Our B2B SaaS Journey

In this episode, Mitchell and Gavin navigate an App Store rejection as the World Police Games launch gets very real, unpack a new Claude-powered development workflow that is dramatically increasing output, explore how MCP could connect the different sides of SixSides, and celebrate Gavin surviving his 100km bike ride.

Chapters
  • (00:00) - Gavin’s 100km ride
  • (07:27) - Testing week and Apple rejects our app
  • (13:35) - Racing towards the World Police Games launch
  • (16:25) - The new SixSides marketing site
  • (19:21) - HubSpot, Snitcher and our ICP workflow
  • (22:37) - Why Mitch is moving from Cursor to Claude
  • (29:19) - Connecting SixSides to Claude with MCP
  • (33:16) - The real-world ROI of AI development
  • (36:00) - Mitch is moving out of his office
  • (38:31) - Wrapping up

In this episode, we cover:
  • Gavin completing his 100km bike ride from the Gold Coast to Brisbane
  • Testing week for the World Police Games app ahead of launch
  • Why Apple rejected the app while Google approved it
  • Preparing an iOS and Android app for App Store review
  • Adding account deletion and removing unused NFC and background audio permissions
  • Preparing SixSides for thousands of World Police Games attendees
  • The remaining work around athlete CSV imports, HubSpot and welcome email automation
  • Progress on the new SixSides marketing website and design workflow in Figma and Slack
  • Early signs that SixSides blog content is being indexed and found through Google
  • Using Snitcher and HubSpot to identify and follow up with ideal customer profiles
  • Why Mitchell is moving from Cursor towards Claude Code and Solo
  • Using MCP to connect AI agents with Cloudflare, AWS, Sentry and other business systems
  • Cutting infrastructure costs from roughly $2,000 per month towards $300 with the help of AI
  • Building an MCP integration so SixSides data can be accessed through Claude and ChatGPT
  • Using AI to connect product development, marketing, sales and operations
  • How AI is changing the role and workflow of experienced software developers
  • Mitchell's decision to give up his office and return to working from home
Got questions or topics you want us to cover? Email us at journey@sixsides.co

If you enjoyed this episode, please leave us a 5-star rating and a review on your favourite podcast app. It really helps us reach more people!

Connect with us

Creators and Guests

Host
Gavin Tye
Sales and Marketing and Co-Founder of SixSides
Host
Mitchell Davis
Developer and Co-Founder of SixSides

What is Our B2B SaaS Journey?

Join the SixSides.co team as we navigate the highs and lows of building a B2B SaaS company. From finding product-market fit to scaling sales and community-driven growth, we share real insights, tough lessons, and candid conversations about what it really takes to grow a successful SaaS business. Whether you're a founder, marketer, developer, or just SaaS-curious, this is your backstage pass to the journey.

Mitchell Davis:

Hey. I'm Mitchell Davis, CTO and Laravel developer.

Gavin Tye:

I'm Gavin Tye, CEO and a recent 100 kilometer bike ride finisher. My gosh.

Mitchell Davis:

How's that? To that. That's not a bad intro. You can put again, I think you you need, mate, you've got room in the email signature, and I think that's exactly where that should go.

Gavin Tye:

Yeah. Fair enough. I could just do one comma two comma three comma all the way to 100 and go, that's the recent ride I completed. Yep.

Mitchell Davis:

That's a long signature. We are into year two of running our remote startup, sixsides.co. It's a community led events platform. And on this show, we're documenting both the business and tech of our journey as we build our SaaS. How are you, mate?

Mitchell Davis:

Mate, pretty good. Pretty good.

Gavin Tye:

I had my

Mitchell Davis:

100 kilometers. How was that?

Gavin Tye:

It was easier than I thought. Yeah. It was The hard part was getting up at 03:00. Yeah. Right.

Gavin Tye:

Early. Yeah.

Mitchell Davis:

The photos you sent me, it was like pitch black.

Gavin Tye:

Yeah. Well, mate, it was dark at 06:00 still. So yeah. No. It was great.

Gavin Tye:

So I caught a train in from here, like around like, there's a train station not far away. So the drove down there, caught the train, went into the city. It was pretty cold. I it wasn't that bad, though. And then, mate, I made sure I had a lot lot more food.

Gavin Tye:

I'm glad I did that ride last week. Like the one before I was so underprepared and your and then I just course corrected and I was alright. I got to forty k's called Mel. I messaged Mel and said, I was I'm doing pretty good. Got to Mel smack on 60 k's.

Gavin Tye:

I was pretty good. And the last 10 k's was hard. I lost a group of people I was riding with. And when you you know, when you see the Tour de France and people follow

Mitchell Davis:

each other?

Gavin Tye:

Is so much easier to ride when you're with a group of people. And I lost them because I stopped for my cousin Mel's cousins who were on the side of the road. And then you just in the afternoon, the wind on the Gold Coast blows up the coast. It's always up the coast. And so you're riding into the wind and it was the last five k's was brutal.

Gavin Tye:

Yeah. But it was great. I got there, the kids were there, they all made me signs and

Mitchell Davis:

That's awesome. Yeah. Must have felt Yeah. So

Gavin Tye:

Well, Mel missed the finish line and she was around the back and she's notorious for missing the finish. And oh, mate. She missed my first half marathon once. I messaged her and I said, where are you? She goes, I'm still at Paradise Point having a coffee.

Gavin Tye:

Where where are you? And I was like, I'm nearly done. And she's like, oh, fuck. I missed that one. Yeah.

Gavin Tye:

I had him try come on a half iron man once. And then I was running down to the finish line and I see Mel sprinting with Harper on the under her arm, like, following you trying to get to the finish line.

Mitchell Davis:

Slow down so she can catch up.

Gavin Tye:

No. Don't slow down.

Mitchell Davis:

No cheers.

Gavin Tye:

If I see the finish line, want it over. Yeah. That's right. But it was pretty good. It was great.

Gavin Tye:

I saw some really nice roads around from the Gold Coast to Brisbane that I hadn't ridden on before. Yeah. It was it was really good. So

Mitchell Davis:

One. That's great. Yeah. Well, I it's funny. I intended to send you a message on Sunday and say, like, good luck, mate.

Mitchell Davis:

I actually forgot to, is the truth. But I had good intentions. But by the time I would have sent that, you would have probably already been done. When did you finish? How long was the race?

Gavin Tye:

I actually did it faster than I thought I did it in just under four hours, three hours fifty. So I started at ten to six through 05:55, and I finished just before ten. Nice.

Mitchell Davis:

Good one.

Gavin Tye:

Yeah. It was done. Fantastic. Yeah. But as soon as they got to the park, like because there's a big kids playground, one of those max super max playgrounds that the kids play on these days.

Gavin Tye:

And as soon as they they just got there and so as I was done laying on the grass and the kids were playing for ages, and I was I was I was hurting by then. Yeah. Slept for twelve hours last night. It was a great sleep. Yeah.

Gavin Tye:

I don't doubt it. Yeah.

Mitchell Davis:

I saw on your Strava, you broke like it gives you those achievements of like your fastest segment in this area, whatever. You had, like, a 100 and something achievements that you yeah. It was something like that.

Mitchell Davis:

It was it was a lot.

Gavin Tye:

Mate, I was to be honest, I was hooking down there, like, for some some and it was only because I was riding behind groups of people. Like, don't get me wrong, I wouldn't have done it by myself. So I am That helps. Yeah. I'm going to 54 achievements.

Gavin Tye:

Yeah. I had quite a bit.

Mitchell Davis:

54.

Mitchell Davis:

Yeah.

Gavin Tye:

Yeah. But one of the things I really like doing is riding with a group of people. I haven't done it for years and it's really good. So I followed 99 bikes. There was a group about 10 people in 99 bikes.

Gavin Tye:

And I followed them, I reckon, for 30 k's and they were great. So then there was these big Samoans. They had a big JBL speaker where a water would be, and they were just all riding together. And they were pretty fun. It's true.

Mitchell Davis:

Yeah. A good one.

Mitchell Davis:

That's awesome. Well, my weekend wasn't as achievement full as yours, but I did Nicole and I actually did go on a walk last night. It was it was our first one in a little while. It's been probably a week or so. And, yeah, that was a nice six and a half k's.

Mitchell Davis:

So not quite the 100. Yep. But it was a good start. And yeah. So gonna try and keep that going.

Gavin Tye:

So Yeah. Okay. Yeah. Yeah. I'm I'm moving towards now that twelve hour walk.

Gavin Tye:

That's gonna be my next challenge. I'm gonna start walking towards that. Yep. Was good. But yesterday, Mel is so not understanding.

Gavin Tye:

I had a sleep and then she went out and she'd come back and she wanna do some gardening. And I was like, I hope you know that I'm not in a position to sit there and help you. This is a what you're about to start is a three hour job. And she's like, it's okay. It won't be that long.

Gavin Tye:

Fuck. Twenty minutes later, I'm manicing the ground up and and I was well, wouldn't be that hard normally. I was out of breath and I'm going

Mitchell Davis:

No rest for the wicked.

Gavin Tye:

Thanks for being understanding. She was like, are you being sarcastic? I was like, do do what do you think? Yep. Anyway.

Mitchell Davis:

She'll learn one day. Yep. Yep. Get there on the finish line.

Mitchell Davis:

They give you some breathing room after that.

Gavin Tye:

Some signs, and Mel's cousins were there yelling out. I got a good there's a good photo at the end because I had these photographers. And I saw these old friends that I used to do triathlon with. I haven't seen them since Harper was one, and they happened to be there. Mel saw them a second a second before the end and they ran over and I there's a photo of me like smart like, sure.

Gavin Tye:

It was really for the sign, and then I saw Diane's the but yeah. Yeah.

Mitchell Davis:

Nice. If you like, we can put

Gavin Tye:

that as the cover art for the the episode. Maybe. I gotta buy the photo, though. So

Mitchell Davis:

Oh, okay. Yeah. Maybe not then. Anyway. Alright, mate.

Mitchell Davis:

Good one. That's really good. Good on you. Okay. Let's get into the meat of the the subject.

Gavin Tye:

Mate, let's get into the meat and potatoes of this episode. You that's the words right out of my mouth.

Mitchell Davis:

Yeah. Something like that. So Apple rejected our police games app, which was very rude of them. Well, hang on.

Gavin Tye:

Hang on. Set the context, mate. This week is testing week. Fucking hell. Come on.

Mitchell Davis:

Alright. You restart the segment then.

Gavin Tye:

Yeah. Yeah. Come on. Let's You don't have to cut it out, but let's this week has been testing week. We've we set up test flight for the World Police Games team.

Gavin Tye:

We've invited them in, and we're trying to get the clock ticking on the release of the World Police Games app because it can take up to two weeks. Right? So we've kind of done a quicker release than normal than you would have got some things earlier on. Yeah. But I don't think they have rejected and come back, but I also don't think they would have been things that you would have addressed anyway.

Gavin Tye:

Right? So

Mitchell Davis:

Yeah. Exactly. Yeah. So so I've now submitted, I don't know, five plus apps to the App Store, like on other projects and and the Play Store. And typically the first time you're trying to get an app out the door takes way longer than future updates because they kind of just wanna see, it's my understanding, what this app is.

Mitchell Davis:

Is it doing anything that it shouldn't? They then give you a bit more leniency, and I'm guessing they test a lot less thoroughly for apps that have been already released. Right? So, yeah, typically this can take like two weeks or so for the first one, and that's generally I've noticed that sort of delay on the Google site, and Apple generally has been a lot faster. In this case, it was the reverse.

Mitchell Davis:

So we put the app up and then within a day or two, Google had already approved it for production release, which is great. No issues. They didn't raise any issues at all. So maybe their testing isn't as thorough or stringent, but it works in our favor. And then Apple surprised me because it took them a whole week before they actually even started reviewing it.

Mitchell Davis:

And then they have rejected it. So we got some feedback. They've got like a bunch of guidelines that you have to meet. And generally, most apps, it's fine, but they've raised some of these ones. So we needed a way to support, like, deleting your own account.

Mitchell Davis:

And fortunately, we had done some work on that already, but we hadn't rolled it out yet like two months or so ago. And so I was able to take that work over the weekend, leverage Claude, which I'll talk more about in a bit, and then get that rolled out. So we now have some UI in the app and the back end process to support it of deleting your own account. So that will check that box. We had a couple of, like, extra packages or functionality in the app that we weren't using yet.

Mitchell Davis:

So one of them is the NFC tags, which is like those, you know, the pay tag things for like Apple Pay and whatever. You tap your phone on top of some little thing and then it'll do something. We had that functionality kind of under the hood of the app, but nothing that actually used it yet. And that was me kind of future planning because we will at some point wanna do something like that to tap on a sponsor's table or whatever, you know, and have that trigger some points or something. But given we've raised that permission that we needed that already in the app, they then want a demo of how it works, and they want like a video of you doing that with your physical phone tapping a physical NFC tag.

Mitchell Davis:

And, of course, we're not ready for that. So I've had to pull that out. Yeah. They raised something about like background audio, which is like how Spotify and Apple Music and that can play music even while your phone's locked. And we didn't need that at all.

Mitchell Davis:

I don't know how that got in there, but we took that out. And then, yeah, the last one is the demo account that we give them apparently didn't have access to any events or communities. And so they're looking at basically an empty app, and that's something that I should have realized before putting it in. But there's been a million other things to worry about. So, yeah, I'll just go through and that can become a part of the regular checklist basically before doing any releases.

Mitchell Davis:

Just make sure any of the functionality they're gonna wanna actually see is full for their accounts. And this goes for Apple and Google. Right? They're using two different accounts, but we can do the same thing for each of them. So yeah.

Mitchell Davis:

So those are the things that they raised. And then on Saturday, it's now Monday as as we're recording. So on Saturday, I spent the whole day basically working on that deletion, user deletion, and fixing up some of these things. Yeah. I think we'll the plan now is to get this get a final build done on Wednesday or Thursday at the very latest.

Mitchell Davis:

And then that will include a bunch of changes that Raymond's been making to fix some of the the UI in the app, some changes that I've made in terms of some styling and and some of these new things that I had to add like deleting users and stuff. So, yeah, we should be looking at getting the final build on Wednesday or Thursday at the latest and then submitting that for review on both stores. I'm gonna do an expedited review with Apple, which you can do. You can ask them, hey. This is important for a date.

Mitchell Davis:

Can you put this through review? The general idea is you don't wanna do that too often or they they they'd start ignoring you. Every release can't be super important, you know, but this one is. So, yeah. So that's the plan.

Mitchell Davis:

That's what this week should look like. It's pretty exciting. Right? How are you feeling about it all?

Gavin Tye:

Oh, look, I think we're well within our timeline. I was looking at the timelines today because we've got a timeline to the police games at the moment, the week of the seventeenth. We're now in we're now into the week of the twenty fourth. So I think we're I'm a little bit behind in my email welcome series, which I'm comfortable with. I can knock that out.

Gavin Tye:

I'll do that either today or tomorrow. Then we gotta give Ash access to Ash once you're ready for that. So Yep. But yeah. Look, I think we're yep.

Gavin Tye:

Yeah. I think we're doing pretty good. There's a couple of things that that we haven't discussed yet around uploading the CSV with the athletes and their

Mitchell Davis:

Yeah.

Gavin Tye:

Which is what it's reasonable project. Yeah. I don't think it's

Mitchell Davis:

It is. Yeah. Yep. Yeah. So I I do still that's like the last big project that I have.

Mitchell Davis:

So I need to get the app kind of finished. And then, yes, this week I have to deliver on that this Yeah. Sure. It's time.

Gavin Tye:

The other one that we haven't spoken about is getting new users into HubSpot because that's where they're welcoming and tag them appropriately. Right? Whatever that tag may look like. It'll probably just be an attendee of a world police games. Right?

Gavin Tye:

Something like that. Like an attendee. Because that's where the welcome email series will come from.

Mitchell Davis:

Yeah. I'm sure we can put that together pretty quick. I'm not

Gavin Tye:

too worried about that. No doubt. No doubt. So, look, we've got lots of different things and yeah, it's going pretty good, I think.

Mitchell Davis:

Feels like Yeah. I wish we were a bit further ahead. It feels like we're chasing slightly behind, but not like not too far. You know? I'm a little stressed.

Gavin Tye:

Yeah. Yeah. Yeah. Got it.

Mitchell Davis:

Yeah. I wish everything was done already. It's not, but we're pretty bloody close. So, yeah, we're we're within I feel comfortable that we're well within our timeline. I just it's like,

Mitchell Davis:

oh gosh.

Mitchell Davis:

The app is potentially going out a week from now to five Yeah. Thousand

Gavin Tye:

We've built in some redundancy here. Right? So a little bit, not a huge amount. So as long as it's out by the thirteenth and and it's everything is working. So yeah.

Gavin Tye:

Anyway. But it's good, mate. Like, I think we would would have we taken this position six months ago and you go, fuck, yeah, we would. Would we? Oh,

Mitchell Davis:

yeah. Yeah. Absolutely. Yeah. No.

Mitchell Davis:

It's good. It yeah. We'll we'll get it there. It's just

Gavin Tye:

One thing. How's it going with Robin and the design project?

Mitchell Davis:

Good. So the he's given us back kind of the final wireframes of the the landing page on the marketing site. So this is a project that, yeah, we kicked off earlier this month, and he's now, I think, into week two of it. And, we've got the we've got the wireframes and he's now working on actually like fleshing that out. So he's happy with we're happy with the layout and everything like that.

Mitchell Davis:

I think there's enough variation there. I really like the design on page. Lots of white space, which is something that I've always really liked. Just keep it really simple. I think it's gonna look pretty good, pretty modern.

Mitchell Davis:

I mean, you'd hope so. But, yeah, all in all, it's going pretty well. It's it's pretty nice working with him just on Slack and, like, using Figma and stuff. Like, feels so much more modern than, like, a back and forth email thread of, hey, could you change this or do you like this? Whatever.

Mitchell Davis:

You know? Yep. It's a lot more efficient. So, yeah, it's pretty cool. It's it's a way that I would like to try and work with my clients in the future is to try and adopt some of that stuff.

Mitchell Davis:

Yeah. It's just it's been a bit of an eye opener.

Gavin Tye:

Yeah. Because you can enter any time, right, and you're not taking up. Yeah. It's actually pretty good. It's good for him on his side of the fence.

Gavin Tye:

Right? So

Mitchell Davis:

Yeah. Like, Slack is not really any different from email. Right? It's still asynchronous. I can reply whenever I want.

Mitchell Davis:

It doesn't matter, You know? But, like, it's nice to kind of get away from email. Yep. Yeah. Yep.

Gavin Tye:

I was just looking at Snitcher. Our our blogs are starting to get picked up by Google.

Mitchell Davis:

Okay.

Gavin Tye:

So I was looking at Surrey County Council. They've they've was referred by Google, the hidden problem behind low event engagement. So people are starting to actually get referred. And then I'm also looking at there's a Adelaide University of Adelaide. They people are going to they're looking at the design of our like, which is what you created, but they've also looking at contact us.

Gavin Tye:

I think we can do we'll change all that up to contact us so to submit a form and stuff. Right? Yeah. And also the I guess the chat functionality is not popping up either. Maybe because I've changed platforms and and scaled it back.

Mitchell Davis:

That's probably why. I think you just need to configure that in HubSpot and then it'll pop back up.

Gavin Tye:

Yeah. Yep. So that yeah. That that probably brings us to HubSpot this week is I'm still going forward to I'm configuring it quite a bit at the moment. I showed you a little bit about that last week.

Mitchell Davis:

Yep.

Gavin Tye:

So How's

Mitchell Davis:

that going?

Gavin Tye:

Oh, there's lots of different things to do. Getting one one thing that we spoke about last week was snitches not set up on wasn't configured properly, but I've now configured that that if any of our ICPs come to our website, then it'll actually flag it on Slack and and let us start straight away. But then I'm also implementing a workflow that it will open up three to five ICPs. And then I'll send them an outreach email to say, hey, look, some of your team may have been on our website to try to kick off an engagement. But now I've gotta try to filter between those in our CRM and those not because because I don't want to do that to people in our CRM.

Mitchell Davis:

So

Gavin Tye:

Yeah. Yeah. It's coming along. When

Mitchell Davis:

you say ICP, that's insane clown posse, bro.

Gavin Tye:

Yeah. Yeah. Yeah. Yeah. International clown posse.

Gavin Tye:

Yep. Ideal clown posse. Insane clown posse.

Mitchell Davis:

That's the band. Do you know the band?

Gavin Tye:

I think it's ICP. Yeah. You know me. Right? Isn't that a lyric?

Gavin Tye:

Yeah. ICP. O OPP. Okay.

Mitchell Davis:

Oh, mate. Talk about a dog's breakfast.

Gavin Tye:

Talk about a boomer.

Mitchell Davis:

That's your generation of music, though. Like, it's

Gavin Tye:

the thing am OPP. I'm as illiterate with music as it comes. Yeah. I of all the thing I'm right to think I'm reasonably knowledgeable on certain things, not music. You are killing me with your with your lack of musical references.

Gavin Tye:

It's it hurts. Do you know what I put it down to? When I when we were young, we were so we were pretty poor. We had no no stereo. And and when we had a car, was only a shitty old car that had an AM radio.

Gavin Tye:

So there was just no way for me to listen to anything. And so in my influential years from say 12 to 16, no radio, no music whatsoever.

Mitchell Davis:

You got nothing. No nothing. Okay. So now I'm the asshole because I'm teasing you about

Gavin Tye:

Way to bring it up. And I used get beaten a lot as a kid. So yeah. Thanks, mate. And now you're laughing at me.

Gavin Tye:

It's just getting you.

Mitchell Davis:

Oh, mate, mate. I think we just killed the podcast here. That's you done me in here. Oh my gosh.

Gavin Tye:

Oh, thank you. Now we now have your emotion of blackmailing me. Tye, so you're bringing it all up. Dog, that's fantastic. Fucking cat's lighter.

Mitchell Davis:

But alright. Alright. I think we should move on.

Gavin Tye:

Moving on. Yeah. Moving on. Dashboard is where the blogs are starting to to what's it called? Index, which is good.

Mitchell Davis:

Yeah. Okay.

Gavin Tye:

Look at me. Sounds like I'm talking about.

Mitchell Davis:

Yeah. Yeah. That's great.

Gavin Tye:

That probably could run to the chat we were having last week about I'm starting to use Chord and connecting HubSpot with Snitcher and then even potentially Apollo in doing some outreach there. And you're using Chord quite a lot as well. And I think we were discussing that we think it's a really good opportunity to bring your side and our side, left and right brain together through Claude. Right? Like at this point in time.

Gavin Tye:

Yeah.

Mitchell Davis:

Yeah. That's right. So I've been working so long term listeners of the show would know I use cursor a lot. Over the last month, that's changed. So I'm now using a mix of two tools.

Mitchell Davis:

One is Claude Claude code specifically, and the other is solo, which is soloterm.com. And Solo basically is like a wrapper around Claude Code or any of the other like coding agent tools that you can run on your computer. And it gives you, like, the ability to have to have one agent talk to another and kind of plan things out or ask questions to another. And that's quite powerful when you've got code bases, like multiple code bases like ours. So we have a code base for the web back end, a code base for the app, and then another one for our infrastructure as well.

Mitchell Davis:

And it's really cool to be able to have them all talk to each other. So strongly recommend solo and

Gavin Tye:

What made you drift from cursor? Like, what was it impetus that made that happen?

Mitchell Davis:

Mainly what I just talked about of, like, it was difficult to get one thing to talk to another. And also, I've had like pretty good success lately with like the quality of the code that Claude or Anthropix models were giving me and trying to run those from Cursor is expensive. Like, we had an overage of like $200 or something on Cursor when I started using Claude models. And then I can just go use the Claude plan instead, use Claude code, and then that is all included. Right?

Mitchell Davis:

And so I think I'm now at a point where I'll probably get rid of Cursor. The only thing that would maybe keep me on there is they've just announced this new Cursor's been acquired by x AI, which is SpaceX, and they've got this new Grock bot thing, which I've heard a lot about recently on on a couple podcasts of people using it. I wanna investigate what that is before I fully give up cursor, but that might be something that you and I can then use to do certain tasks for us in the background. Seems like it's kind of a hosted agent experience. So I'll have a look at that.

Mitchell Davis:

But if if that isn't worth keeping, then I think I'm at the point where I'd just get rid of cursor because I don't think I'm I need it anymore.

Gavin Tye:

Yep. Okay.

Mitchell Davis:

So yeah. So this new flow of using Solo and Claude has unlocked me to be super productive. So on the weekend, I mentioned I did a bunch of work on six sides and getting through some of these App Store issues, but I also was able to work at the exact same time on some other projects that I'm involved in. And specifically using like the connectors, they call it in Claude. It's this m it's MCP basically where you can connect Claude to, you know, in my case, it's HelpScout for support tickets on this other project or to Cloudflare to go look at our logs.

Mitchell Davis:

And I'm working on some security rules at the moment so that anyone that's trying to like spam our domains and see like, hey, do they have a index dot PHP or, you know, what any hidden files, I can detect that and block it. And that's something that Claude can hook up for me because it knows it can talk to Cloudflare. Same thing for AWS, our hosting. Like, I had it work on a database cost problem that we've got on another project, And it's cut our costs down from like somewhere around $2 a month to I think it will end up being around $300 a month. Yeah.

Mitchell Davis:

Right. Which is insane. Right? By having access to the database metrics on AWS, the logs in Cloudflare and then the code base. And so it can kind of do all those things and then roll out changes and we can test them, and it was an immediate drop.

Mitchell Davis:

So, it's really the If you're not using MCP tools to connect to your systems that you're using and have your agents know more stuff, like you're missing out a lot. Same same for Sentry as well. Like Sentry is what we use for bug tracking and that Yeah. That's huge to be able to have it all just hooked up. So, I might be late to this, but I reckon there's a lot of people that are way further behind than me, and I'm telling you, it's here.

Mitchell Davis:

Like, it is very cool what you can do So

Gavin Tye:

in saying that, like, you worked over the weekend and so you were doing two or three things at the same time. Without AI, just for context, how long would it have taken you or could have you done what you did on the weekend without it? Like

Mitchell Davis:

It'd taken me a whole week. Easy. Yeah. Like maybe more. But yeah, this database thing in particular, yeah, could have easily taken me a whole week, you know, and it took it maybe three, four hours in the background of like trying things, testing it, rolling it out, and then repeating that loop.

Gavin Tye:

So, you just side note, you told me about this, that you showed me this, that bit of work you did on the weekend. This is a blog post that you should put out Yeah. Around reducing costs. And I also think it's probably an app like your site, your other business is something that you could offer other businesses and audit like to

Mitchell Davis:

Yeah. Yeah. Yeah. Like the MCP stuff is gonna be half of my talk that I give at LaraCon. Yeah.

Mitchell Davis:

So, yeah, this is giving me more like stuff to talk about in that. So yeah.

Gavin Tye:

You set up six sides to have its own MCP, right, to connect in the cord?

Mitchell Davis:

I sure did. Yes. So, we now have an MCP. It's not documented, and only you and I are using it. But, yes, our customers can now add six sides to their Claude or ChatGPT.

Mitchell Davis:

It works as well. ChatGPT makes it a little scarier. They call it like developer mode and it just seems silly, to be honest. But anyway, yeah, it's a really simple, easy process. And then you can talk with your data in six sides.

Mitchell Davis:

It's very similar to we have a conversation area in our dashboard and- But you get the benefit of it being in Claude is that Anthropic and OpenAI have spent billions on making that like the best interface to talk with an agent. Yeah. We haven't. Right. So, we can you can use their agent, but our tools to access our data.

Mitchell Davis:

It's like it's incredible. This is so powerful. So, I'd suggest that's probably the way that you should set up any events that we have for the next little while. Like, I I know obviously we're we're getting customers to set up their own events as well. But if ever there's anything that you need to get in and work on, I'd suggest trying to do it in Claude.

Gavin Tye:

I think we run the because they got a convention attached to the World Police Games. Right? A couple thousand people are going to it. It's our biggest event that is bundled in, which we've never spoken about. I think we set that team up, Aaron and Neil, and we show them how to do this through the MCP.

Gavin Tye:

And we I'll I'll work with them on it to as a living case study on on using it. Right? And but in all that, tying that all together, like one of the things we spoke about last week was we could do marketing or pull marketing content out of your work without because I I don't know what you're working on to the minute detail. Right? So but we can pull it out a linear and go, hey.

Gavin Tye:

We wanna do a post around this or HubSpot is referring us to giving us ideas on how to trend on AEO in LL LLMs. And we could pull some of that stuff out of what you're doing. And I think it'd be a way to unite the business. I think it'd be really And we did speak about that. The final thing we spoke about that last week was we need to visually represent all the systems that we have in the business so we know what we're talking about at the moment.

Gavin Tye:

It's in your head, in my head. Yeah. And we need to know where to go. Yep. Yep.

Gavin Tye:

Yeah. Okay. It's awesome. It's exciting, mate.

Mitchell Davis:

It is. Yeah. It feels like it's finally like an unlock, you know. Wow. This is all the stuff that's possible.

Gavin Tye:

When we first started working together, you're like, no, I'm more like an artisan developer. I don't want to use any AI. I really pride myself on my code base. Like, I really love it. Like, I like to hand hand chisel my knives, handcraft my knives, and and I'm a real blacksmith with banging out code.

Mitchell Davis:

That's what I said, did I? I don't really say it any of

Gavin Tye:

applied it. Yeah. Yeah. Okay. But almost I'm almost paraphrasing, but it's pretty close.

Mitchell Davis:

Yeah. Sounds exactly like me. Yeah. No. Look, the My rebuttal to that is I do still feel that way, but that's not the reality that we live in anymore.

Mitchell Davis:

Like, things move quickly now so much quicker than they did before. And I just simply don't have the time to be a manual developer where I write the code. Like, it's it's not the way the world's moving. We without AI, we wouldn't have been able to put in this bid for the police games. Like and we've known that.

Mitchell Davis:

Right? So it's like I have to use these tools. And it it does mean that there's often a less quality aspect to it because I'm not as stringent as I would like to be with reviewing every line of code and making sure I'm not adding any obvious bugs. But in general, I'm testing everything that we Yeah. Put out and then monitoring for changes and issues in Sentry and whatever.

Mitchell Davis:

And so in general, it's gonna be pretty good. And the quality of the models has continued to improve, you know. So some of the stuff that was getting produced eighteen months ago when I was first starting with some of this was seriously pretty bad. And now a lot of it is actually pretty good.

Gavin Tye:

So Yeah. Yeah. When you look at like when I look at what you do and stuff like that, is the cream is still rising to the top. Like, started with cursor and said cursor is probably good for a mid level developer or a lower end developer. But you are quickly your skill set is quickly outstripping that and you're moving into other things.

Gavin Tye:

You're going up. You're automatically going to the top and your output. Your quality of outcome is probably the better term here. The quality of your outcome is significantly like within three hours, you just saved yourself $1,800 a month, like $20,000 a year. Right?

Gavin Tye:

Yeah. You're actually turning AI into a practical outcome. Actually listened to a podcast the other day, and they were still talking about they were talking about the ROI of AI is still largely theoretical at the moment. Yes. You get better productivity and stuff like that, but it hasn't really translated yet into businesses getting 30% more profit or revenue.

Gavin Tye:

Yeah. And you're actually doing that. Right? And it's and it's your I think the quality of the code over time will become less less of a focus because AI models will will increase to, yes, we know it's perfect. It's how do you apply that and get the outcome, which I think is really interesting.

Mitchell Davis:

Yeah. Me too. So, yeah, pretty pretty big weekend and nice to see like, okay, this is feels like this will be my future for the next year, and I finally feel okay with it. So we'll see. See how that goes.

Mitchell Davis:

See if that continues. Or if in a couple months time, I'm like, god, this sucks. I hate this. That might happen. Right?

Mitchell Davis:

But I had a good weekend with it. So That's good. Anyway So

Gavin Tye:

the benefit of you haven't really furnished your office has really come to a has really paid off. Right?

Mitchell Davis:

It has. So, yes, I've made the decision that I'm actually gonna move out. Yeah. I've been thinking about this for a little while. And, yeah, I'm not really leveraging this space at all.

Mitchell Davis:

It's nice to not be at home all the time, but I am happy to to at least give it a try of working back from home. I've been doing that off and on for the last couple weeks. Last week, I worked from home three or four days, I think, to just kind of transition out of this. And, yeah, this will be my last week in here. So Yeah.

Mitchell Davis:

I'll go give my notice today. I'm nervous about it because it's like this it's a nice office that I've got, but and to give it up, but I think it just has to happen. It's the right move for me right now.

Gavin Tye:

Yeah. Yep. So where are you gonna go? Back home?

Mitchell Davis:

Yeah. Just home. Yeah. So I've already started thinking about what like, how I wanna update the layout of the office room. We've got one of the you know, we're only in a three bedroom.

Mitchell Davis:

So one of those bedrooms. Yeah. Just just make it work. So see how we go. It's exciting, though.

Gavin Tye:

We are probably heading into warmer weather here, but it would be depending on how keen you are to walk during the day or early or whatever is maybe we can structure we we started talking about catching up more. So maybe we could turn one of our meetings into a walk in meeting or something to get you out of the house. Right? Yeah. Get us out of not you.

Gavin Tye:

Get us out of the house so we can at least don't get stuck inside all the time. I don't know if you're open to that.

Mitchell Davis:

Yep. Yeah. Absolutely. Yeah. I'm looking at doing two walks a day plus so one by myself, one with Nicole, and then also walking the dog to trying dogs to try and get them try and get myself moving more.

Mitchell Davis:

So, yeah, I would love to do that. We could start that like next week would be great.

Gavin Tye:

Alright. Cool.

Mitchell Davis:

Well, we both got a hard stop. So why don't we wrap up now? Where can people find you online, mate?

Gavin Tye:

Mate, you can find me online on LinkedIn, Gavin Tye, t y e. You can also find the Changemakers podcast, which we released our third episode last week, which is about Paparan bringing it to Australia with Tim Oberg. Yeah. So that's a yeah. That was a that was a really interesting podcast.

Gavin Tye:

But, mate, what about yourself?

Mitchell Davis:

LinkedIn, Mitch Dav. That's where you'll find me. Also, Strava. My ever long campaign to get more followers on Strava for some reason. I just think it's a bit of fun.

Mitchell Davis:

Yeah. I guess that's it for this week. So catch you next week, and we'll we'll give you an update on where we landed with the final Build of the Police games.

Gavin Tye:

Alright? Cool, mate.

Mitchell Davis:

See you, mate.