Slightly Caffeinated

TJ Miller (00:00)
Hey, everyone. Welcome back to the slightly caffeinated podcast. I'm TJ Miller. So, Chris, what is new in your world?

Chris Gmyr (00:05)
and I'm Chris Gmyr

Oh, Just more of the same, really. It was nice. We didn't have any real events going on this weekend because the past two weekends we've been traveling and stuff. So it was nice to just kind of chill at home, get a few things done, do some cleaning and catch up and all that. And then at work, just still very much into the onboarding world, like with the company, with the engineering team, with the projects that our new team was

slated for. So just trying to get the lay of the land with all the things, new culture, like all that. So one of the things that I thought would be interesting to talk about real quick is I've been trying to utilize Cloud more, and more specifically like Claude projects. So we have a lot of decent documentation around like some of the apps and services and even like some of the data models.

but they're a little bit old. We have newer videos of people doing walkthroughs and stuff like that and some of the more recent meetings. So for this one specific application, I've been making a new project. with each of the projects, can have so many files uploaded. So usually text files, things like that. And then you can

TJ Miller (01:26)
Mm-hmm.

Chris Gmyr (01:28)
create chats based on those files or additional context that you import into those individual chats. So I dumped a bunch of this information in the global project and basically said like, hey, given this past information that is in the artifacts and this new information that came out of this more recent video transcript or a meeting or something like that, or this newer documentation, show me what

the past architecture and data flow is and what is the newest or the changes. And that way, I was able to generate a bunch of additional documentation for how everything is currently flowing through the system and related systems. And then also, it's going to be a lot easier to pull that out and continue to work on it.

and make the changes that we need to. So I asked Claude to describe a whole bunch of things and generate Mermaid.js files and diagrams to map out the data flow, services, and a few other things like that. And it did a pretty good job. with Mermaid.js, you can just extract those and

TJ Miller (02:27)
Nice.

Chris Gmyr (02:39)
put it in ID or whatever, or put it back into Notion or Confluence, and it'll just be editable in the future. So I'm hoping to get a lot more solid documentation around just the workflows, the connective apps and services that it used, dependencies, data flows, workflows, stuff like that, and start putting all that stuff back into the application repo and just being the source of truth.

there and hopefully with like the mermaid JS files make it easy for people to use and update them in the future. So been tinkering with that whole workflow. I know we talked a little bit about Claude and projects before with blog and creative writing. So it's a little bit different, but there's like some really cool use cases with Claude and using it like this. So it's really cool that you can just dump a bunch of stuff in there and then start asking a bunch of chats and

TJ Miller (03:08)
Yeah.

Chris Gmyr (03:31)
getting a little bit deeper dives into whatever you put in there.

TJ Miller (03:35)
Yeah, dude, I love it. That's actually what I did. I was using Cloud projects this past week too. I use it for all sorts of stuff. So like I have like a persistent Prism project where I upload like typically before I dive in, like I'll upload like all the documentation and the entire Prism code base fits inside of like projects global like context uploads. So I can literally put all of Prism in there and then just like go to town asking questions.

generating documentation, suggestions for refactors. I'll do that a lot too where I'm like, hey, I don't really like this. Give me like three different ways I can architect this just to kind of get some like inspiration for what my options are. But at GeoCodeo, I recently released a change. Right now, the way that the product works is that you supply the API key as a query argument in the URI.

And that's worked fine forever, but for our enterprise customers, we wanted things to be more secure. So we opted to also enable like authorization header authentication. So you can pass your key in either of them. If you're an enterprise customer, you have to, and we'll throw an exception if you don't. But for our like regular tier customers, it's optional. But all of our

libraries. So we have like a node library, a PHP library, and a Ruby library. And there's a couple other out there, I think, but we don't maintain those. So the ones we do maintain, I had to update all of those to use header authentication. That way it continues to work for regular customers, but also now these tools will also work with enterprise customers. So

I can't remember the last time I've written a lick of Ruby. So it was like, how do I install these dependencies? How do I run the tests? I'm like, I don't remember how to do any of this. So I just uploaded the whole library to our cloud project. And I was like, one of two things I'm like, well, how do I like, give me a script to bootstrap this so I can at least work with it. And then furthermore, like just go ahead and make the updates for me. Like turn this into header off.

And it did an excellent job of doing that. I was able to knock out all three of our libraries in the afternoon. It was fantastic. So yeah, I think your use case is perfect. Being able to just upload that stuff and ask questions to Claude, it help you drive out some of those things. Because that's really what it's going to be. That's something that LLMs are really good at. And Claude's particularly good at that kind of stuff.

Chris Gmyr (05:42)
Nice. That's cool.

TJ Miller (06:01)
Did you use, did you happen to use the extended output or are you just using the like regular output?

Chris Gmyr (06:08)
extended output with artifacts.

TJ Miller (06:11)
Yeah, well, so it's, it'll have the artifacts, but then there's also, it's like in the UI, you have your input and then below it, it'll tell you like, there's like a little dropdown for which model you want to use. You can also choose the output type, whether you want it to be like concise output or explanatory output, but there's also an extended output with thinking and reasoning. So it'll actually like, that's what you typically want to use with coding.

because it'll have a whole step where it like thinks about and does reasoning before it starts outputting code. So it'll have this whole like thing where it's like having thoughts about, you know, the tasks at hand and like what it needs to do and maybe the different steps involved in doing that. And then it goes off and performs the work. So definitely recommend like using that for more, not just coding, like anything that I think is going to be like,

Chris Gmyr (07:00)
OK, cool. Yeah, I just had the, yeah.

TJ Miller (07:08)
of a complex or compound task, definitely use the extended reasoning and thinking.

Chris Gmyr (07:14)
OK, cool. Yeah, the initial chat that I had was just set to normal. So I'll have to choose the extended, which actually says best for math and coding challenges.

TJ Miller (07:26)
Yeah, yeah. Anything that feels like it's going to be potentially multi-step or something that you just feel is like complex, or you're going to maybe ask more than one thing of it. think the ex like moving to that extended thinking output mode makes a ton of sense. And I've, I've found it to make

It 100 % makes better coding decisions doing that. Before they released the Extended Thinking, granted, they've moved from Sonnet 3.5 to 3.7, so we've got a better model all the way around, but that Extended Thinking does make a significant difference. So something to try out for sure.

Chris Gmyr (07:48)
Nice.

Cool, and then anything with the style. So there's a normal, concise, explanatory, and formal. So I've just basically kept it as normal. I expect that that's probably fine, unless I want to change it to explanatory, where it could go into more detail on how it came to these results, or explaining exactly what it did, or what it found. Is that correct, or what have you seen with that?

TJ Miller (08:31)
I've not used any of that. actually, for all of my, there's, so like in a Claude project, not only can you define context, but you can also define like custom rule set. And in that custom rule set, I have two different things. I have the Nova character base prompt from Collaborative Dynamics. So I kind of already control the like style of output through that as a base prompt.

So I've never tried additionally changing that with Cloud, because I just don't want to get conflicting results or something. So I have Nova in there. And then something that I also do additionally for coding style projects, like if I'm working on Prism, I actually take the cursor AI rule set from PushPack's Larisonic project, and I dump that in there too. And that's got

all sorts of coding style rules. So it talks about using Pest for tests, different object oriented stuff. Like it's just got like a whole bunch of rules in there. And Claude actually picks up on that really pretty well. So with that all shoved in my custom instructions, if I just go into the Prism project and ask it to like, just like, hey, write a test for OpenAI's

something or other, right? Like it will automatically hop in and start writing a Pest test rather than like randomly choosing PHP unit or Pest or something else. Like it already knows what all my preferences are for those things. So I found that like also incredibly helpful as like giving it coding style rules. And now the Larisonic stuff is definitely very specific to like PHP and Laravel.

But you could, I'm sure you could go take a look at that and kind of like extract things for Node and also like Ruby. Or you can probably find other projects that like have these like cursor AI rule sets. And it's really just like a series of comments. I threw it in there once on a whim and I was really impressed with how big of a difference that all made. So something to check out for sure.

Chris Gmyr (10:37)
Nice, yeah, I'll check that out. So yeah, that's about it for me. So what's new in your world?

TJ Miller (10:43)
man, jeez, I always struggle with this because I have the memory of a goldfish, so I never remember what I do. So I guess things that I can think of.

We had our Science Olympiad practice tournament, which we can talk a little bit more about later. We had that on Saturday. That was a mostly all day adventure. So we had that. then I've been like deep trying to like, haven't been getting a lot of

work done on Prism, but I've been thinking very deeply about Prism quite a bit and like different things that I want to maybe pursue on it and kind of priorities for what's next. There have been a couple of contributors working very diligently. Chris has been just crushing it. I swear the last like three or four releases have been almost solely like...

stacked full of like five or six PRs each of Chris's work. So I'm massively appreciative to Lericon India happened. That was super cool. Pushpak. I'm to see Pushpak's talk because it was on Prism. So the really, really cool picture came out of he had posted a couple of pictures from Lericon India.

Chris Gmyr (11:48)
Sweet.

TJ Miller (11:55)
One of them behind the stage had like Prism, like big in text on the wall behind him. And man, was something to see, dude. My mind's still totally blown about it. Yeah, it's nuts, man. Really, really cool. So looking forward to, I'm hoping videos from Leracon India get released and...

Chris Gmyr (12:03)
Nice.

Yeah.

TJ Miller (12:15)
I can catch that talk. I'm sure it was great. So yeah, that's really all I've had going on. Nothing super cool or crazy lately, but that's also been kind of nice to have a little break. And we're just jamming over at Geocodeo right now. And so that's been fun too. We've made some progress on ClickHouse stuff. Huge shout out to the boys over at Honey Badger.

We'll talk about that a little bit more later too. Yeah, man.

Chris Gmyr (12:41)
Yeah, that's awesome. Well, before we continue on with the bigger topics, let's go into our little coffee talk segment. And I thought an easy, hopefully easy, fun question to kick us off is, do you have a favorite coffee cup or mug or accessory or something like that that you use on a daily

TJ Miller (12:48)
Yes.

100%. I do have a favorite mug. So like my, have favorites for like each type of thing, right? So I have like a mug, like a ceramic mug. I have a preferred one of those out of our like set of ceramic mugs. And mostly just because it's like all the rest of them are like a like sort of cream color, but this one's matte black and it says witches brew on it.

which is brew is kind of the text is whatever it's the fact that it's matte and black that it makes it my favorite. but my daily driver, I use it every single day. Almost every single cup of coffee is the Yeti mug that they gave to speakers at PHP tech last year. I love it. It's I, since I came home from that conference, it has been my everyday coffee cup.

pretty much more or less any cup of coffee I have is going through that mug. So yeah, that's definitely the go-to. How about you, man?

Chris Gmyr (14:01)
well first, I love

TJ Miller (14:02)
It's my first Yeti thing.

I've had many other brands of things, but this, I like it a lot.

Chris Gmyr (14:07)
Yeah, yeah.

Yeah, I have a bunch of the like I have two like 32 ounce tumblers and then like a 24 ounce. One and then I have a big cooler one of their Tundras with the wheels on it and that thing has been so sweet. We travel a lot for like holidays and stuff like that. We split up a bunch of the food responsibilities so a lot of times for like Thanksgiving.

TJ Miller (14:24)
cool.

Chris Gmyr (14:34)
We'll bring the turkey because we usually get pretty decent ones from ButcherBox subscription meat company. And I just throw it in there and it's like completely good. Like by the time we get to either New York or Virginia or Florida or whatever and put a bunch of other stuff in there and it's just been solid. If you get the Yeti ice backs and put a little bit of ice on top of it, those things last for like a week. It's crazy.

TJ Miller (14:58)
Wow.

Chris Gmyr (15:00)
Yeah, so we could have like a whole episode on Yeti stuff and drinkware, coolers, cooking stuff, whatever. Probably need a whole different podcast for it. So we'll skip over that right now. Go back to the question. My favorite, at least current, daily driver is by KeepCup. And it's basically like a glass.

cup. I think this one is like a 10 or 12 ounce, so relatively small. But you can customize the band that goes around on it. You can get either a cork or a silicone one. And all the silicone ones are different colors. And then you can get different colored caps. then because there's an opening to drink through the cap, there's a little plug thing that goes on top of that.

get different colors for all those. So what I have is similar to your favorite one. It's all blacked out. It's the glass cup. It's the black band, black top, and black little toggle thing to close it up on the top. And it just looks super cool, especially with a bunch of coffee in it. So that's my go-to.

TJ Miller (16:00)
Nice man, I'm looking at these keeper cups, these keep cups right now. This is like pretty neat, man.

Chris Gmyr (16:06)
Yeah. when we started, because my wife and I both have one, but we've got it. I don't know. think we did the Kickstarter, or I think that's the company that did a Kickstarter way back when they started. So we've had these, at least the glass cups forever. And every, I don't know, five years or so, we'll just replace the tops and the little toggle closer thing on the top, because those will get

a little worn out and not be as tight anymore. So we've changed up the colors. And I just keep on going back to the all blacked out version just because it's just cool. I don't like it.

TJ Miller (16:43)
Yeah, man, these are neat. Makes me want to buy one for sure. Yeah, I'm looking at the glass ones right now. And then they've got like, so they got like the glass ones, right? But then they also have, it looks like metal travel ones as well.

Chris Gmyr (16:46)
You totally should.

Yeah, they have a whole bunch of stuff now. They have water bottles and different canteen type of things and different size coffee ones. And they've definitely expanded. They've definitely expanded a lot over the last couple of years because I just replaced our tops and the little toggle things on the top probably about a month or two ago. And I was surprised coming back to their site because I don't

TJ Miller (17:04)
You know, like Tumblr's and...

Chris Gmyr (17:19)
go there that often. So I'm like, oh, wow, they really expanded their product offering. And it's like, oh, if you ever need new water bottles or something like that, it might be something to check out.

TJ Miller (17:29)
Yeah, so do you like travel with the glass ones or do you have like a something specific for travel?

Chris Gmyr (17:35)
For travel, we'll typically do the Yetis. We'll bring those around because then they can sit a lot warmer in the car, especially for longer road trips. Free COVID, we used to take the keep cups to coffee shops and they used to fill them, but it's kind hit or miss now on places that actually will take personal reusable cups. It was just something that we tried to

you know, cut down on the waste and, you know, cups and stuff like that.

TJ Miller (18:01)
Yeah, I'm

actually shocked that there's like any trouble using your own cup.

Chris Gmyr (18:07)
Yeah, mean some some places still don't allow you other places do so I don't know unfortunately we just keep them home because it's more of a hassle especially with trying to wrinkle the kids and then having two extra cups if they don't fill them and you know all that so

TJ Miller (18:22)
Yeah,

what size did you go with? I'm seeing they've got like extra small through large.

Chris Gmyr (18:29)
I believe it's the 10 or 12 ounce. I don't have it with me right now, but I can. It's one of the smallish ones.

TJ Miller (18:35)
Yeah, the 12 ounces

they're medium, the large is 16 ounces.

Chris Gmyr (18:40)
Yeah, medium then. It's the 12 ounce.

TJ Miller (18:42)
Cool. I don't know, man, these are pretty neat.

Chris Gmyr (18:44)
And check them out.

TJ Miller (18:46)
and they got all sorts of cool colors, man. This is going to be a trouble.

Yeah, this is cool. I'm not sure the difference between all of them, like the difference between the long play and the brew glass. I don't know.

Chris Gmyr (18:55)
Yeah, there's so many of them now. I... Yeah.

TJ Miller (18:58)
Very cool, dude. Yeah, let's check that

out for sure.

so I was also looking for it. I cannot find the cursor rules anymore that I was talking about for,

for the Claude stuff, so I'll have to track some of that down. But I will, before this goes live, I'll at least toss what I do have in a gist, and we can link to that.

Chris Gmyr (19:21)
Okay, cool, sounds good.

TJ Miller (19:23)
As I was looking around at these coffee cups, I was like, we should probably find that for show links later. But what's cool is I did just find, you know the awesome repositories, like all the awesome lists? There's apparently a cursor rules, like an awesome cursor rules list. I know you're working with, like I can provide one for like Laravel and stuff that I acquired, but it looks like there's like

Chris Gmyr (19:26)
Yeah.

Mm-hmm. Yep.

Cool.

TJ Miller (19:48)
different ones for like Node and Ruby and there is a Laravel one in here too. There's like a Laravel tall stack one also. So that might be something cool to check out as well. So I know you're going to like with your new gig, you're working with an array of things.

Chris Gmyr (20:04)
Yep, a little bit of Ruby on Rails, a little bit of Node TypeScript, a couple of frameworks on top of that. So there's a Node framework called Fastify. So a lot of our apps use that. So yeah, any of that would be fantastic to find.

TJ Miller (20:23)
Cool man, yeah, definitely something to check out. Sick.

So, you want to talk a little bit about Science Olympiad? I would love to talk about this a little bit. it went very poorly. I was, you know, we went into it and we talked a little bit about it, that I was a little concerned going into it. I was not feeling the most confident. and that mostly was coming just from looking at the syllabus we were provided.

Chris Gmyr (20:29)
Yeah, let's hear it. How did everything go this weekend? no.

TJ Miller (20:50)
And then looking at like what we had covered, I was feeling like we maybe missed some stuff and we've between snow days and scheduling conflicts and whatnot, we've missed a handful of practice sessions. So I was feeling like maybe they were going to be underprepared. And, it ended up being, you know, I actually don't even know. We couldn't get past, they had some team dynamic issues. So.

Chris Gmyr (21:06)
Mm-hmm.

TJ Miller (21:15)
Just looking at the whole competition, we show up, they have like opening ceremonies, and then everybody like splits off to their competitions. We had our competition at like 10, 15 in the morning, which is great. Cause then like award ceremony wasn't until one 30. So we kind of had like the rest of day to like chill. And then, in between competitions, they also have some like non-competitive free form stuff that you could also participate in as kids. like, was, it was pretty cool.

But we spent a little time after the opening ceremonies, like they were allowed to bring in an eight and a half by 11 cheat sheet. So the morning of the competition, I woke up like super early, sat down with Claude and we knocked out a cheat sheet, like geared, like I kind of narrowed it down to like front and back, what like geared towards fourth of sixth graders.

for Python, like I dumped in like the context of like what the competition is, the syllabus and everything. And then it gave me like a, had to finagle it a little bit, but we ended up with like the cheat sheet. So I went over the cheat sheet with them, helped them put together a game plan for going into the competition. Like competition's two parts. There's a written test where it's like a multiple choice fill in the bubble, like quiz. And then there's also like an interactive practical.

I didn't really know what the interactive practical was going to be. And my assumption going into it was also that it would be like structured. Like, let's say they had half an hour for competition. They would be 15 minutes on the like multiple choice tests and then like 15 minutes on the interactive like practical thing. Turns out it's

Let's just say it was half an hour. It's half an hour. Here's the written, here's the interactive thing. And so like they were totally unprepared for like how to split their time between the two. and so I guess they decided on like dividing and conquering between the two tasks, which I think is a fine plan. I much rather would have had them all put their heads together on one thing at a time, just to like make sure they're getting the best possible answer.

Chris Gmyr (23:03)
yeah.

TJ Miller (23:22)
But they decided to divide and conquer. I guess, going into it, the plan was that one of the particular, teammates would be, he was like a, everyone thought he was a better typer. So we're like, all he's going to do the typing and the other two can kind of like do the thinking pair programming style and like work together. And in all of our practice sessions, like they've worked really well together as a team. So I figured we were like,

Chris Gmyr (23:43)
Hmm.

TJ Miller (23:46)
probably good on that front. Turns out the dividing and conquering was not a great idea. Also turns out the student that was doing the typing wasn't necessarily listening to the other two students. So like the other two team members would like maybe make a decision on what the answer was going to be or like what the solution was going to be or like what to type, you know, like a print statement, a conditional, here's what to do.

And so I guess there was like a bunch of infighting with them over like the one student that was typing wanted to type his answer or whatever he thought was correct and not necessarily the consensus of the team. so like, I think things just like really fell apart on team dynamics. It sounds like they kind of struggled with some of the content, but it's the way they made it seem was that the team dynamics, really they came out and immediately, my son told me he's like, we argued the entire time. I was like,

Chris Gmyr (24:25)
Yeah.

TJ Miller (24:38)
Like is not the first thing that I wanted to hear coming out of the competition.

Chris Gmyr (24:43)
Yeah,

TJ Miller (24:44)
So wasn't feeling great about it, but they were all off having a good time. Like we decided the competition was down the street from our house and we didn't have anything else going on. Ellis had done all of the like free activities. And so he was just right. Like we just came home for a couple hours until the award ceremony. So we go back for the awards. They award. There was like 30 something schools there, I think. I think there were like 32 schools.

So for the award ceremony, was like in a gymnasium, all of the schools were like grouped together, like out on the gymnasium floor and all the parents were like up in the bleachers. And they get to the Code Warriors and they award the top eight teams. So basically once I knew that they didn't get like eighth place, that they like basically didn't place.

Cause I just coming out with that report of like, we fought the whole time. I'm like, great. This they didn't, they did not do well. so they didn't end up placing my son took it super hard. he was just like, my wife and I were sitting in the bleachers and we could see he was just like bawling. So I like head down there and try to get him to like, talk to me at all. Like, let's try to like deescalate the situation. I was.

Chris Gmyr (25:35)
Yeah.

TJ Miller (25:55)
I was totally prepared to like, do we just need to like step out of the gym for a minute and kind of like gather ourselves and come back. but he, all he was just like, leave me alone. I don't want to talk like, just leave me alone. Like, all right, fine. I'll let you be. And like, I head back up to the bleachers and I guess like, as I was walking back up, one of his teammates, I guess was like trying to console him or was like tapping him on the shoulder or something. And Ellis ended up like.

throwing an elbow like back at him. I was like, no, that is not cool. So we ended up leaving. I take, I take Laurel and like my wife and my son home and, she's, she's dealing with him. I go back cause I still have like two other teammates that I want to like, at least recap with afterwards and just, know, at least

Give the like, I know we didn't place it's okay. You guys did great. We'll like pick things back up where we left off. You know, there's plenty of time to train before the next one will work out all the kinks. Um, so anyhow, I like, I head back up there to do that. And I ended up talking to like one of the team members, give that whole spiel to, uh, his mom was just like, is there anything like they could take home in practice? Which we'll talk about.

Chris Gmyr (26:47)
you

TJ Miller (27:11)
So I got to work on that. And then I went and talked to the other team member and like, I get it, but like totally got like mama bared by the other teammates mom, which did not feel good. Like I was already like, kind of beating myself up. I'm like, could I have done more? Like what?

what could I have done differently to have not resulted in this? I'm kind of beating myself up. In the back of my head, I'm also like, I know my son's doing the same thing. And I guess this team member was also really upset. yeah, the last thing I needed was to get it totally mama bared. So I navigated that conversation as best as I could.

But we've got practice tonight. I'm kind of stressed out about it because like, don't know.

It's a tough situation. And like, I feel like all we had time for leading up to this competition was like just trying to go over all of the content and like really didn't have a ton of time to like talk about team dynamics and like the game plan going into the competition. Plus I was like totally flying blind into like how the competition's conducted. so we know a lot more coming out of it, but

And honestly, I think if I would have foreseen team issues, I would maybe have tried to deal with it. But during practice, they have been so cohesive and so good about working together.

that I just, that was, I was so much more concerned about like the content.

Chris Gmyr (28:46)
Yeah.

Yeah, it seemed like the pressure probably got to everyone and then just started to crumble from the inside out.

TJ Miller (28:53)
Well,

in, so my son's in sixth grade, he's 11. I believe the other two team members on his team are in fourth grade, maybe fifth grade. I know for sure they're not in sixth grade. So they're younger than my son. And this is my son's not his first time learning how to code. Like he got through the majority of learning JavaScript over the summer. So these concepts aren't new concepts to him. So I

definitely put a little pressure on him to like, Hey, you're the oldest here. You, this is not your first time learning how to code. Like, I kind of need you to be the team captain and kind of be a leader and an example here. And so part of why he was so upset that they didn't place is that he felt a lot of responsibility. and he had said at some point that like, let everybody down. like, buddy, you didn't.

Like, this is a team activity and like...

feel kind of bad about putting pressure on him, but like, I also didn't put that much pressure on him. And it's a competition, like, and I wasn't wrong in saying the things that I said. He is the oldest. He is the most experienced in the group. Like kind of got to recognize the position you're in, you know, and it is a competitive environment. So, I don't it was.

Chris Gmyr (30:07)
Yeah. If he wasn't

used to being in that environment a lot, even though you might not have put as much pressure on him, I'm sure the pressure that he feels is probably like five times what you don't thought it was, because he's just trying to do a much better job at that role. And I feel the same thing. I do the same thing, for sure.

TJ Miller (30:29)
Yeah, I put a ton of pressure

on myself.

Chris Gmyr (30:32)
Yeah, and they perceive those things completely different because it's like a parent or someone that they respect or look up to asking them to do something special. So it's it's that much more pressure for them, which is tough.

TJ Miller (30:47)
Yeah, and he's also definitely not,

He's also definitely not used to that kind of pressure. Like he hasn't done a lot of competitive things because he does like, part of the reason why he doesn't do a lot of competitive things is because he struggles with losing. Like he takes it really, really hard. And so we haven't engaged him in a lot of competitive things because it's just, it's such a struggle for him.

And yeah, I get the duality of like, the more he does it, the more, the more you work that muscle, the better you get at it. But at the same time, it's pretty big reactions. and he's also just like not.

Chris Gmyr (31:19)
Yeah.

TJ Miller (31:22)
He isn't, he isn't really that all like all that competitive. so like he doesn't thrive in a competitive environment like some people do. So I don't know it, it honestly all felt kind of bad. but the next competition's in May. So we've got, you know, we got some time to like practice more. We can work on team dynamics. We can work on getting through the rest of the content and really solidifying things.

Chris Gmyr (31:38)
It's time.

You can start doing some dry runs too. So basically do the same thing that they did over the weekend. Here's the test, the competition, the two parts. You have 30 minutes go. And then you can, over time, try out those different dynamics, like everyone working together and one person typing or splitting up in teams. See what worked best, what didn't work as well, and just keep on trying to work with them on that.

It's not always going to be the same stress level, but something a little closer to it.

TJ Miller (32:17)
Yeah.

And here's a factor of it that like really sucks. And I feel really bad about like the one student that I got like mama bear'd from. I hope that's not super derogatory. It's just, think the best way, and I think she even put it that way too, like, I'm sorry, I'm being such a like mama bear about it. And I'm like, it's, I get it. Your kid's upset and I'm the coach. Okay. Like I understand. But,

he's the alternate on the team. So this was like a practice test. So like, this is like a practice competition where it's like you place, you get medals and everything, but like, really this is kind of a practice competition for the big regional competition coming up in May. So our team is three members, but only two members get to compete in the like big regional competition. So he's actually an alternate. So there is a chance that this was the only competition that he's going to be participating in. So.

I feel like extra bad that things didn't go over so well, but

I also didn't anticipate them having those like team issues and like I think those were so.

I think the team issues were so cumbersome that like, I don't even know how much they struggled with the actual content. like...

All I can think of dealing with, I mean, maybe I get more details tonight when I talk to the team about like what happened and like talk more about the competition, but.

Maybe it's we should go and focus on like the more complicated pieces. Like, I don't know. Speak.

Chris Gmyr (33:41)
Sounds like they're

gonna be introduced with team and project retros.

TJ Miller (33:45)
Yeah, right.

They're already pair programming, man. Like, let's, let's introduce retros and we're already like in like a couple of weeks sprints. no, this is hilarious. so speaking of study guides, I talked, I talked a little bit about, like, I wanted to provide them study guides to like take home. Cause what we've been doing for our team training is, I just signed up for code Academy.

And we've been going through the Codecademy Python course. I couldn't think of a better way of approaching it. Like, I don't think I would have done as good of a job putting together like cohesive learning for them. It's got like the explanations. It's got videos you can watch of like the explanations of what to do. You've got the interactive like terminal right next to it. So you're actually like, you're typing and you're writing code and you're running code. So I really couldn't think of a better way to like approach it.

so we've just been doing Codecademy, but I haven't given them the like credentials to take home and practice Codecademy stuff. Like, I don't know. like I'm not even planning on getting like reimbursed for it. I got just like paying for it because it's what I can donate to the team. Like, I don't know. and it was like my choice. Like I'm, I want to use this as the learning mechanism and it's been working out well so far, but I wanted to provide something for them to like take home and study with.

So I thought it would make for an interesting live stream. I went and did a live stream in, I'm like using Prism to like build out artists in command that generates these study guides based on the topics. Worked out really pretty well. They're the only thing that I don't, I think they're all really pretty good. The only thing I don't like about them is like, think

they could have a more standardized structure. maybe come up with a standardized set of headings for it to fill in. And then could just fill out each one kind of in the same way. But yeah, we can link that in the shout-outs. think it wasn't the most exciting stream. There was a lot of me not talking and just typing out code. But I think the end result and the topic was pretty neat.

Chris Gmyr (35:41)
Yeah, that sounds good. I think that would be interesting to watch too.

TJ Miller (35:44)
And then,

yeah, and then they'll have something to like go home and study with and like hopefully reinforce some of the like technical pieces a little bit better. wish I would have maybe done that from day one, but Hey, here we are. Yeah. All right, man. I think we're getting close to running out of time. You want to wrap up? Awesome. Thank you all so much for listening to the slightly caffeinated podcast.

Chris Gmyr (35:54)
Thanks.

That's right. do what you can.

Yeah, let's wrap up.

TJ Miller (36:11)
Show notes, including all the links of things we've talked about in social channels are down below. You can find them also at slightlycaffeinated.fm. Thanks for listening and we'll catch you next week.

Chris Gmyr (36:21)
See you.