05just sort of like leaning into it a
42:07little bit more of like
42:08>> what if I think what I'm trying to say
42:11is when we grew the company, we brought
42:13out a lot of really talented people in a
42:15lot of very various areas
42:18>> and I kind of like in some ways I feel
42:21like retreated mentally of like these
42:24people know these people are
42:26professionals.
42:27>> Yeah.
42:28>> I am was more of like a hobbyist that
42:30got lucky in many ways. [laughter] You
42:32know what I mean?
42:33>> Yeah. I'm trying to get a get away from
42:35that a little bit to where like these
42:37people, it is true, these people are
42:38professionals,
42:40>> but I actually have done a lot of crazy
42:42[ __ ] too. They're very successful
42:45>> and um should lean into those opinions a
42:47little bit more because I think like
42:48what I bring to the table is like a nice
42:51counterbalance to um what other people
42:54bring to the table. A little bit more
42:56like crazy, a little bit more like
42:58swashbuckling, you know? Like
43:00>> just ship something now. Like what if we
43:02just shipped it now? like what if we
43:03just hit the button right now?
43:04>> Like what would really happen? Is it
43:06really that scary, you know, like um so
43:09trying to push on that a little bit, you
43:10know, push the pace a little bit, push
43:12the energy a little bit.
43:14>> Um you know, which I think is like what
43:17I can bring to the table. Um and I think
43:19other people at the company bring
43:20amazing things to the table, but I think
43:22we all need to bring our unique selves
43:25to the table. And I think I was trying
43:27to be a little bit more like blending
43:30in, you know, with what everyone else
43:31brought to the table. And that's really
43:33probably not what the company needs from
43:34me actually. You know, we already have
43:35that.
43:36>> Yeah.
43:37>> And we need me to be me. So it was me
43:39trying to like a lot of it's me
43:40preaching to myself, you know, of like I
43:43need to be
43:43>> No, this is true. pushing the pace, you
43:45know, it is so easy like when you're
43:47building a software company like
43:51>> um especially remotely I think
43:54>> it's hard to feel the like intensity and
43:56the energy when you're all remote and so
43:59like I do want to be someone at the
44:01company that sort of is like always sort
44:04of like you know like okay what are we
44:06working on this week like what are we
44:08trying to like bring excitement and
44:09energy to the
44:11>> to the space cuz I mean startups is a
44:13competitive world you know like there's
44:14companies out there that are locked in
44:16like right now, you know, and
44:17>> I used to think about this back when I
44:19started Laravel.
44:20>> Um I would absolutely not be outworked
44:25by anyone for sure. I knew that um you
44:27know back when I first started Laravel
44:29in 2011 2012 I was like okay well
44:32>> they may be like smarter than me or like
44:34more like advanced than me but like
44:37>> I'm working going to work harder than
44:38them and like in a on a long time scale
44:41it's just going to compound
44:43>> and I'm going to beat them you know just
44:45out of sheer like grit and
44:47determination.
44:48Um, and but that's that can be
44:51challenging as the company gets larger,
44:52it's remote, like
44:54>> it's hard to maintain. So, I try to, you
44:56know, bring that to bear a little bit.
44:58>> Yeah. Well, it's been really fun and I I
45:00I I love what you've described here. It
45:02absolutely makes sense of a you are a
45:05self-aware uh you know you can read the
45:09room you understand people and so as you
45:10bring in a whole bunch of people who
45:11especially like a lot of people coming
45:13from the existing startup world a lot of
45:15people come from the existing you know
45:17enterprise corporate world people who do
45:19things differently than you and there is
45:21a type of founder who would come in and
45:22say you all are going to work my way you
45:26are going to it's shaped to be like me
45:28you're at my company now uh and then
45:30there's a founder like you who says I
45:32want to learn from you and I understand
45:33and respect what you're bringing and all
45:35those things. Um but in doing so you can
45:37lose yourself a little bit and that's um
45:39that's something I'm very familiar with
45:41having done in my personal life and in
45:42my professional life. Um, and you hit a
45:45point where you're like, "Wait a minute,
45:47you know, like I've learned, I've
45:48listened, and I've also squashed
45:49myself." And I want to to to say that I
45:53not to say to disrespect other people's
45:55opinions or to say they were wrong, but
45:57to say like there's also needs to be
45:58space for me. And I have been so good at
46:01deferring and listening and considering
46:02others that I forgot to be myself, too.
46:04And so I'm like, "Oh, I love you doing
46:06this. This is so good for my friend and
46:08it's so good for our community."
46:10>> I feel that, too. And I think it it
46:12really I think what you said really hits
46:14home for me too and it goes outside of
46:16Laravel like just I am a I as a person
46:19am very like very like compromising and
46:22willing to like go with the flow and not
46:24really that hard to get along with you
46:26know not really polarizing like in
46:27person I mean like you know just in my
46:30interactions with friends and family and
46:31stuff. Um, but yeah, you can kind of
46:33like squash yourself so much
46:36>> that like it's actually a disservice to
46:39the to the company and to the people you
46:41want to be working with. It feels like,
46:43oh, I'm kind of like
46:45>> I I'll be this person that just sort of
46:46gives everyone space to do their thing
46:48and not really bring any particular
46:50flavor to bear. And I think that's, you
46:52know, that's not good either. And uh
46:55>> you know, so that's what I'm trying to
46:56correct for a little bit is like
46:58>> be a little bit like crazier, be a
47:00little bit more opinionated, be like a
47:02little bit more genuinely me.
47:04>> Yes.
47:05>> Um because I think that's just another
47:08um that's another, you know, um
47:12ingredient in the recipe of Laravel that
47:14is needed. And without that, um it I
47:17think it is like missing something.
47:18>> Yeah, I couldn't agree more. I'm I'm
47:20very happy for it. Um, I know we're just
47:23about at time, so I did want to kind of
47:25end with just a little bit more talking
47:26about you. Um, I get to know you as your
47:29friend and so I get to kind of like see
47:31what you're doing in your life, but I
47:32know everybody doesn't. What is
47:33something that you've done outside of
47:35work recently that was really, you know,
47:37did you guys travel somewhere or have
47:39you taken up hobbies or have you gone is
47:40there something you or you and the
47:41family or you and Abby have done
47:42recently that it's just like, oh man,
47:44this was so great. I want everybody
47:45should do this or I want to share about
47:46it or anything.
47:47>> Oh man, you know, it's been a busy
47:49summer. Um well we did all go to Tokyo
47:52and this is not exactly workrelated. I
47:55was there for Larave Alive Japan but we
47:56all went we had to turn it into like a
47:58family vacation. It was very different
48:00actually than all of the um
48:02>> most of the other vacations we've done.
48:04Tokyo's massive by the way. [laughter]
48:06>> It made like New York City feel small.
48:09>> But it was it was fun for the the kids
48:11to like
48:12>> go to Asia in general. um you know I had
48:16been to India and and other places but
48:18um they had never been and just all hang
48:20out in such a huge city that is is quite
48:23different you know than America
48:25>> um and that was a lot of fun.
48:28>> Uh my son James, he's been fishing a
48:30ton. So we've been chatting like fishing
48:32strategies, baits. He went to a fishing
48:36store in Tokyo because they like make a
48:38special kind of like bass baits, which I
48:40didn't know that fishing was um
48:42especially like bass fishing and stuff
48:43like that was a big thing in Japan, but
48:45apparently it is.
48:47>> Um so we uh packed up the suitcase with
48:49a bunch of baits and brought them back
48:51to the
48:53>> And do you fish? I feel like he's got
48:55like an uncle he fishes right with,
48:56right? Did I make that up? Yeah, I I can
48:58fish, but um he fishes a lot honestly
49:00with his friends and then uh my uncle
49:03has a boat and takes him out, but um
49:05yeah, fishing has been a big thing for
49:06our family for the last year or so.
49:09>> Um yeah, but yes.
49:11>> Okay. And and the kids I I I know we try
49:13not to say exact ages, but I feel like
49:15the kids are, you know, late middle
49:17school, early high schoolish ages, so
49:19you guys are staring down the barrel of
49:20empty nesting.
49:22>> Have you thought about that at all? Is
49:23there any big changes coming in the
49:24future? Like we'll get to it when we get
49:26to it. I I can't say we've like cracked
49:28it on what it's going to look like, but
49:30definitely we've talked about it
49:31because, you know, it's like coming up
49:33relatively soon, say in the next like
49:35seven or eight years.
49:36>> Um we're kind of we're kind of there. I
49:38don't know what it will look like for
49:40us, you know, like um I grew up here in
49:43Arkansas, but my wife
49:45>> did not. Um
49:48>> you know, I my my whole like personal
49:50situation looks so much different than I
49:52thought it would in general. Yeah.
49:54>> And I think this I mean this goes back
49:55to trying to predict things years in
49:56advance, right? It's just so like
49:58impossible.
49:59Um you know my my parents have both
50:02passed away. Didn't expect that. Um that
50:05makes you know what is what where do we
50:08want to be you know in the future? It's
50:09a lot more like open-ended now like I
50:11you know um so we'll see. Um
50:13>> yeah. Okay.
50:15>> Uh I think I'll still be tinkering
50:17around on tech stuff. I'll probably
50:18still be tinkering around on Laravel
50:20even [laughter] at that point. hopefully
50:22if the company's doing well.
50:23>> Um, so, you know, we'll see. But I I
50:26think it'll be fun. Um, because there's
50:27just so much. It's so open-ended. You
50:29know, there's
50:29>> Yeah,
50:30>> we'll see where the kids end up and
50:31we'll we'll figure it out.
50:32>> I love it. Um, is there anything you
50:36want to share or anything we didn't get
50:37a chance to talk about today you'd want
50:39us to cover?
50:40>> Oh, well, definitely stay tuned for
50:41Laron. There's going to be a bunch of
50:42stuff. We may have to do a follow-up
50:44recap. I love it.
50:46>> But, um, yeah, there's going to be a
50:47bunch of stuff um, coming out. I'm
50:49really excited about it. And I think um
50:51a lot of it is stuff I've been excited
50:53about for a long time. And Larona is
50:55always kind of this moment to like, you
50:58know, take like a deep breath. Finally,
50:59like all of these things are out and we
51:01can talk about them and there's no like
51:03secret projects for preparing behind the
51:05scenes or at least we get to start fresh
51:07doing that again.
51:08>> Yeah.
51:09>> Um I can either steal you away at Larcon
51:12or get you the week afterwards so we can
51:13do a follow-up.
51:14>> Yeah.
51:15>> Sorry, but it'd be fun. I'm proud of the
51:17team. you know, there's been a lot of
51:18like heavy technical lifting and work
51:20that's gone into a lot of this stuff and
51:23um you know, but that's what really
51:24makes the products, you know, great is
51:28um tackling these sort of hard problems
51:30that are very hard to solve on your own
51:32with your own sort of like small team or
51:34DevOps team.
51:36>> These are pretty complicated spaces that
51:37we're playing in now, but that's what
51:39really gives the product like value and
51:41um differentiates it, you know, from
51:43what's out there. So, I'm excited.
51:45>> I love it. Well, thank you so much for
51:47your time, especially I know that you're
51:48super super busy this week. Uh I will
51:51see you next week. A lot of us will see
51:52you next week and then I'll I'll try and
51:54catch up with you a little bit
51:55afterwards. So, look forward to um so we
51:57got this Taylor episode. We've got a uh
51:59episode coming after that and then after
52:00that it'll be one more Taylor before we
52:02wrap the season and we can do a full
52:04debrief of everything you announced at
52:05the at the conference.
52:06>> Nice. Yeah, it' be fun. We can uh talk
52:08about everything.
52:09>> I love it. Well, thank you for hanging
52:10out and for the rest of you, we'll see
52:12youall next time.
52:14>> [music]

About this transcript

This page contains the full transcript of Taylor Otwell on Why Laravel + AI Pair so Well, and the Future of Programming by Laravel Podcast, generated from the public captions YouTube serves with the video. The transcript has 11,141 words across 1,585 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.