Subscribe
Share
Share
Embed
Greg and Michael discuss the Laracon AU venue announcement and tickets going on sale, Greg's coaching pricing experimentation, and more considered approaches to using AI in your applications.
Show links
Show links
Creators & Guests
What is Ripples?
Greg and Michael discuss the thrill, the uncertainty, and all the other emotions that come with putting yourself out there and never really knowing where or when the ripples you send out into the world will overlap again.
Hi. This is, episode 10 of the Ripples podcast. I'm Greg.
Michael:I am Michael.
Greg:And we are guestless again today, but a lot of stuff a lot of stuff going on in both our worlds, I think, at the moment. So some interesting stuff to talk about since you probably have the biggest news in terms of switching on Biggest. Laracon. Mhmm. What do you tell us what you've been up to, mate.
Michael:Well, last last episode with Marty, we talked about all of the things that were going to happen, and they have since happened. So, we announced the venue, which you have known about for quite some time. But, we managed to to steer Stephen Rees-Carter off the trail. He was he was convinced it was one venue, and then he came around and decided it was a different venue. And then the different venue was the correct venue.
Michael:But I was like, we can't have Stephen having the correct answers. So I carefully and intentionally crafted some false messaging in some public locations that I knew he would see. And he took the bait. So threw him off the trail. So, yeah, we're at the QUT Gardens Theatre at the Queensland University of Technology this year, which is a beautiful, very modern theatre.
Michael:It's got all the bells and whistles. It's got pretty lighting and comfy comfy seats inside, which I'm sure everyone will be happy about. It's got a huge foyer. So they do cocktail functions in there for, like, 300 plus people, which is good. And then we've got the campus grounds.
Michael:We've got the botanic gardens directly next to it and things like that And loads of fun and exciting things, in in the vicinity. So very excited that information is out there. I no longer have to keep that secret. Tickets went on sale went on 18th May. So we've had 6 days, if my maths is correct, of ticket sales.
Michael:We have sold out of our blind bird allocation. So we we kind of forecast that to to last us until sort of the 1st week of June. So we're 3 weeks ahead of schedule, which is very, very exciting for me as the event organizer that that we're, you know, taking along nicely. So we
Greg:The nerdy, bit nerdy changing changing cities and wondering whether or not the recipe will still work. So it's good to hear that, that everyone's coming out of the woodwork and and heading along. It's gonna be an absolute banger. Like, there is easily twice as much going on this year versus last year, and last year was, like, massive. So
Michael:Yeah. Yeah. If you bought
Greg:a ticket, they're still cheap, just not as cheap. Go get 1.
Michael:Yeah. So on on that, you know, we we sold out the the blind bird allocation, which is for those eager beavers that wanna get in 1st, the early risers that that wanna catch the the best waves as it as as it were. And so, you know, 3 weeks ahead of schedule, we announced our first three speakers for this year. So Matt Stauffer. Everyone everyone should, I think, who listens to this podcast know who Matt is.
Michael:Matt spoke at the very first Laracon AU back in 2018. And I have actually been plotting since June of 2022 to get him to come back to Australia. So I'm excited that we've made that happen. Simon Vrachliotis is speaking as well. He also spoke at the first Laracon back in 2018.
Michael:So excited to have him back. And scheduling conflict after scheduling conflict over a number of years. And then even COVID in the middle of that, we've managed to get, Mandy Michael to come and speak this year as well. So very excited with with our first three speakers. The Laracon US schedule went up, today as well.
Michael:And I had been kind of waiting for that to see the kinds of talks that are going to be presented in August. And we kind of base our line up a little bit, not on necessarily the exact topics, but the breakdown of kind of technical talk to soft talk to business talk to, you know, to kind of have a good balance of of similar styles of talk. So that that kind of helps with balancing out the the rest of the schedule. So I've got 60, I think, talks in in the shortlist that I now get to whittle down to the final. I think we're gonna do 17 again.
Michael:And I'm and I've got a few ideas around a panel that we can do this year as well, which I think, you know, 3 or 4 people in different stages of or different parts of different businesses, that we can get on stage. So not necessarily a Laravel specific panel this year, but, I've got some ideas on on what we can do. And we've got, a good selection of speakers that we can kind of fill that out a little bit. So, yeah, really, really excited with with how things are going to date. Excited to to see how they continue to unfold.
Michael:My focus now for the next 6 days is to iron out that that speaker lineup and, send out invitations early next week, you know, early this week at the time that that this recording comes out. So if you're listening here and you have submitted to speak, then, keep your eyes peeled for invitations. We will try and and send out an email to speakers that submitted but weren't selected this year as well. Just to kind of say, yeah, thank you. Thank you for putting in the time.
Michael:We encourage you to try again next year because, you know, when when you have a 130 and you need to you can only put 17 speakers on stage. You know, there is some expectation that not everyone's gonna to make it across the line. And and as I said, it's it's not it's not because they weren't good talks, or I didn't think that they were gonna be good presenters. It's just that we have, one, so many people submitting and, 2, a certain balance of topics that we wanna fill. So keep submitting.
Michael:And if you if you don't make it, then please still come along to the conference and and see, you know, support the speakers that that are there because I'm sure that they would be there in return if you were speaking as well. So, yeah, very, very excited to get that all going.
Greg:Yeah. Cool. Yeah. It's, the the location's been really, really great for those who are in Brisbane. It is right in the middle of the scene.
Greg:Just to put it to bed once and for all, the tweet that you sent out that did send Steven off the off the scent was completely accurate. It is it is very close. Anyone who thinks a 20 minute walk is, like, a long distance is, like, kidding themselves. It's it's not that far. Really isn't.
Greg:Yeah.
Michael:Definitely subjective. I go a 30 minute walk every day. It's it's fine. Don't worry about it.
Greg:Yeah. Get your steps up going. It's good for you. But, yeah, even if 20 minutes is too far for you, just a quick, quick stroll over one of the many, many, many bridges. We have lots of bridges in Brisbane.
Greg:1 of the many bridges across, to Southbank. There's heaps of cool stuff.
Michael:Yeah. There's like a little walking bridge, from Southbank right in to Gardens Point, basically. So
Greg:Yeah. Yeah. And I said, like, the, Gardens Point, like, in the the whole sort of town gardens, area is is a real sort of nice kind of sanctuary there. Like, you you know you're in the middle of the city, but it doesn't feel like you're in the middle of a big concrete jungle
Michael:out there. It's all It's a it's a baby baby central park.
Greg:Yeah. It's one of, one of 3 gardens in Brisbane, or public garden like that, and it's probably it's probably the nicest. I guess I think it's the oldest, so it's it's established. Lots of things. Yeah.
Greg:It's both having spending a day walking around that should be really cool. Yeah. Awesome. Well, that's, that's some great news. I'm glad that glad to hear that the ticket sales are going well.
Greg:Laracon sell out pretty pretty reliably around the world, I think. So Yeah. Yeah. Should should be should be a good one. Yeah.
Michael:You know, funny headspace for me as well because, like, we intentionally put tickets on sale early to give people plenty of lead time if they need to, you know, if their employers are covering it to to make sure that they have either plenty of time to get it into this financial year's budget or plan for next financial year. And also like, you know, we we want to appeal to as many people as possible. I think that the tickets are very competitive in terms of what you get and not just the conference inclusions. But, you know, if you look at other events, other professional events, you know, it's their orders of magnitude more expensive, like some of them, twice, 3 times, 4 times as as expensive. So I think it is still really good value at the price that it is.
Michael:I think, you know, some people are not at the same level of being able to just say, yep, Oh, oh, buy that now. They might have to plan for over a over a few weeks. And that's why there's there's plenty of time. But like I said, those those blind bird bird tickets went really quickly. So
Greg:Yeah. Yeah. Cool. So Tell
Michael:us about your your consultant. What's what's going on? You've you've had a a renewal couple of renewals?
Greg:Yeah. Yes. I, I started to play around a little bit with, so I'm coming up in the financial year, and everyone's putting prices up. Next financial year, some of that's just because costs going up. So I decided to play around with, some sales y strategies, which is very much outside of my my wheelhouse.
Greg:So I let all my my existing clients know that, you know, the prices are going up marginally, next year. But if they're, you know, happy to renew now, that I'll hold them at the price that they paid, for this financial year. And I also sort of kicked in and said, look, if you'd be so I I I sell my time in in in blocks of 5, so, you know, you'd be by 5 hours with 5 1 hour meetings. And I sort of said, look, if you buy, 10, I'll give you one of those 10 for free, which has, sort of helped a few people, along, I think. And it's I think it's pretty good value.
Greg:I get a lot of good feedback from people that I do end up speaking with that they get a lot out of it. So, you know, getting being able to sort of offer, a bit of an incentive to get sort of my, sort of front load the start of the next financial year. I still don't take on that many clients just because of, you know, home and, day job commitments and things that go into that. So, yeah, it was really good. Both, people who renewed were probably ones that I wasn't necessarily expecting too, which is lovely surprise.
Greg:It's good. So yeah. I mean, I think it takes there's probably 2 types of 2 types of people when it all kind of boils down to it. The ones who kind of need a piece of advice, for a short period of time to get them over some, challenge or set of challenges that they're having. And then there's people who, either want to have or need to have, sort of more ongoing support.
Greg:So it's hard to know speaking like, it's hard to know in the in the moment whether or not like, which of the categories people fall into. So, yeah, that's been, that that's been, like, hugely successful, which, which I'm I'm really glad of. I was kinda getting to the point where I thought it was petering out a little bit, but, you know, just look at, like, how how the offering goes together and and and and and refining that and refining that as I learn, like, effectively how to run a side hustle, which is pretty new to me still. So stumbling around in the dark a little bit on the on on the things that I'm not, I'm not you know? But it isn't my wheelhouse, I guess, which is the the joy of running your own thing.
Greg:So, yeah, that's, that that that's been really, really good. Got a opportunity to speak at Brisbane Laravel meetup. We say opportunity to speak. I run the Brisbane Laravel meetup, so there's a lot. So we got a new sponsor for that.
Greg:It's, that's Trilogy Care, which is really interesting business, Laravel shop in, in Brisbane. They're they're in the aged care sector. They had a really flash office. So they, offered to sponsor this round of the Laravel meetup. So we'll be in their offices next so the Wednesday after this, this recording goes out.
Greg:If anyone wants to come along, I will be digging into a little bit more about some of the stuff I spoke about at Laracon last year. So one of the, one of the things that keeps coming up is, people come to me. They say, I saw you talk at Laracon. You talk a lot about small steps, and it sounds like it's a good idea, but how do you actually do it? Like, what is the what what are the mechanics of actually being able to do that?
Greg:And, as I've said before that there just simply isn't enough time in a 30 minute talk to, like, dig into full of the mechanics of all the different things you can do. So I'll be picking apart, sort of some real actual strategies and pitfalls and anti patterns and patterns of working in small steps, which which I think will be good when I actually roll my sleeves up and write the talk. Yeah.
Michael:So,
Greg:yeah, that that that that's that's coming up coming up really soon. Steven and Rizka will also be speaking there, assuming he doesn't go to, the convention center instead of the offices
Michael:at.
Greg:Oh, I'm gonna copy from him when he uses this.
Michael:Yeah. I, was it you and I? I think it was either that I was talking to that we should get, like, some special signs made up and, like, just have them at at the convention center and just take photos and send them out. They're like, don't go here.
Greg:Yes. Welcome to the convention and exhibition center. If your name is Steven, you were at the wrong place.
Michael:Oh, we just we just put a sign out the front that says this is not the Brisbane Convention Center.
Greg:Oh, no. Yeah. Well, you will you will mess around and try and spoil all the surprises. So so yeah. So, yeah, that's, the sort of news news on my front, I guess, for for what's going on with me, at the moment.
Greg:Very busy at work. Very busy in the day job, at the moment. So we're doing a bunch of work to, right size the the platform and the technology to the sort of size of the team. Mhmm. So so during COVID, we we expanded a lot.
Greg:We we built lots of experiments and so on. So there's a lot of code that's just floating around that that,
Michael:we're
Greg:having to maintain. So stuff. Yeah. Yeah. So lots of work on that front that we, that we're having to having to tidy up, which is keeping us keeping us really busy, playing around with some fun AI tools as well, which is, always a joy.
Greg:So, yeah, that's, that's that's where I'm at, at the moment, I guess.
Michael:Yeah. Nice. Yeah. We we talked a little bit about AI tooling and and how that fits in with us as a business. And it's it's kind of difficult in financial services because you're on the hook for whatever that AI says, basically.
Michael:Mhmm. And, you know, you you you've got to caveat it with this is not fun. And, like, our customers because we're an aggregator, our customers really are the brokers. So the brokers in in so far as, like, you can ask questions about documentation kind of thing, like, you know, which lenders support this this set of criteria for whoever it is. But, like, at the end of the day, the broker is on the hook.
Michael:So it's like a really fine line between this is not financial advice and this is like pointing you toward the financial advice that is already there. So it's an interesting thing. We keep we keep putting it off because we've got too much stuff to do, that is not, you know, playing around with AI. But at at some point in the future, maybe.
Greg:Yeah. We had a bit of an interesting journey towards it. I think when the sort of all the AI stuff went and kicked off, so sort of when Chat GPT first came out. And I think Ian Landsman actually responded to a tweet that, Ian Landsman had last week. Just trying to dig it up.
Greg:I can't find it. Everyone seems to go to chatbots because that that's the sort of economical example of Yeah. Of AI. And we we sort of did the same thing. Like, what can we use chat GPT for to, improve, or reduce workload or or whatever?
Greg:And we had some very mixed results, but there's definitely some parts of the business that that are able to use that and similar tools. But when what we found when building features is that the best thing we can do is to leverage AI, you know, almost invisible way, that, you know, improves a feature, improves something that already exists or, or so and so. You know, can you lodge a job on our platform using natural language as opposed to speaking to a chatbot, which Yeah. In our case would feel kind of unnatural, and and you'd you'd you'd end up with, you know, a lot of hallucination and so on. So, being able to sort of convert and put out sort of get rich tool does is it converts a blob of natural text into a JSON payload that is compatible with an existing feature.
Greg:So we can stop people from having to click through a multistep wizard in order to to to to launch a job. So that's that's been really successful. We're playing around with call transcripts, and call summarization at the moment, which is designed to speed up some of our internal, internal functions, which, which is really neat. But this is you know, it it it is all powered by AI, but it it it isn't sort of you know, it's not a chatbot where you can go and speak to and, have it pretend to be a human being and, get really awkward and promise there's a dollar
Michael:piece of equipment for $1 or something like that.
Greg:Yeah. Yeah. So, that that's that's it's been an interesting sort of, journey for us. Just uncovering where where the usefulness in that in that technology actually lies. And, I think you've also gotta take into account what your audience I mean, your case, you you're dealing with brokers, you're dealing with financial advice, and those kinds of things.
Greg:So, having sort of an open what is effectively an open prompt that, you know, people can prompt engineer their way into any sort of they want is probably not not super cool. But, you know, getting to a point where you can, you know, have somebody say a thing or type a thing and partially fill a form in just to save some busy work. Yeah. You know, it's potentially, potentially a safer use case, especially for the human who's then, like, auditing the output that comes out of it. Yeah.
Greg:It's certainly not, like, sexy, and it's not gonna land you, like, massive amounts of VC funding like, you know, some of these, like, chatbot businesses, are getting, but, I I think it's probably more useful. I think there's sort of invisible AI is is probably more if we've always sort of have a bet, and I'm always terrible at these bets. But I think that sort of invisible use of AI is probably more likely going to be where most of the use and most of the value comes out of that that sort of technology. So, yeah, it's fun and exciting. It's it's it's an interesting sort of space to to play around in, fiddling with all the new tools, AWS Bedrock, and Claude and a bunch of other things, which is, yeah, really interesting.
Michael:Yeah. Yeah. I think I mean, ultimately, for us, there is 2 layers of protection between, like, an end user, an applicant. And, you know, getting any financial support is, you know, the broker has to do the things and the lender ultimately has to sign off on serviceability, availability, you know, all that kind of stuff. So but I think, yeah, broadly, everyone's like, we're using AI to do this thing.
Michael:And it's like, well, I don't know how much AI is the selling point versus what you do with the AI being the selling point. And I think there's a there's a strong bent to, you know, we're using AI because everyone, you know, AI is the buzzword. But, finding ways
Greg:to use a really care
Michael:to a business where it actually empowers your users. Yeah. No. I don't think users care. Yeah.
Michael:But finding ways to get it into your business where it empowers your users to do either something that they couldn't do before or do something that they could before with, like, the form stuff in a in an easier way, then, yeah, that's that's a terrific use of it. And I think, like, at the moment, we've got this click through thing that's like, find the loan type or find a lender that will support this loan type. You know, like, click drop down, click button, and then you, like, scroll through the list. So if you could specifically say, like, I have a client that has x, y, and z. Show me lenders that can do, you know, this kind of, like, lending facility or commercial asset loan or something like that.
Michael:And and then then that comes back with with the information that's already there as, like, this this is an option you can explore as, like, a starting point. Ultimately, there's, like, so many layers of stuff that you wouldn't just go from, like, that initial thing to suddenly someone's got, you know, a loan that they can't repay. But it's, you know, just simplifies that process. And I think that kind of augmentation of existing things is is where AI is actually going to be useful.
Greg:Yeah. I mean, you can kind of think of it more like, like yeah. Or it's it's more a replacement for for someone doing the the the typing as opposed to someone doing the thinking. And I think too, like, in a lot of cases, people think AI and because something like ChatTBD has been wildly successful, they think AI, they go straight to chatbot, and then they ignore all of the other or or they they become closed off to all of the other potential applications for that for that piece of tech. When you see it all the time, you see it in, you know, whenever any, you know, new technology becomes, like, the invoke thing, people tend to copy the the the initial example as opposed to, you know, trying to think out.
Greg:So we were definitely we were definitely locked into that kind of thinking like, oh, well, how does a chatbot make sense for our our use case? And it it probably doesn't. But it it may do, I think, for things like frontline support, and it may do for things like document searching and things like that. But econ closes you off to to to a lot of the other applications. And and once you start exploring some of the other things you can do with it, you can you you really open the doors to to a bunch of things that you either couldn't have you may have thought would have been cool to solve but didn't have a solution for it before, and then suddenly you're unblocked and you can do all sorts of cool crazy things for, you know, relatively low cost, or it opens up a whole bunch of things that you you probably couldn't have dreamed up in your wildest imagination.
Greg:So, yeah, it's, you know, early days. I mean, we've been fiddling around with this stuff now for probably
Michael:Yeah. Those those are ripples that are going out. So we're figuring out which ones are gonna resonate. Yeah. Yeah.
Greg:Yeah. And I don't I don't know why I don't know. In in some in some ways, I think, there's a lot of parallels in the sort of a a AI revolution as there was with the, the tulip craze that was web 3. There's a lot of hype. There's a lot of it can do all of these things, and I don't you know, pessimistically, I would think that, you know, the reality is gonna be far less shocking and surprising that than people think it is going to be.
Greg:But that said, there is definitely value. There's definitely value in doing it. And, you know, the the fact that it can we we we we've been able to, like, set up, you know, sentiment analysis and, you know, communication categorization and so on in inside of a day, which you know? Okay. Possible to solve it in other ways, but Yeah.
Greg:To solve it that quickly, probably not. Seemed like a really big problem, like, so big that, you know, maybe it it it wasn't worth, expanding the effort. But, yeah, yeah, it's it's, it's been really interesting, and there are a few other ideas that we've got kicking around. So a busy time on the on the, on the main job front on on that kind of stuff. And Very cool.
Michael:It'd be interesting to see how that tracks over time as well. See what you plan with with all of that.
Greg:For sure.
Michael:The right sizing of the application. Like, I assume you've got lots of metrics in there that will tell you what isn't isn't actually being used and how frequently so you can Yeah. Pick apart bits and pieces.
Greg:Yeah. So, I mean, the business that at core does a lot of, you know, has a big strong marketing backbone. We we sell we effectively sell a marketing technology for the construction sector. So, there's a lot of attribution tracking that goes through everything. And, probably in the last 6 months, we've, you know, done a full post hoc implementation.
Greg:So for those that aren't aware, post hoc is like a product operating system. So it gives you product analytics, funnel analytics, AB testing, experimentation. So we can run that over. We we we can bolt that up across, you know, a form fill or a feature set and then try and understand what the user journey is through that and then get some ideas on, you know, this feature is used by x percent of the user base. And if that percentage is low and the cost to maintain it is really, really high, we can have a conversation about whether or not there are other reasons to keep features around that that are beyond it it's expensive to maintain and not many people use it because sometimes the just the existence of a feature is a sales hook.
Greg:So, you know, it serves a purpose sort of quite beyond just being useful. But, yeah, we've got we've got some fairly good insight into what and how things are being used, which is is sort of also in service of being able to sort of work experimentally and work in small, like, work in small steps. Because, you know, sometimes the step length is so small that you can't really just go and ask a user or a set of users whether or not it's improved, anything. You have to kinda look at the micro adjustments in the in in the measures to see whether or not you you're tracking in the right direction. So it's super important for our feedback to be able to be able to have that, and we're gonna put that to use in terms in other ways, in terms of, like, you know, should we continue down this path?
Greg:Should we jettison this feature? Should we, invest in this feature? Because, you know, it's it might be incomplete but has some signs of life in it in terms of, you know, if we were to add a little bit more to it, it might become 10 times more useful. So that's, that's a pretty useful tool to be able to have.
Michael:Yeah. Nice. I'll have to check that out. Might be might be Yeah.
Greg:It's a really good it's a they're free. The free tier is actually, like, really generous, like, really, really generous. It's a community. There's an open source version of it if you want to stand it up yourself, if you, like, hate yourself and wanna use Kubernetes, you can stand up your print stack, and they're free tiers.
Michael:Yeah. I know Big Brother has got some like they do some stuff with Datadog, maybe some stuff with Honeycomb and things like that. And they've got some other, I think, 2 different feature flagging platforms. So they've got, like, heaps of stuff in there that we can leverage Yeah. To check out.
Michael:So
Greg:Yeah. We just play with, LaunchDarkly, which is also, you know, that operates in this space, and it it it's just expensive. It's very, very good. It's incredibly good. Very, very useful tool, but it's very expensive.
Greg:And, also, kind of built it's kind of built assuming that you have a long running process and that you can stream events to it. So PHP doesn't play super nice with it, just because of the request response life cycle. You you end up having to have this little Docker container that sort of does the streaming for you that, you know, you you report your metrics into, and it maintains the connections. Yeah. Whereas PostDoc's kind of really front end heavy, so it's it's all just sort of JavaScript events.
Greg:And if you've played around with anything like Google Analytics, it's kind of in that same vein. You trigger events on the front end, and they all flow into a big analytics platform. And you can cut and slice the data however you want, compare it across time periods, bucket analysis, all that kind of stuff. Yeah. Really cool.
Greg:Well, I mean, that that is that you can get that level of power for free is, is is is pretty insane. Mhmm.
Michael:Nice. Right. Well, I think, good conversation there. We'll chat. We will we've got still we've talked about this.
Michael:We've still got some people in the pipeline that we want to have on to interview, to talk through their own Ripple stories. So if that is you, we will get you. If you are interested in sharing your Ripple story, then certainly reach out to us on Twitter. It's probably the best way at Michael Drinder or at Greg Swimmerman or at Ripples FM. You can watch this.
Michael:I mean, if you're watching this on YouTube, then you know you can watch it on YouTube. But if you're listening to this in audio form, you can watch us on YouTube at ripplesfm as well. This has been episode 10. Buy buy early bird tickets to Laracon AU. We would love to see you there.
Michael:Can't wait to start. I mean, it won't be until August that we're putting speaker announcements out, but we've got lots of stuff, lots of exciting stuff in the pipe for this year. So I can't wait to start sharing that. I mean, August will be here before we know We're already almost halfway through the year as it is. So Crazy.
Michael:Yeah. Right.
Greg:If you're in town or you wanna jump on the live stream, we will be live streaming the Brisbane Laravel meetup, as well. But if you're in town, come along because it's better to come in person for these things. Please don't go and sign up on meetup and then just sit at home. If the weather's good, just come across. Have some free pizza and free beer and meet a whole bunch of people who are doing cool stuff and have similar interests to you, it's always a good night.
Greg:And we'll get you home by, like, 8 o'clock, so it's not gonna be like a big party in the middle of the week.
Michael:Alright. Well, until next time, everyone. See you later.
Greg:Bye.