laravel make:podcast
Yo. What's up? Welcome to the Slightly Caffeinated podcast. I'm TJ.
Chris:And I'm Chris Gmyr.
TJ:Awesome, man. So, first inaugural episode of a podcast that we have spent years talking about. I think we've been sitting on this name for at least 2 or 3 years and and have talked about it and have guested on several other podcasts and, finally sitting down to record, man. How's it going?
Chris:Finally. It's going great. Glad that we were able to do this after all the talking. And I I think it's definitely been, like, much longer than, 2, 3 years. Probably, like, closer to 5 or 6 compared to, like, how long we've known each other and all other, guys in the group.
TJ:Yeah. All these Podcasts. Other guys in the group. They've got they've got podcasts. They've got, you know, Dyrynda's got, several podcasts.
TJ:Yep. Yep. And, I know we get to do the annual, bonanza of of all the podcasts, plus plus TJ, plus Chris Gmyr. So now now we actually have a a titled podcast that we can we can join in with.
Chris:Yes. Yes.
TJ:Cool. So for this first episode, I think, man, we can, talk a little bit about the podcast, some things that we're gonna cover, do, do some intros, and, kinda touch on a few other things. How does that sound?
Chris:Yeah. Sounds great to me. Why don't you kick it off?
TJ:Awesome. Yeah. Yeah. I'll kick it off. My name is TJ Miller.
TJ:I'm based out of Metro Detroit. I live, I don't know, on a good day about 30 minutes north of the city. You know, I've got an absolute zoo here at my house. I've been married for eons. We're we're coming up on 16, 17 years.
TJ:We have 5 dogs, 4 cats, and I also have a 10 year old son running around. So it's, super chaotic, but it's a it's a ton of fun. How about you, man?
Chris:Yeah. Chris Gmyr, just south of Raleigh, North Carolina, about 20 minutes south of there. Been down here for, I don't know, probably about 8 years now. Previously, like, upstate New York, Syracuse area, getting out of the snow and colds, all that. Really loving it down here.
Chris:Almost no snow, which is awesome. A little hot in the summer, but not too bad. Been married to my wife for almost 11 years now. Have a 8 year old son and a 2 year old daughter and a dog. So not quite as big of a zoo, as you guys, but, you know, a little close with the the 2 kids, for sure.
TJ:Yeah, man. Especially a young one. That's, you know, 10's a fun age. And, you know, I just right before we sat down, I got my butt handed to me in a in a Pokemon card game. So, it's it's definitely fun fun getting into a little bit older ages, but I there's definitely times where I, I miss the young the young ages like that.
TJ:There there's just something special about it.
Chris:Yeah. Totally. There's there's pros and cons to each one. But, yeah, our our oldest is all into Pokemon, which we've talked about before. I'm sure we'll talk about it at some point on here.
Chris:But, yeah, it was just going crazy, like, making different, like, decks and trying things out, and I don't have that much time to put to it. So I just have my my, like, old crusty deck, and then, like, he just figures out which cards that, he can, like, whoop me with. And, then with all the other, like, weaknesses and attacks and stuff like that, he just has it dialed in. So, like, I haven't even been playing with him that much because he's just tuned it in that much. But gotta get back into it for sure.
TJ:Same. He's he's dialed in, I think, I think I've won 2 out of the last dozen matches. It's it's been absolutely brutal. Brutal. So I think we could also talk a little bit about our, our dev journey and kind of where we got to you today.
TJ:How about you kick that off?
Chris:Yeah. For sure. Man, I started off back in high school in early 2000s. It was like 2000 or 2001, that I was tinkering with stuff on the web. I had a computer early on and was like building them and doing like other things like very early, even like AOL and, you know, AIM like Messenger and like chats and stuff like that.
Chris:Doing like band websites. I, played music, like, all growing up and, drums and different bands and stuff like that. Jumped on, like, GeoCities and Oh, yeah. Made, like, the the flame, like, buttons with flash and, you know, all that stuff, back in the day. And, just started taking, classes luckily.
Chris:Like, we had some in our high school for, like, web design and some, like, early, like, programming and stuff like that. And ended up, doing so well in those because I would I do, like, so much on the side that I would come in. I basically had my work done. And eventually, I got to, like, work on, like, the school's website and the district's website. And then, like, some of the teachers were coming to me for, like, friends and family who needed websites.
Chris:So I started getting a little, like, side income, you know, late, high school and stuff like that. I started doing contracting on the side. I was, trying to think back. It was like, Creative Moonlighter and then they changed their name to, guru.com. I think guru.com is still up and running, but it's basically like an earlier version of, Upwork, which is pretty popular today.
Chris:And, basically, long story short, I, didn't do too much, time in college, but I started up this side business and agency and doing programming and, like, all the things on the side and, running my own thing, full time for about 7 or 8 years, until after, 2008, it started slowing down and kinda into 2010. And then I was just kind of, like, burnt out on it and didn't really wanna do it anymore. It's just hard finding, like, good people to, like, contract with and, doing all the client management, all the programming, all the all the things except for, like, design and some, like, heavier, like, animations and stuff like that.
TJ:Yep. Such a struggle, man. Yeah.
Chris:Yeah. It is. So I definitely didn't do it as well as I could have, but it was a great learning experience to kinda get that, like, well rounded, not only like full stack, application development, but just handling different types of people and companies and, you know, costs and all the things associated with, like, running the business. But I'm super grateful for, that time because I I think it added so much to my experience, over time. For sure, man.
Chris:Yeah. And then after that, because I was just tired with that, I was just like, I need to get, like, a full time contract or a job or something like that. So I ended up getting, a couple of full time contracts over the next, like, few years and, ended up, getting into some, like, regular, like, standard, like, w two, like, type jobs. And that's been for the past, like, 6 years or so, I think. Yeah.
Chris:So yeah.
TJ:Yeah, man. That's it's crazy. We we have such a a parallel story. Like, I I also got started, back a little bit before high school. My dad worked for Xerox, and so they were really, like, early tech company.
TJ:And so we were really fortunate to have, pretty decent computers really early on, and we got Internet also super early on because he was able to get it through work. And, yeah, I got I got started just kind of messing around on on GeoCities and and taught myself by, you know, just viewing source and and figuring out, like, what did what, Messed around with, like, yeah, GeoCities, Angelfire, Myspace. Like, I think my my very first paid gig was doing a
Chris:a band's profile on Myspace. Yes. Myspace profiles. Yep.
TJ:Oh, dude, that was that was so much fun back then. I I I miss I miss being able to customize all that stuff now. That was that was, something else. And, yeah. In high school, I started getting, some of my first paid gigs.
TJ:Also, had, like, an early intro, like, web course and just super excelled there. Also worked on the school and district websites and then, actually partnered up with, a couple other students and the instructor. And we started a small little agency and, went around to, like, local businesses and just found anybody who needed a website and just started putting things together. And, that's kind of where I got my intro into PHP is we wanted to do, like, static sidebars and headers and kind of dynamic body content, and that's where we found we kind of stumbled across PHP, and all of us learned it, and then that spilled into, like, WordPress. But, like, I spent, 12 years in the bicycle industry.
TJ:And in Michigan, you know, we have very full, hot summers and deeply cold winters, and so, that job was super seasonal. So I'd I'd be working 60, 70 hour weeks all summer long and then get laid off for the winter. And, for me, I would I would pick up a job at a coffee shop because during the winters, they'd lose their all of their staff going back to school. And it was perfect for me to slot in for a few months. And then in between working at the coffee shop, like, without I would only be working there, like, 30 hours a week.
TJ:So I had plenty of extra time, and that's when I would, you know, pick up a couple clients or just start exploring different things that I could do code wise. And, right when we were ready to have our our kid, I was like, you know, I gotta I gotta transition to a real job where I can make, like, sustainable money for a family and, got really fortunate. A buddy of mine through the bicycle industry, had his own agency, and he was working on a WordPress site and needed some help and was able to jump in and start learning WordPress. And he knew, another local agency that was looking for, you know, someone to to get in and start working more WordPress stuff, and that that was that. And I, like, jumped in as a full stack dev there, and their senior dev left.
TJ:And when he departed, there was, a few, like, Cake PHP applications, and I absolutely hated working in Cake. Like, I don't know. I just I really didn't like the experience. And I think I think part of it was the way that those apps were written. I know with, Cake, you could dynamically build the HTML outputs for the pages, and I think that developer took it on as a a personal challenge to see if he could get the entire page in one echo statement.
TJ:And, that was you know, if you had to go back in there and drop, like, a wrapper div around something, it was an absolute nightmare. So I started, I started looking around for other options and honestly was about to jump ship to Ruby and Rails. And I was looking at, Falcon PHP, and somehow looking at that led me over to Larevo, and I would just blew my mind. I never knew, I never knew you could write PHP that way. And so I got I got bit, and I've been now working with Laravel for the last 11 years.
TJ:And that's been just, you know, such a it it just holds, like, such a special place in my heart and and something I'm super passionate about. I think that's, you know, that's kinda what what led what led us together too. Right?
Chris:Yeah. Yeah. Totally. And, like, very similar to you, like, got into, like, some of the the older frameworks. Like, I was, big into, CodeIgniter and, that slowed down its development over those years and that was kinda like peak, like, my, like, agency time and just wasn't a lot of development going on and hearing about, like, you said, like, Ruby on Rails and some of these other, frameworks or languages or tech stacks.
Chris:And it's like, man, like, maybe I should, like, jump ship or start learning something else. And then, like, out from the side comes Laravel. And I think I found it at, the very end of version 3 Mhmm. Which was like it wasn't composer yet. It was like bundles.
Chris:Yeah.
TJ:Where you had
Chris:to, like, bring things in, like, by hand. And, version 4 was starting to get, kinda like more solidified, like this new thing, like with composer and stuff like that. And it's like, okay. Like, I'll start, like, kinda looking at the ecosystem and, like, what you can do. And this this seemed like, such a breath of fresh air even look looking at like v3 compared to like old coding error stuff.
Chris:Mhmm. For sure. I gotta I gotta like take a look at this and like stay on it. So basically when, like, v 4 came out, I was like right over there, like, started, converting apps, like, new apps, like, it's all, Laravel, like, using this, like, new fancy composer thing and, like, that was all new because it was basically just, like, you know, FTP and stuff up to, servers and, you know, that's all you could really do. Right?
Chris:Yep. And bringing in, like, downloading packages from, like, all these different sources and, like, SourceForge and, like, a little bit of, like, GitHub, like, that wasn't, you know, as big or, like, as easy to use, like, back in the day. And just, like, rode that, Laravel train, like, all the way through, like, to the present. And
TJ:Yeah.
Chris:I think that definitely saved, like, me and, like, my career for for, like, staying in, PHP. And it, like, hands down has helped the PHP ecosystem incredibly. And I think everyone is just, like, leveling up in the space. And even over the last, like, handful of years, we see all the positive change in PHP and, like, obviously, Laravel, you know, with that that it's just so nice to work with.
TJ:Yeah. I think I think Laravel was such a major catalyst for a lot of improvement in in the entire PHP ecosystem. Yeah. I got started around the same time. I it was tail end of 3.
TJ:I remember, I was looking at 3, and then, like, a month later, there was, like he he Taylor had announced 4, and all the changes there. And I was just like, what what is oh, what's composer? Like, that was the nail in the coffin for, like, FTP driven development for me. It was from then on, it was like, nope. We're going with, like, VPSs so I can shell in and use, you know, GitHub deployments and, like, Git deployments.
TJ:Like, it was, yeah. It was it just revolutionized the way that I, like, thought about, like, code and, like, what you could actually do with it. It was wild. I think, I was I was so into it, and we were still the place I was working at was still so heavily WordPress and, like, word WordPress driven app development that, I had built a small framework wrapping around WordPress so I could use composer, and I was pulling in, like, Illuminate components and, the IOC container. And I I got absolutely crazy with it.
TJ:The the last WordPress site I did was super gnarly with all sorts of, like, Laravel sprinkled into it. That was, that was wild. Yeah. Yeah. I'm I'm so happy that I don't work with WordPress anymore and made made the transition to you, like, app app development full time.
Chris:Totally. I'm so grateful that I didn't have to do too much WordPress. Like, I've installed it, like, installed a couple plugins, like, did a few couple customizations, but nothing, like, big development or anything like that, fortunately. So
TJ:I feel like so many people had the, like, CodeIgniter, Talarvo, transition, and I I never I never messed with CodeIgniter. I went I went straight from, like, my first PHP framework, was the experience was, like, with Cake. And it was, from there, it was, like, from Cake to Laravel, and, I had I had no no other no other journey beside that. So I was, consider myself pretty pretty blessed to have that.
Chris:Yeah. Yeah. Totally.
TJ:Yeah, man. So I think with, like, Laracon coming right up, you know, I think before we get into that, I think we can talk a little bit about, like, what what we're looking to cover in the podcast. I think, for us, I think it's it's probably gonna have a pretty Laravel and and dev heavy focus. We both we both work in the industry. Actually, we we both work together at the same same company, and I I've got That's right.
TJ:Who to thank for that.
Chris:You're welcome.
TJ:And I think
Chris:My pleasure.
TJ:Yeah. No. It's been it's been great. I was, I was looking for the next thing, and and you said, hey. I'm I'm I've got I've got an opening where we're at.
TJ:You should you should come check it out, and, I've been there for the last four and a half years, and I think you were there about a year and a half before me.
Chris:Yep.
TJ:So that's, that's super cool. So I'm sure we'll get into, like, musings about that and I think just kinda, like, personal life updates. I think if you follow us on socials, like, you'll you'll probably see a lot of similar stuff that we talk about. It's it's got kind of, like, a dev dev focus and spin to it, but you'll hear plenty about, like, what what we've got going on, things we're working on, and personal life updates and stuff.
Chris:Yeah. Yeah. Mostly mostly dev, with, like, a sprinkling of, like, personal interest and, like, interesting little tidbits, here and there. I think it just kinda keeps it fresh. Like, it keeps it personal and personalized and, opens up the door for, like, more connection with, like, people who have those, like, similar interests or even who might be, like, interested in taking on some of those interests.
Chris:You know? So even, like, you know, you said biking and you've been into skateboarding and Yeah. Some of the more, like, extreme sports, like, you know, touching on that. It's like might get some people who are interested in that, like, in the the dev or, like, Laravel space and I don't know. More more connection, the better.
TJ:Yeah. For sure. For sure. I've always appreciated the podcast where, you know, like, you really at the end of the day, you listen to a handful of episodes, and you really feel like you've gotten to know gotten to know the people on the podcast at, like, a a a personal level and hoping to bring bring some of that here, you know.
Chris:Totally. A 100%. And that's kind of a interesting feeling. Like, I listen to a ton of podcasts, like, either within, like, the Laravel space and outside and, especially going into, like, Lyricans and other, like, meet ups and and things like that. Like, hearing, some of the hosts talk on, like, a weekly or even, like, monthly basis.
Chris:And it's like you get caught up with all the things, like, work or dev or personal related. And then you, like, walk into the room and be like, oh, hey, you know, I have this too where, like, we ran into this same thing and you're, like, already, like, on the same page in, like, seconds. And, like, I would love for us to, be able to provide, like, another space for people to, like, talk and connect and even, like, come up to us in, like, future conferences and stuff like that and just have that rapport, you know, with the community. I think that'd be pretty awesome.
TJ:Absolutely, man. Absolutely. So I think that kinda brings us to, like, this is, like, a few we're recording this a a few days before before Lair Con kicks off. And, bringing that full circle, that's that's how we met, I think, back in 2016. So that's, jeez, that's, like, 8 8 ish years ago now.
TJ:And Yeah. That was that was my first Lyricon, and I think that was not too far past when I really started getting into it. And there was, this was in so it was in Nashville. Right?
Chris:Let's
TJ:see. I've got I've got a much more. I'm pretty sure that was out out in Nashville or,
Chris:Louisville?
TJ:It was Louisville.
Chris:Okay. Yeah. I did it 1 or 2 before that, I think. I was in, like, one of the smaller ones in New York City. I think it might have just been, like, the year before in New York City.
Chris:And then we had the 2 years or they did the 2 years in Louisville. Mhmm. So I went to those 2, and I think the first one, that was the one that, you were there. Right?
TJ:Yep. Yeah. That was, like, my my first dev anything conference. And, gosh. That was so cool.
TJ:There was, I think Evan Yu was there was was it was a real banger. Like, Evan Yu spoke about Vue. Sandy Metz spoke. Yeah. Gosh.
TJ:That was that was such a cool conference. I remember, having lunch and just, like, sitting down on the steps out back and then, looking over and, like, Taylor was sitting sitting next to me, and we were just, like, chopping it up. I'm like, this is so cool. And I think it's it's been the same at every Laracon I've ever been to. Like, it's a special place.
TJ:Like, I've been to several other conferences too, and just everybody at Laracon is just so approachable and so friendly. I mean, and that's the whole reason why I go to these conferences. Like, the talks are are just bonuses for me. It's all it's all about going and meeting people, getting to to meet new people, getting to see people I've met at other conferences or, you know, some of the some of my best friends that we've, like, met through Twitter, you know, never, you know, never get to ever meet in person outside of Lair Con. So, yeah, that's awesome.
TJ:And I've I've went to I've been to I've been to 4, and I think all of them are, like, back to back, up until I think, like, COVID happened, and I haven't been able to make it back since. Super bummed about not being able to go this year. But
Chris:Same.
TJ:Yeah. Yeah. There's there's some serious hype going on. I'm I'm I'm I'm on the edge of my seat waiting to see what comes out. I know I know the the Laravel folks have been just cooking over there.
Chris:Yeah. And just with the, increase in team size, like, over the last few years and, like, it's just amazing, like, what they've been putting out and just the quality of work that you would expect from, like, the Laravel team. And, even more impressive that it's not just specifically from Taylor, like, it's been, you know, in years past. But just like everyone is just on, like, such a high caliber, of engineering and people and team together to produce some of this stuff. It's just been, like, awesome to see.
Chris:And they keep on coming up with new and better, and bigger ideas. And it's just so awesome to see, like, year after year, like, what, Laravel can do.
TJ:I mean, we've just had, like, absolute fire. There's been, pulse and prompts and, huge improvements over on Forge, like, wild, the amount of stuff that's been getting done and just yeah. Like you said, that that high high quality high quality work that just, you can't find anywhere else. Yeah. I know they've been they've been hyping up.
TJ:This is, I think, gonna be the the biggest drop, I think, they're they have of stuff this year. So, I I can't wait to see what happens.
Chris:Yeah. Same. And that'll be a a good follow-up for, like, us to take a look at in future episodes is just, like, diving into some of the updates, some of the things that, like, really caught our eye. And, hopefully, we can bring some more of that information, you know, to our audience and even building it into some of the stuff that we're working on.
TJ:Yeah. Like, what would it it's, brings the question up, man. Like, what what part of the stack do you are you, like, passionate about or or, like, bits and pieces that are that are, like, near and dear to your heart? I know for me, APIs are, like, a really, really big piece of what I like to focus on and and also the the CLI side of things. I like I like really working on, like, dev tools.
TJ:So, like, prompts is big. The the Artisan CLI is is a big deal for me. Like, I like writing CLIs in Go also. Like, what kinda what kinda parts of the stack are you are you passionate about?
Chris:Yeah. I'm very similar. I've been mostly, like, back end focused, in the day to day work that I do. And definitely, like, the the CLI, like, even building, like, artisan commands or, like, file generators for, like, some of the the features that, I've done at work. Like, making you, like, make commands and just, like, automatically, being able to scaffold, whole sections of applications with, like, a boilerplate, like, test suite, like, already, like, out of the box for other engineers to, like, come in and just, like, jump in and have tooling available to them, like, has been awesome.
Chris:We've been working together on some, like, API stuff at work too, which has been fantastic. And just, I'm sure we'll get into this, more in like future episodes, but just like, converting our very, kinda spaghetti code and, complexity, like, over the years and, making it more manageable. And we've done, like, a lot of, like, modularization, efforts and starting to explore, like, other architectures and services that we can use. So really like looking into how I can, like, make the application itself like more performant, you know, cleaner, easier to use, and, kinda dev friendly. So, all that stuff, like, you know, on the back end and just making it more of, like, a joy to to work on.
Chris:Like, that's what I really attached myself to.
TJ:Yeah. I know you've been really driving the, the modular monolith, and really, like, beating that drum, and I think that's made massive improvements. It's it's a it's a it's a big, gnarly monolith that's been around for a long time and, has not aged well. And I think I think you've you've done a a really good job leading the drive to to kinda clean that up and and make it a a much nicer place to work in. And you're you're pretty fairly full stack as well too.
Chris:Right? No. I try not to be. Yeah. I try and stay on most of the the back end.
Chris:If I need to, like, tinker with or, you know, give some help or insight on some things that we can do on the front end, It's definitely more conceptual. It's definitely not like in the weeds of making all those, changes. I I definitely burn way more comfortable in the back end than the, the JavaScript craziness of the the front end ecosystem.
TJ:Yeah. I feel like I I I tend to to bill myself as a full stack capable back end developer with a flair for DevOps. So, like, definitely prefer yeah. I definitely prefer, back end development. I I really like the last few months have been kind of serving as the the glue between our SRE team and app development, doing a lot of a lot of containerization, a lot of work with Docker, and, it's it's a fun a fun spot to be.
TJ:But if I need to get into the weeds in the front end, I definitely can, but, boy, am I do I do I feel rusty staring at anything JavaScript right now?
Chris:Yep. Totally.
TJ:We've been, I know we've been exploring. We recently, like, implemented a bunch of stuff in in Next. Js, and that was pretty cool. And I know some other team members have been experimenting with the remix. And I went in to, like, review a handful of PRs for some of them, and I'm like, oh my gosh.
TJ:I am I feel so out of my comfort zone right now.
Chris:Yep. 100%.
TJ:Cool, man. So I think we can probably wrap this up talking about, you know, some of the stuff that we're, like, working on or focused on at the moment. I can kinda kick that off with, like, I've I've been working on Sparkle, and that's it's been nice to get to a place where I've been able to make some fairly consistent progress on that. Sparkle is a AI package for Larevo for for working with large language models in a, kind of, like, agent agentic focused way. And as far as agents, they're, you know, they're kind of a, you think of an agent as a large language model or, like, an AI with a a task, like, a a personality and a task to do.
TJ:So one of my, like, agent style workflows that I have that I I really enjoy is I set up a a Telegram bot, and then I can send it voice messages. And kind of the pipeline after that is I send it a voice message. I have one, AI agent do the transcription, so it takes it, turns it into text, and then it passes it off to, like, the next agent. That agent, turns that transcription into a bullet note, And then the 2 subsequent agents after that, do different things. They one will extract any action items that were pulled out of that transcription, and then the other one will look for any decisions that were on that description, like and then takes it all, drops it into a Notion page, and puts that in a side of a database in my my Notion, like, my my little personal knowledge management Notion, area.
TJ:And then I've got, like, all that to go back and reference. So I've now got, like, searchable, all nice and formatted and easy, and that's that's kinda how I like to explain, like, what agents are is, you know, each of those steps were an agent specifically designed to do those tasks. And, yeah, yeah, I've been working on that for the last little while. You know, I I started getting into it. I spoke at PHP Tech about it.
TJ:I spoke at the Laravel Worldwide meetup earlier this year about it too. And then, my wife got sick, and so she was she was down for the count for almost 3 months. And with that, it was just, an onslaught of life hurling lemons at me and just me frantically making as much lemonade as possible. Her health's improved. Life's settled down.
TJ:So over the last, like, handful of weeks to a month, I've been able to actually make some, like, pretty steady progress on it and, really enjoying where that's going.
Chris:Yeah. That's awesome. Yeah.
TJ:Yeah. It's been it's been fun. I've spent the last 2 little over 2 years now, just my ADHD, like, hyperfocus latched on to AI, and I I just haven't been able to let it go. And, I mean, I use it on a daily basis. I have, a couple different agents that I use for various different, like, chat tasks, but I it's a it's a daily daily thing for me.
TJ:I don't know what I would do without him at this point. I've I've kind of adapted to him. Yeah. How about you, man? What are you
Chris:what are you working on? Well, yeah. I just, wanna keep going down, your
TJ:Oh, yeah. Yeah.
Chris:My plot there. But we'll definitely get into, like, more of the AI and Sparkle stuff in future, episodes. But I just think it's so interesting how, like, the the non tech, like, general public sees, like, AI as, like, that it's just this, like, new thing or that it's, like, only chat GPT. Right?
TJ:Yeah. But,
Chris:like, there's so many more, like, use cases. There's so many more models. There's so many things that you can do with it either, like, using these big models or starting to train your own. And, like, what you just explained with your, like, Telegram, bots and and workflow to Notion, it's like you set up individual agents to do very specific things. It's not like one thing to do all the things, right,
TJ:in
Chris:the workflow. So, like, knowing how to, like, break those apart and what's, like, agents or models to use, like, I think would be, like, a really interesting deep dive to understand, like, what's available, like, how you can plug into, like, each one. And I think, your sparkle package is going to be, like, amazing for that, especially in the the PHP and Laravel space too.
TJ:Yeah. Because that
Chris:I hope it
TJ:just, like so much
Chris:that we can do, like, with that.
TJ:I hope it just breeds like a a playground for people to, like, go and and learn and explore more about it because you nailed it, I think. A lot of people just see what's on the surface with chat gpt. And while that's great, that is a very, very general purpose tool. And so you're only gonna be able to get, you know, comparatively to something that's like a specialized agent. You're gonna get very mediocre results depending on what you're looking to do.
TJ:And being able to, like, tool Toolify, I I guess, all of it, you can build some, like, really powerful workflows. For example, when you're just talking to chat gpt, that that's, like, one cycle. Right? You send it a message, it generates a response, and then that's just what you get back. One of the things that I do immediately if I want, you know, higher quality responses is I will actually have, like, the agent that I'm chatting with, send its response off to another agent, and that agent offers feedback.
TJ:And I I call this the actor critic pattern. So there's, like, the actor that receives that response, you know, maybe receives receives your request, and then you take that response, you send it to a critic and that critic is gonna offer, like, feedback on that response and then feed it back to the actor and then you're gonna get immediately, like, a higher quality response and, like, it's significantly better, especially when you're asking it to do, like, programming based tasks, like, hey. Like, generate some code for me. You know, that critic is gonna be able to spot all that stuff. Because even with chat gpt, you get a response back.
TJ:And even, like, even if you just ask it, like, are you sure? Like, it'll go back and iterate on itself, and it'll be 9 times out of 10 go, well, actually, I think if we do it this way, it'll be better. And so you just kind of, like, avoid that whole, like, feedback cycle, and and you only really get that if you like Toolify. And there's tools in lots of other languages, like, namely Python, but, you know, my my heart's really with, like, Laravel and PHP. And, you know, when I got into any of this stuff, I had to go off and, like, learn Python, and I always just felt like I was fighting the tooling.
TJ:I was just I could it it was on, like, at the tip of my tongue what I wanted it to do, and I just couldn't, like, bridge that gap. And now being able to bring something over to PHP, it's, it's so much easier to, like, you know, have an idea and be able to execute on it.
Chris:Totally. And I think the the pattern that you just explained would be great to go more into detail, like, in the future, and I think would be, like, a fantastic blog post for you to put together, also.
TJ:Oh, yeah. For sure.
Chris:Be, like, fantastic.
TJ:Yeah. Once I get some of the tooling down, like, one of the things that I, I appreciated when I was, you know, first getting into this is there were a lot of, like, Python frameworks to do very similar things, like, namely, like, langchain and there's a couple other ones. But they all had, like, either in the projects repository or in a, like, in a company repository, had were, like, thick of tons and tons of different examples and use cases of, like, hey. If you wanna do this actor critic type pattern, here's here's a way to do it. Here's a a prime example.
TJ:And that's something that I wanna, like, offer up with this too is just, like, as many different, like, tools and examples as I can, like, put out there. I wanna put out there so people can just go, like, source dive and, you know, be able to grab something, copy and paste, and get up and running, and then kind of, like, fine tune it and customize it to their needs. So, yeah, I'd I'd love to be able to put out, like, heaps of examples and tutorials and stuff like that. I definitely smell a, a video course somewhere screencasting course somewhere in here. So
Chris:Yep. Totally. I And that's how I learn best. I like watching, then doing, and then researching after that. Like, it's really hard, at least for me, to go through, like, large documentation sites and try and, like, piece together, like, some of the especially, like, in, like, a new language or a new concept.
Chris:It's like I have all of this, like, expert knowledge built up in, like, PHP and Laravel. Right? And it's hard to go back to that beginner state. But if I can see someone, like, work through that, like, live or like in a video course or something like that. I have enough pieces that I can put together more easily and then know what I need to search for or figure out, like, after that.
Chris:Right?
TJ:Yeah. I mean, I'm a I'm a a dive into the deep end and and figure out how to swim kind of person.
Chris:Yep. Totally. So I think, like, any of the the the tooling, the educational pieces would be fantastic. And what I've seen from, Sparkle so far and all the things that you've, shared with me and, like, us in the the group chat, like, has just been, like, awesome. So it's looking amazing, and I can't wait for you to get through it to share with the world.
Chris:And I I appreciate that. Use case.
TJ:This is a great use case for AI too, man. Like, like, learning new stuff is is so good. I mean, you gotta you gotta have your your wits about you because, you know, if if you've ever spent any time messing with chat gpt, you know it'll kinda come out of left field with some stuff that's definitely not correct. But,
Chris:if you've got
TJ:your if you've got your wits about yourself, you know, when I that that's how I learned Python in order to do the things that I needed to do was, a lot of times, I'd write some, like, pseudo code in PHP and be like, right. How do I do this in Python now? Like, here's the libraries I'm using. Tell me how to, like, translate between these two things. And that's how I picked it up.
TJ:Like, I don't just just, like, reverse engineering some of the the langchain examples and then, doing a little bit of source diving, but so much of it was just actually using AI to, like, understand AI and, like, understand Python, and being able to translate that between, like, languages that I do know and understand, is is a great use case for for AI.
Chris:Totally. It was a great use case. And I
TJ:know you had, you had talked a little bit about, starting to dabble dabble in Golang a little bit, and I think I think it pairs so well with, with PHP. And I think there's you know, as I've been brushing up on Go myself, being able to, like, pseudo code that out in PHP and then, you know, get it to translate that over to, like, to Go or the approximation of it, It's really good at doing that. I think I think that's, like, that'd be a good way to go.
Chris:Yeah. I've definitely been, like, dabbling with that as I, like, work into, Golang because, you know, just being used to, like, basically one programming language for, like, so long, you just get used to, like, how everything works and, reads and, like, converting that over to, like, kinda like an Otter, you know, syntax, is definitely, like, a little jarring at first. But then, like, once you understand, like, some of the concepts and using, some of the chat and AI tools to kinda fill in those gaps, like, it definitely, like, opens up the world, a lot quicker, you know, than just going through the docs or, you know, something like that. And being able to compare 2 languages together and like you said, like, I wanna do this from, like, PHP. Like, how do I do this in Golang or, you know, Rust or, you know, whatever is super helpful.
Chris:So, yeah, I'm definitely enjoying, like, the journey of getting into something new after, like, being in, like, just, PHP for the most part, for, like, all these years.
TJ:Yeah. I think it's it's literally, so much of just to, like, drop in there and go, hey. I'm a PHP developer. I'm learning Go. Explain this Go code to me and, like, paste it in, and it will do a pretty fantastic job of, like, explaining it to you through the eyes of a PHP developer.
TJ:Yeah. Yeah. Figuring out, like, all little little tips and tricks, on how to, like, prime prime and prompt. The AI is is is a little tricky. But yeah.
TJ:No. Just say, like, I'm I'm a PHP developer. Explain this to me as as such, and, yeah, you'll you'll get very contextual answers out of it.
Chris:Yep. Totally. Where it does unfortunately fall down a little bit is, like, a lot of the other work that I do is a lot more abstract and higher level. So, like Mhmm. The different, like, architecture type patterns or how do we get from, you know, our current state to this, like, ideal, like, future state.
Chris:Like, it's hard to get past, like, some of the more just, like, basic definitions or, like, fundamentals of, like, going to the next step. So, like, going from, like, a monolith to, like, a modular monolith or, like, a service oriented architecture. Like, it kinda breaks down a little bit on there because it's just so much information and so, like, abstract and, like, it doesn't know enough about, like, your specific, like, instance to connect the dots. So Yep. It's still helpful, like, in certain, like, instances if you need to, like, compare, like, 2 things together to, like, make it a little bit more, like, solid in, like, your mind, it helps with that.
Chris:But, like, those bigger, more abstract jobs, like, it still, obviously, you know, has a hard time. But, you know, it is what it is at this point.
TJ:Yep. Yeah. No. There's still a lot of onus on on you to kinda figure out the specific steps, but it can help you with, like, the broad strokes or or giving you the the details once you kinda get into those decisions. And I think a lot of people are starting to find find that out about about AI, and that's where we're seeing a bit of the hype die down was as people are realizing that, like, it's it's not gonna take over everyone's jobs, at least right now.
TJ:And there's still quite a way to go before we get to, like, AGI or anything like that.
Chris:Totally. 100%.
TJ:Just the full Internet of
Chris:the box.
TJ:It's it's it's still just fancy auto complete. Like, at the end of the day, that's really just what it's doing. So
Chris:100%. Yep.
TJ:Cool, man. I think that'll probably wrap us up for this episode.
Chris:Yeah. Let's wrap up. This has been a great one.
TJ:Cool, man. Well, I think we're gonna we're gonna try to, to keep this on a a weekly clip at least for a while, and I'm sure we'll have some gaps in here and there. But, yeah. Great first episode, man.
Chris:Yeah. Awesome. And thanks everyone for listening and tuning in. We'll have some other information coming out, soon as far as, like, socials and how to get in contact with us. If you want to join us for a chat or ask a question or, whatever you wanna share with us, would be awesome.
Chris:So we'll have that in future episodes and future share outs. And, yeah. Thanks again for listening and we'll see you next time.
TJ:Yeah. Thanks. Bye. Bye.
Creators and Guests
