How should we work now? (part one)
00:00:07.77
Chris Morrell
All right. Welcome back to Overengineered. It's only been a little over a year. ah The podcast where we ask the question, what's the absolute best way to do things we already have a perfectly acceptable solution for?
00:00:20.90
Chris Morrell
Um, I am back with the full Internachi crew. um It's been a long time and we've been. ah in a you know totally stable, normal period of web development. um nothing new happening and we're just going to talk about the nothing new that is happening right now.
00:00:39.84
Chris Morrell
before we get there. ah let's Let's just go around and let everyone say hi because we've got a couple new faces. so Maybe. Skylar, you want to start?
00:00:49.16
Skyler Katz
Yep. Named Skylar.
00:00:51.31
Skyler Katz
I don't know what i don't know what what you'd like me to say. I live in St. Louis, Missouri.
00:00:59.59
Bogdan Kharchenko
Nice. uh well i'm happy to be back here on a podcast again obviously my name is bogdan i'm sure many of you guys have Yeah, it heard me ramble on the podcast before, but I am now officially a forward deployed engineer.
00:01:13.60
Bogdan Kharchenko
That is my title.
00:01:15.22
Chris Morrell
ah ah
00:01:16.16
Bogdan Kharchenko
ah So anyway, glad to be here. but um Yeah, go for it. What do you guys think?
00:01:22.11
Chris Morrell
Ben you want to say hi?
00:01:23.19
Ben Carr
Yeah, sure. Hey, I'm Ben. I live in New York City. Long time listener. First time. co-host, I guess.
00:01:28.13
Chris Morrell
ah
00:01:28.78
Nathan Mayo
Yeah.
00:01:29.82
Chris Morrell
And Nathan.
00:01:30.89
Nathan Mayo
Yeah, I'm Nathan. I live in... Good old Phoenix, Arizona. It is actually cold enough right now to be. wearing a jacket. Well, this is more like a thin.
00:01:41.65
Nathan Mayo
Thin jacket compared to what people wear where it's actually cold, but yeah. All right. Happy to be here.
00:01:49.04
Chris Morrell
Yeah, so... ah We've added... um a couple of people to the team. and also all of software development has changed. And. um you know It's a thing that I've been thinking about a lot.
00:02:04.49
Chris Morrell
both sort of you know We're going through this interesting... interesting place where we're transitioning a team to twice its size it's still a small team but you know going from three folks to five folks is is changing the way we have to think about our processes. And then at the same time, you know AI has changed software development in a million different ways.
00:02:26.63
Chris Morrell
And there's all these questions about. you know, what. processes even make sense. You know, how do you approach software development anymore? um These are... conversations that we're constantly having.
00:02:38.55
Chris Morrell
And so I've wanted to sit down with everyone and just. talk a little bit more. Um, I don't know if abstractly is right, maybe from a higher level about some of this stuff. And I just thought.
00:02:49.92
Chris Morrell
Why not? Pull the ah the old podcasting mic out and and do it live. um And I think what. Finally. Pushed. pushed me to say let's actually like do this is um The folks who run or created AMP. uh have this podcast, Building an Agent.
00:03:09.24
Chris Morrell
They did a couple of episodes sort of broadly about. how they work. at amp. um that I think is both very different from... our process and sort of traditional processes but also I.
00:03:26.50
Chris Morrell
ah you know i i can't believe that i'm saying this because a year ago i certainly would have Like. laughed laughed in your face if you told me I'd be saying this today. Um, But I think a lot of the things that they're talking about now, I'm like.
00:03:40.22
Chris Morrell
Maybe we're not there yet, but I can see getting there. you know this down to the idea of like, do we even do pull requests anymore? Um, So.
00:03:51.02
Chris Morrell
I sent. Everyone off said. Take some. work time to actually listen to these two episodes and let's like reconvene. And. um Talk about them. So I think. I have a few things that I want to get to and I'm curious about, but I'm curious.
00:04:05.97
Chris Morrell
If there's anything for any of y 'all that just like. stood out. Um, you know, out of the gate as something particularly interesting or noteworthy that that came out of listening to this.
00:04:17.19
Bogdan Kharchenko
um The only thing, you know I was actually talking about this with Ben the other day. And I feel like, you know, I think number one, I have to establish the fact that they are advertising their product.
00:04:28.37
Bogdan Kharchenko
So I have to approach this with like a grain of salt. of everything that they're saying and uh but you know i do think that um
00:04:31.61
Chris Morrell
Yeah.
00:04:36.87
Bogdan Kharchenko
you know It's kind of refreshing to hear somebody talk about some of the process changes that are happening. um Simply because, you know, like like you said, Chris, like a year ago, it was like stigmatized that you were using AI to write code or whatever. And now we've kind of all accepted this. And, you know, we're we're now like trying to. figure out what does that future look like and you know I do think that they have like a pretty compelling product that we can obviously get into but
00:05:07.78
Bogdan Kharchenko
In general, I think it's worth having these conversations. you know To me, personally, like one of the things that really kind of... continues to be a reoccurring theme is like this idea of Uh...
00:05:22.54
Bogdan Kharchenko
having a little bit more agency around various
00:05:24.92
Chris Morrell
Yeah.
00:05:27.23
Bogdan Kharchenko
parts of not just the software development process but like the business process because you can now kind of operate at a higher level. um you know Just being able to do multiple things at once, whether it's using an agent for research or you know looking at some sort of like product.
00:05:45.91
Bogdan Kharchenko
uh inform or like data or like doing some sort of like data sciency type of thing um so i do think like that is something that is super exciting and you know i do like the idea of amp personally i have not used it much
00:06:01.91
Chris Morrell
sure
00:06:02.03
Bogdan Kharchenko
But it's an interesting. product and some of the interesting ideas. so
00:06:06.84
Chris Morrell
Well, and I mean, realistically. I think a lot of what. They. were talking about obviously it was through the lens of their product But. None of the things that...
00:06:16.93
Chris Morrell
All right. particularly interesting to me at least are things that i feel like are specific to amp as a tool you know it's more just like the process. And, and yeah, I mean, I think.
00:06:28.92
Chris Morrell
By the time anyone's listening to it, it'll be obvious from the title that. I think you're absolutely right that like sort of. the questions about trust and agency and sort of like what Um,
00:06:40.54
Chris Morrell
where you're approaching the code, you know, like the... the name of this the the tentative name of this is like moving from a thousand foot to a 10 ,000 foot view on code. like I think that that's a big piece of it for me is just like.
00:06:54.84
Chris Morrell
The scale of everything is totally different, right? The the amount. that a person can do and like What? parts are hard are now like changed or There are fewer hard parts, so now we have to focus on different hard parts more.
00:07:09.73
Chris Morrell
Um, And. Yeah. The ways that like. agency and trust and like. process come into play. ah just are totally different now.
00:07:21.05
Chris Morrell
And I think that's probably the most interesting part of it for me, for sure.
00:07:25.27
Chris Morrell
I think.
00:07:25.26
Ben Carr
i think you
00:07:26.17
Chris Morrell
Oh go ahead.
00:07:27.15
Ben Carr
I think the the question that I keep... wrestling with um With all of. like this development in the industry is kind of whether or not the code matters.
00:07:37.99
Ben Carr
Amen.
00:07:37.93
Chris Morrell
Yeah.
00:07:39.19
Ben Carr
And you know there's part of me that feels like... Uh... you know, i I internally know that it never really mattered. and you know tabs versus spaces or like where does the curly brace go Like. intrinsically i know that all of those things never really mattered and if the code works that's the thing that's important But...
00:07:58.12
Ben Carr
The... The thing that I keep wrestling with is like... What was... The code mattering. A proxy for? And I think there was these things that happened. partly from the code itself and partly from the processes around the code, like pull requests.
00:08:11.87
Ben Carr
Where. there's interesting things that it was useful for that are not really about the syntax, like the typing. part the the actual like indentation and variable naming and stuff.
00:08:25.38
Ben Carr
like mangled JavaScript that is all minified for browser deployment and stuff. has always been like the exact same code technically as... the the Yeah. unminified version that had all of the nice variable names and all the block comments and all of that kind of stuff.
00:08:37.70
Chris Morrell
right
00:08:39.83
Ben Carr
But what what code felt like a proxy for was sort of... care and intention. And what pull requests felt like a proxy for. are like communication and.
00:08:51.49
Ben Carr
ideation in a group setting.
00:08:54.22
Chris Morrell
Yeah.
00:08:54.34
Ben Carr
And. I think you get some of that with agents, but some of it... you can't really get. um you know you can't really ask an agent to give you another human's lived experiences And if you entirely remove...
00:09:07.68
Ben Carr
That. What does that say about what you're building and like how it differs from what you used to be able to build before? given that you had access to those other ideas and that that other communication and the ability to like think ahead. based on someone else's mistakes that they made over time.
00:09:24.18
Ben Carr
Um. That question is like the the biggest thing. i struggle with a little bit. uh It's just like, what what are we losing if we give up some of these processes? Because it's very enticing to do, and it does feel like that's a direction that makes sense to move in.
00:09:39.00
Ben Carr
Bye. How do we get some of that stuff? that wasn't really about the code. in this new world.
00:09:46.08
Chris Morrell
Yeah. I think that's a great way to. Say it. I really do because I think. and And I think that... Part of why I want to have this conversation is. I don't think that we should accept giving those things up. Right.
00:09:59.20
Chris Morrell
It's just figuring out. How do we? How do we handle like making sure that we still work with the same level of care but With.
00:10:11.04
Chris Morrell
the new tools that are available to us you know And like. Yeah, I mean, I think that that I think that's partially why the amp. Like. podcast was so interesting to me is because they So clearly.
00:10:23.86
Chris Morrell
Think about communication and like shared understanding and like um ideation and all those things, right? But they're just doing it in a different way now. right And in a way, honestly, that is really appealing. it's like Was it?
00:10:38.74
Chris Morrell
Really? good for me to look at the code instead of the product in a lot of those cases. Like a lot of times it was just easier for me to look at just.
00:10:49.88
Chris Morrell
The PR. And. my developer brain could turn that code into what I knew the product would do from it. But like. with orbs or with VMs, ah you know, or or preview like Laravel Cloud Previews or whatever. like You can just... send me a fully working feature.
00:11:09.82
Chris Morrell
With like three different logins. And I can poke around. End. Get. Like. probably a richer. understanding of what you're trying to accomplish.
00:11:21.35
Chris Morrell
through that than I was by looking at the pull request, right? And obviously that doesn't work for everything. you know for things like security and performance. and you know there there are lots of other pieces that we still have to approach in a different way.
00:11:36.15
Chris Morrell
But. for a lot of a lot of code, it's like... Well, does this actually... Do what? like do Are we aligned in what this is supposed to do and does it do what?
00:11:48.65
Chris Morrell
I think we both agree it should do. and Could it be better or does it matter if it's better? You know, like all those questions. We were kind of doing through Code review.
00:12:00.34
Chris Morrell
And now it's just so cheap and easy to do it at a different. level. That like. that's really exciting to me you know? Like. The whole irrefutable proof.
00:12:13.14
Chris Morrell
idea of just like Bring me... Videos and screenshots and... charts and you know, sequel queries or whatever the heck it has to be to show me that this does what I think it does.
00:12:26.100
Chris Morrell
um And I can share those around first. so that when it... does come down to looking at the code because we still do look at the code like You're looking at. the parts that actually need a developer to look at them. you know?
00:12:40.07
Bogdan Kharchenko
I will say the other thing that I guess the other point, Ben, that maybe would i don't know maybe connect the dots is like you know i feel like a lot of uh times we've been as developers in a position of like here's a ticket and let's let's go go kind of like write this ticket out in code right and i feel like right now we have an ability to kind of say well instead of instead of spending a lot of time writing the code
00:13:08.09
Bogdan Kharchenko
you know maybe we meet together with another team member in terms of like the social kind of element and kind of discuss of like what does this feature kind of look like and maybe just talk about like broader architecture decisions and um you know I think that That's still not going to go away. you know, and, you know. participating in that type of um you know, activity is still kind of like having this like...
00:13:37.37
Bogdan Kharchenko
you're not reviewing code maybe but you're kind of like thinking out of like how this thing is going to play out and you can kind of even so solve some of these things earlier and obviously the other benefit is you know because of the rapid development cycle like the things that you may have agreed on it may not be true when the feature is built.
00:13:56.71
Bogdan Kharchenko
um you know uh you could just rebuild it to something else so i think like it's just shifting a little bit because i just know personally like from you know, pre-AI era, you know, somebody would like write a thousand line.
00:14:10.29
Bogdan Kharchenko
a piece of code and they've spent like days you know curating it and making it look perfect. And it turns out that there's some sort of like fundamental architecture issue that, you know, even though the variable names are beautiful and everything's like abstracted.
00:14:25.61
Bogdan Kharchenko
in the wrong way you know and you have to like now it's very painful you know uh especially if you're like on the receiving end of that review it's like oh shoot, like I just spent quite a bit of time. you know putting my heart and soul into this and now i have to redo it so i do think that there's like some Uh,
00:14:44.90
Bogdan Kharchenko
you know, Just kind of like. a shift in where those things happen and I think if you know, you can solve them early on. There's not a lot of, like, emotional investment into, like, the actual code itself. I personally see it as, like, a benefit. Because now I don't have to like spend a lot of time, like...
00:15:01.95
Bogdan Kharchenko
Curating. the pr's only for them to be rejected later. um
00:15:08.36
Nathan Mayo
And I think...
00:15:08.42
Bogdan Kharchenko
So.
00:15:09.46
Nathan Mayo
I think some of that is... like for me it's it's almost more fun now because there's less emotional attachment to the code that I've written.
00:15:18.30
Bogdan Kharchenko
That's also true.
00:15:18.74
Nathan Mayo
And so I can. work with other team members who I normally like if we're hashing through a development tickets Um, now sometimes my first.
00:15:31.03
Nathan Mayo
point of contact is the person who's actually using uh So if I'm working on one of our... products i'll loop in michelle and i'll sit down with her and be like well so this is what the ticket says Let me just make sure that we're on the same page, that this is actually like.
00:15:47.99
Nathan Mayo
what you want. Is this fixing the problem that you have? And. We can iterate through like several different ways of solving that problem in a very fast period of time. I can send her.
00:16:01.33
Nathan Mayo
you know small demos And that takes... Very little time now, but... it i think is increasing the collaboration with uh with the rest of the team which for me is way more fun than sitting down and writing the thousand line PR for hours and hours and hours.
00:16:20.88
Nathan Mayo
um So I think that's been... That's been a really cool thing with the advancement of of ai and now being able to look at it from this higher level overview of how do you want this to work how does this make your life easier um it's just It's just very different.
00:16:38.29
Chris Morrell
Yeah.
00:16:39.29
Skyler Katz
yeah I'm not sure I'm not sure if the current. flow of like PRs and lots of meetings is the the right thing, but...
00:16:50.52
Skyler Katz
I... I... I listened that there's an episode that came out since you told us to listen. Two episodes and so. I don't know that others listen to it, but they talked about how like.
00:17:00.86
Skyler Katz
during a day. they They were redesigning their dashboard and they released a new dashboard, like 80 releases in one day. And every time they were using the product, the dashboard just kept changing as they were like. trying to make adjustments and I think.
00:17:16.28
Skyler Katz
Like on some level, that's cool. But as a user. like our users. who are not on the frontier of software development and coding, like would not want.
00:17:30.30
Skyler Katz
their user experience to change 80 times in a day so like i don't know how we like can
00:17:36.96
Skyler Katz
But also it doesn't feel good to have 23 pending PRs. constantly now either so like i don't know what the right level of like how do we How do we change to like, yeah, we can move faster than ever.
00:17:51.36
Skyler Katz
How do we also care for the fact that our customers? are generally non-technical Generally. Don't like change. end. Like.
00:18:02.61
Skyler Katz
You know. I don't know. How do we? how do we care for the customer and still, and I'm not saying AMP doesn't care for their customers, but their customers are just different in terms of like.
00:18:05.32
Chris Morrell
Yeah.
00:18:12.50
Skyler Katz
being willing to be like, yeah, they're. we're in this dev product and it's just going to change all the time. um
00:18:19.39
Ben Carr
I don't even think it's necessarily like a technical thing.
00:18:19.41
Skyler Katz
So I don't know.
00:18:21.63
Ben Carr
Like... We've been... I'm sure all in the experience. side of uh you know an a b test that's pretty clearly an a b test and like you open up an app or a website or something and you're like oh everything is different today And it just kind of...
00:18:36.05
Ben Carr
kind of sucks especially if you really liked the thing that you tried and then they decided to take it away or If you hate the thing. the new thing and you want to go back to the old thing, but that you have no real agency or control over it.
00:18:48.60
Ben Carr
um Yeah, it's a good tricky. thing to think about the human side of the the consequences of Shipping. huge changes all the time.
00:18:57.54
Bogdan Kharchenko
I will say.
00:18:57.59
Nathan Mayo
Because as a develop
00:18:57.54
Chris Morrell
Well and I don't think it has to be that. All the time. like I don't think that that's a... inevitable consequence. right like
00:19:06.87
Ben Carr
Yeah.
00:19:07.18
Chris Morrell
Hey. I think that in a lot of ways. The capacity that we have. You know, it's just how do you use it? Right. and maybe I... i agree I agree with both of you that I don't. think that it's I think that.
00:19:23.78
Chris Morrell
probably for amp it's fine right because of their audience Even though like, you know, I just saw they're talking about removing the entire setting screen and making it so that you.
00:19:33.95
Chris Morrell
can only adjust settings through puck, which I'm like. That's a terrible. choice Why would you do that? Like, please don't do that. Um, so i'm you know as i use the tool more maybe i'll i'll be more frustrated by the the constant tinkering with things, but.
00:19:51.58
Chris Morrell
ah You know, I think that. they can choose to work that way because of their audience. Um, our audience is different. But... There are lots of things that we can do.
00:20:03.35
Chris Morrell
That. don't have any customer-facing impact. But make. you know Make life better for them. Anyway. or, you know, introduce new features or improve stability or...
00:20:15.94
Chris Morrell
you know make things integrate better. in the backend in ways that. you know. doesn't. change their the way they interact with the tools that they're using right now but you know expands that or there's a lot you know there's lots of opportunities there where Yeah, it's just another thing to be thoughtful about. Right. And that kind of comes back to the whole.
00:20:34.24
Chris Morrell
the You know, I was thinking about... Um, Like. John Drexler talking about. Everyone has to be a PM. this is like that was his thing before you know, before AI really. And like.
00:20:48.55
Chris Morrell
That feels so true now is like. Yeah, you've got to be. a product person now and thinking about The impact to the customer is a huge part of that. You know?
00:20:59.65
Bogdan Kharchenko
One of the things that also stood out in the podcast as like a up. a word. is they've said feature flag like a dozen times. And maybe that's also like ah one of those things that could potentially help mitigate that.
00:21:09.86
Chris Morrell
Yeah.
00:21:14.51
Bogdan Kharchenko
But yeah, I obviously agree with you, Chris. It's like. you know, um
00:21:18.98
Bogdan Kharchenko
Even what Nathan was saying earlier, you know, rather than just like us, you know, putting on our glasses and writing code and eating pizza and Coca-Cola, like, you know, we can now be product people, not just like, you know, computer translation layers between the product people, right? And we can go talk to the person who actually needs the feature and maybe get some like real feedback, not something that was digested through some PM who didn't take good notes on that call. and
00:21:46.92
Bogdan Kharchenko
uh you know gave us the bad thing and then we also misunderstood the assignment and next thing you know it's like you know a few layers of errors so i do think like yeah that's kind of like the evolution and you know there's this word that's kind of floating around like this product engineer uh type of role you know where you are kind of still an engineering type of role, but, you know, with product focus.
00:22:11.02
Bogdan Kharchenko
And I think for the most part, especially in our team, we've kind of been operating this way just because we don't have those layers of the of the cake.
00:22:15.65
Chris Morrell
Yeah.
00:22:19.72
Bogdan Kharchenko
But, you know, it's... It's something that is just, you know, with the power of AI, especially not even with AMP, like we can now basically allocate more time about actually doing some sort of like discovery and iterating quickly rather than. you know making the variable names look pretty. um So. Yeah.
00:22:41.72
Chris Morrell
And we can make the variable names look pretty too, because...
00:22:43.84
Bogdan Kharchenko
Of course, 100%.
00:22:44.72
Chris Morrell
We have tools that do it.
00:22:46.13
Bogdan Kharchenko
Exactly. AI can do the variable. That's the other thing that I do, I will say, I appreciate. like ah you know to actually have really good code you have to spend
00:22:55.59
Bogdan Kharchenko
quite a bit of time, you know, ironing it out. And luckily with AI, I was just like, oh, yeah, like, I want to just do the rename of or pull this ah method into like its own class and do this abstraction to make it look pretty. And it's like effortless, you know, it's just able to do the thing. um So I'm super obviously excited about this because I feel like it's It's stuff that I always wanted to do, but it's just kind of like...
00:23:21.16
Bogdan Kharchenko
so expensive to do it manually that now it just became free. you know and I can allocate my time. doing other things.
00:23:32.18
Bogdan Kharchenko
like you know being kind of on the product side. you could say.
00:23:37.16
Chris Morrell
I do think that that's like a really.
00:23:37.24
Nathan Mayo
refactoring is.
00:23:39.04
Chris Morrell
Oh, go ahead Nathan.
00:23:39.34
Nathan Mayo
I was just going to say, refactoring is yeah effortless.
00:23:42.48
Chris Morrell
Yeah.
00:23:42.26
Nathan Mayo
No.
00:23:42.63
Bogdan Kharchenko
Right.
00:23:43.19
Nathan Mayo
and And I think the... being able to spend more time. gathering the lived experience of our teammates who are So like a lot of the work that we do is staff facing to make the interactions with members.
00:23:58.54
Nathan Mayo
better. and to make. The... the staff experience more seamless. and so being able to spend more time. talking with them and showing them new features and designing new features with them has just been Like I love that so much more. and um so i think like Bogdan was saying being able to allocate more time towards that product side.
00:24:19.75
Nathan Mayo
um it's just Yeah. Awesome.
00:24:22.42
Bogdan Kharchenko
Yeah, the only thing I was going to ah say is, you know, the the thing that I always come back to is like, for me personally, and this is again just personally, programming was always like a means to an end, right?
00:24:33.82
Bogdan Kharchenko
The thing that I actually wanted was a website that had a login screen that people can go put in their username and password. and you know do the thing somewhere inside some sort of dashboard. It just so happened that you know I had to write the code for that to be the case. But really, that's what I want, right? Or maybe that's what the business wants or whatever the case is. So I think ah that was always kind of like the thing that was driving.
00:24:58.86
Bogdan Kharchenko
Um, you know the code uh part
00:25:02.36
Chris Morrell
Yeah, i I think that that's like an important distinction because ah I do think that there are two. you know, there's kind of. two types of developers that i interact with at least right you've got The folks.
00:25:13.09
Chris Morrell
for whom that's really true right that the code was always a means to an end and like being precious about the code just never felt. important because it's like well if it does what it does And like.
00:25:24.01
Chris Morrell
my customers are happy then I don't care. Beyond that. And then there's the people. for whom like Writing code. was an interesting part of it and I think that that's I think that I fall kind of somewhere in between personally because I do think I've always thought.
00:25:41.26
Chris Morrell
you know. what's the end goal that we're reaching but i also really enjoyed the you know the act of sort of like creative problem solving and thinking about the systems and Um,
00:25:54.16
Chris Morrell
And yeah, I think that I enacted that through. Being thoughtful about code. Um, And I feel like I see. the It feels to me like the folks who are having the most trouble. with this transition are the ones who haven't found how to like shift that mindset.
00:26:16.18
Chris Morrell
into this like so new space. Because I do think. You can still Be. Thinking about. Sssst!
00:26:26.23
Chris Morrell
systems and processes and like. you know maybe you're not you're you know you're again it's like just at a much higher level You're not worrying about like.
00:26:39.33
Chris Morrell
you know, getting the perfect architecture for a single feature right you're thinking about getting like the right architecture for. Uh,
00:26:50.52
Chris Morrell
just everything right you know what i mean um and all of the you know all of the stuff that i've been doing around you know, automating Bug fixes and like.
00:27:03.19
Chris Morrell
you know. uh Yeah. thinking about like responding to exceptions with like automated reviews and just like I think all of that stuff ends up sort of being the same.
00:27:17.29
Chris Morrell
the same thing. It's like, how do I build a system That. is like eloquent. and elegant i guess is even more the word i'm looking for and like Um...
00:27:29.24
Chris Morrell
works well for us. And like, it's just. It's just not as much that. that's just not as much about the code anymore, you know? It is what it is. Um,
00:27:39.73
Chris Morrell
It's just a bigger thing now.
00:27:41.89
Chris Morrell
But.
00:27:42.27
Bogdan Kharchenko
Yep.
00:27:41.99
Chris Morrell
The other thing that I think about is like. you know we keep on kind of talking about the same thing which is sort of like the economy of code right the like um It used to cost. a lot more to write code.
00:27:55.33
Chris Morrell
And that had a bunch of different implications, right? It meant that. it felt bad to throw away code because like it costs you a lot to get it. Right. And it meant that.
00:28:06.23
Chris Morrell
You didn't. write certain code because you knew it would cost you a lot. to have it be written. Um... And like the number of times now that we can just be like, well.
00:28:18.79
Chris Morrell
Yeah, I'm just going to send an agent off for 40 minutes. to build out this entire feature, knowing that like, I mean, how many PRS do I say? I'm just going to close this because like. It's not quite right. you know.
00:28:30.46
Chris Morrell
And no one has any qualms with that anymore. Right. Whereas before, yeah, you. spent. days and days and days working on a feature and it's like well like We obviously we're going to close it if it's not.
00:28:42.93
Chris Morrell
good but like it hurts and now it's like yeah
00:28:45.60
Bogdan Kharchenko
Yeah.
00:28:45.59
Skyler Katz
I mean.
00:28:46.96
Chris Morrell
Okay.
00:28:47.29
Skyler Katz
My like ah very first feature.
00:28:47.60
Chris Morrell
we'll do it again
00:28:50.14
Skyler Katz
at Internachi. Well, there's my second feature. It was like my first week. And I. You're like, all right, build out this like. benefits membership benefits system and i built all this code out did a whole entire module.
00:29:05.54
Skyler Katz
And we deployed it and I've been here for. Four years and. We never.
00:29:11.65
Chris Morrell
We don't use it.
00:29:11.40
Skyler Katz
We never used the feature.
00:29:12.50
Bogdan Kharchenko
Yeah.
00:29:13.02
Skyler Katz
And every time I was like, maybe we just delete this. And you were like, no, like you wrote that code. I feel bad.
00:29:18.36
Chris Morrell
ah
00:29:18.48
Skyler Katz
ah And I'm like, I mean, it was great code, but. No one uses it. We should. delete it or retry it or do something. It just cracks me up like that.
00:29:29.42
Skyler Katz
uh whole Sort of. Yeah, code was expensive. And so.
00:29:34.22
Bogdan Kharchenko
Yeah.
00:29:34.96
Skyler Katz
We have to keep it. or not. Or... there's a problem. Ah.
00:29:40.51
Nathan Mayo
Well, it bugs.
00:29:40.44
Chris Morrell
Right. Well, and that's a good...
00:29:41.46
Nathan Mayo
and
00:29:42.70
Chris Morrell
a case of like Even there, um it It just turned out that. Once the feature was done. the person that was intended to use it.
00:29:54.03
Chris Morrell
didn't want to use it in the way that I expected them to. And so it was like. you know again if you had If we had been able to just like. throw a quick demo together and send it off to Nick and be like,
00:30:07.48
Chris Morrell
Hey, try this out. We could have learned. way, way earlier. That ah oh, he's actually it turns out he's never gonna use it despite the fact that I'm telling him to.
00:30:18.94
Chris Morrell
Um,
00:30:19.36
Skyler Katz
Exactly.
00:30:20.16
Chris Morrell
But. We just, it was. So unreasonable to do that before. Right. Because it's like. before you eat. you still would have had to build the whole feature. You know what I mean?
00:30:30.27
Bogdan Kharchenko
I don't know.
00:30:31.66
Skyler Katz
Exactly.
00:30:32.98
Bogdan Kharchenko
Yeah, I will say I had ah recently had an experience with this cheap code economy thing that you're talking about, Chris. I have an app built in Svelte. And I was like, oh, maybe I should just like. have an agent like rebuild it and react. as I was just like, I wanted to do it. And it was all like very well tested because it was just, you know, AI is writing the test so they were free. Anyway, the agent went off and like rebuilt. everything in react like everything just works one shot and i'm just like
00:30:58.09
Bogdan Kharchenko
yeah i don't want to do it i'm just gonna stick to smelt you know and it was just kind of like oh man that's insane you know that you can take an existing thing and change some sort of core architecture you know especially like the front end like relatively easy you know and it's kind of like
00:31:18.20
Bogdan Kharchenko
it's kind of wild you know before and the the other crazy part was like claude was like oh yeah this is gonna take two to four months to do it i'm like no you're you gonna do it right now you know and uh it's just
00:31:26.97
Skyler Katz
if
00:31:30.60
Bogdan Kharchenko
you know the timeline is kind of like shrinking on a bunch of these features uh and just in general like ah you know maybe we're going to transition to this other thing that on the podcast they talked about is like the idea of even like story points or the size of the tickets you know like what is that anymore you know I hear that all the time where it's like oh yeah it's an eight point ticket like what is that you know for for an agent it's 20 minutes of work or something you know but yeah traditionally it took
00:31:59.15
Bogdan Kharchenko
Quite a lot of effort to you know break down this ticket and create epics and you know, planned three sprints ahead uh and then you know like Skylar was just saying the person who we built the feature for it doesn't even want to use it. you know it's It's kind of crazy to think about ah that being the workflow.
00:32:19.27
Nathan Mayo
yeah now we show up on monday and chris is like here's a 7000 line built this whole product. Like for the higher sight.
00:32:29.84
Nathan Mayo
It's like I did this.
00:32:30.23
Bogdan Kharchenko
7 ,000. It was 22 ,000 lines.
00:32:32.06
Nathan Mayo
Oh yeah, yeah, it was a lot more than that.
00:32:32.25
Bogdan Kharchenko
of
00:32:33.37
Chris Morrell
ah
00:32:35.18
Bogdan Kharchenko
And that was the second push that Chris did. so it Um...
00:32:37.56
Nathan Mayo
Yeah.
00:32:40.99
Ben Carr
There's something in this. like I... code being difficult to produce and time consuming. Thanks. It's been... It's been pretty easy. to hire people that don't live in the United States.
00:32:54.44
Ben Carr
And. their economy as much. smaller to produce code for you for a very long time. Bye. the amount of code that you could have written for you for very cheap. Bye.
00:33:05.70
Ben Carr
it has been significant for her. quite a while. And at the same time... The idea of prototyping and testing smaller iterations of things has been like a product mentality for a while.
00:33:16.66
Ben Carr
So in this in this world, like... why did we build the whole membership benefits thing instead of doing like Up. a wireframe. to see if the product would be useful and like why haven't we
00:33:27.28
Skyler Katz
There was a wireframe.
00:33:28.84
Ben Carr
Yeah. Like, why haven't we just been hiring uh cheaper developers for every amount of code. like there's There's some like tension there for me in this. oh, it used to be so expensive and now it's so cheap. Well, it's like, well, it could have been cheap before.
00:33:43.62
Ben Carr
I. And maybe the quality would scale with that. But the quality also scales with having agents do it. and
00:33:49.32
Bogdan Kharchenko
but go
00:33:50.18
Ben Carr
Yeah. that the like third component of this is kind of the an agent will take X amount of time to do it. And I think. While it will give you something the first time, every time. it like the thing that it gives you the first time is.
00:34:01.95
Ben Carr
pretty I would say 100 % for me, not the thing that I want. that I actually want to ship. So there there is like a... Yeah, there's there's some amount of work still. involved in getting the thing to work the way that you want it to.
00:34:15.93
Ben Carr
And we do still have this history of. Like... it has been pretty easy to throw together a wireframe to test out whether or not it features is has legs to even go to the next step. for a while, and it's been cheap to. pay someone to develop code.
00:34:29.11
Ben Carr
For a while. so like that Yeah, there's just some some tension there.
00:34:31.46
Chris Morrell
oh
00:34:32.99
Ben Carr
that I struggle with a little bit. in the it's almost like there's ah a simplification. Of like... oh everything used to be one way and now everything is a different way and there's like a gradient in there that i feel like isn't usually brought up in these types of conversations.
00:34:48.47
Chris Morrell
Yeah, that's interesting. I mean... i i think I think you're right to a degree, but I also think that... I'm... the One big part about. Part of it is when I say cheap.
00:34:59.90
Chris Morrell
I don't necessarily mean like the cost of labor. I do think that that's. one part of it. But I actually think... that the uh there's like the human emotional cost that's as big or bigger for me of like No matter who you are.
00:35:16.51
Chris Morrell
Like what? No matter how well or... poorly you're paid. Like being told that all your work is going to be thrown away, like sucks, right?
00:35:26.74
Bogdan Kharchenko
s
00:35:26.88
Chris Morrell
And. I'm not. I'd like to believe that I'm not like. A bad. person and i don't want to do that to anyone like regardless of what i'm paying them um But I don't mind. like but you know An LLM doesn't. care.
00:35:42.87
Chris Morrell
Um, And I think.
00:35:43.93
Ben Carr
I feel like... There there's something in that exact sentiment though that still kind of gets me which is like If the code... is being thrown away. but an agent wrote it.
00:35:55.66
Ben Carr
And I should feel better about that. Now. it kind of implies that I didn't make that thing. So it's okay because they made that thing. And it's like removing... my own Like I'm still the person that is making the feature right and just because the implementation was created by another thing I should still feel bad if we're throwing away that feature.
00:36:15.04
Ben Carr
Because I put work in to get the agent to do the thing, you know?
00:36:18.20
Chris Morrell
well yeah i think i would feel equally bad If you. Ben. Uh, Like. came in with like a fully functioning feature that you felt. was like useful and good and like you had done all the process around like getting it to a point where you were ready for someone to like review it and merge it into the production code base.
00:36:39.92
Chris Morrell
I would still feel bad now saying like. Actually, no, like this is. this isn't good let's get let's just like get rid of it I think that it's more like. We just.
00:36:51.08
Chris Morrell
Don't. Like now we cannot get to that point. Because you, Ben. can iterate. a hundred times on a hundred different things.
00:37:01.88
Ben Carr
Yeah.
00:37:02.40
Chris Morrell
And find the one that feels good to you. like a person who is smart and thoughtful and i trust you know like you get to make that call
00:37:11.71
Ben Carr
Yeah.
00:37:12.36
Chris Morrell
And... I think that's the difference, right? Where. Um, The human side is still there for the people in this room. Right. It's that each of us.
00:37:24.73
Chris Morrell
can do like You can wake up you know or you You can get off this call right now. for a Friday afternoon. And be like. You know.
00:37:35.52
Chris Morrell
There's this thing that Lisa said that's been bugging her and I just like have an itch to fix it And you can... Send an agent off. to build 500 ,000 lines of code.
00:37:46.80
Chris Morrell
Look at it. and say, Meh. And just. Delete all of it. And like, you won't feel bad about it. I won't even know about it. you know and but like one day you might have that same thought and be like oh man i actually think this is great I'm gonna polish it up a little bit.
00:38:04.52
Chris Morrell
and send it Send a little video to Lisa. you know. 4 p .m lisa's like oh man this is amazing And 430, you can be like. I think I just built something great. like i'm gonna open a pr you know like that's more when i talk about the costs i think it's It's...
00:38:21.36
Chris Morrell
within that. that space, right? like There's... there's still the emotional toll of like investing in something and then like having it not work out.
00:38:33.09
Chris Morrell
um That. we are all going to feel sometimes and and like we're trying to avoid. But like. the Both the cost of like trying something and deciding. that it didn't work out. And the cost of like going for something, knowing that like.
00:38:49.48
Chris Morrell
Actually, it costs way less to totally shift gears now. Like both of those actually. come with that Both of those things, I think. solve the like human emotional side of it it's like we're we're removing some of the like Psychic weight. of the work. Knowing that like either a.
00:39:11.16
Chris Morrell
If it doesn't work out, I can just. Literally. delete the branch and It didn't really, I'm not. emotionally invested in the work yet. like Just didn't. Didn't feel right, and I just am fine with saying, actually, never mind.
00:39:25.82
Chris Morrell
And like. If I do invest in the work. And then it turns out we meet with a stakeholder who we didn't think to like. consult with first and they were like, oh, what about this other use case?
00:39:36.89
Chris Morrell
And you're like, ah. shit like that actually changes everything. It's kind of like, okay, well. we'll just change everything then it's not a big deal you know like
00:39:45.82
Ben Carr
Yeah.
00:39:45.82
Chris Morrell
I think both of those things. I don't know, does that... Does that feel like? It gets at what you're, what you're trying to get at.
00:39:52.57
Ben Carr
yeah I mean, I think maybe I was closer. to that separation. before this transition. like part of Part of the nature of... growing in your career and like working with lots of different people is sort of separating yourself from the output.
00:40:06.58
Ben Carr
And.
00:40:07.00
Chris Morrell
Hmm.
00:40:07.45
Ben Carr
It's not that it's easy by any means. but it is like a thing that i've put in a lot of effort Dude. try and do because you don't want that. no matter where the the thing happens, whether it's in code or review or whether it's like once the thing goes to production or whether it's like customer feedback that comes in.
00:40:22.63
Ben Carr
third hand.
00:40:22.98
Chris Morrell
Right.
00:40:23.67
Ben Carr
or whatever like you never want that and the design side as well like you produce a design some people especially early career designers They tend to think of the design as like a proxy for their.
00:40:34.80
Ben Carr
thoughts and beliefs and opinions.
00:40:36.34
Chris Morrell
Yeah. their self-worth.
00:40:36.92
Ben Carr
And it it becomes like a personal attack when someone is like, I think that is the ugliest button I've ever seen. Uh, Emmett. It's difficult to separate yourself from it. and sort of fall in love with the problem not with the execution
00:40:49.74
Chris Morrell
Yeah.
00:40:49.54
Ben Carr
And I think this is just like another flavor of that. Um... So, yeah, that that does get ah at what I was getting at, though. And the the cost analogy, I think. There's like the... The monetary cost, but then the...
00:41:01.16
Ben Carr
the emotional cost. plus the opportunity cost of. doing that thing and not something else.
00:41:06.91
Chris Morrell
Yeah. Yeah. It's like the economics of those things is what. has changed.
00:41:11.15
Ben Carr
Yeah.
00:41:12.00
Chris Morrell
Yeah.
00:41:13.08
Chris Morrell
yeah i think okay so I want to step back for just one second. And like, say. Okay, this is how... This is how we have worked. and where we are right now. Right. Which is.
00:41:24.51
Chris Morrell
We use linear for ticketing. Right. We use. GitHub. PRs. Um, we typically ah you know, we have CI. So like. pushes to the production branch, go directly to production after everything through a merge queue now.
00:41:40.80
Chris Morrell
um And so our workflow has always like. Then we meet. weekly to triage tickets we meet occasionally to do planning. Our planning had been every cycle. Now we've moved it to every month.
00:41:53.27
Chris Morrell
Um, And... a lot of that work is you know getting into a decent amount of detail i think that like Some of that comes from the fact that.
00:42:04.40
Chris Morrell
Our code base. you know, much of our code base for much of. The time was. very much my code. And it took me a long time to like.
00:42:15.96
Chris Morrell
let go of that and so there was a lot of like let's make sure that I'm on board with with a bunch of the decisions before we move forward. i think that some i hope to believe that some of that has changed in the more recent years Um,
00:42:30.71
Chris Morrell
but yeah a lot of it was getting pretty deep into the like okay what's what are the decisions that we're making here that Um, are going to impact the code base in the long term, right? Are we introducing new patterns? Are we introducing...
00:42:44.32
Chris Morrell
Um, new approaches to data like I want you know for me because I've been working in this code base for 20 plus years almost 30 years like Um, I care how the code. is gonna be in. in a decade, you know what I mean?
00:42:59.49
Chris Morrell
Um, And so that's kind of where we are. Is there anything else in terms of our process that I'm missing?
00:43:06.09
Skyler Katz
I mean, we do a daily standup.
00:43:08.22
Chris Morrell
Oh yeah. The Daily Stand-Up. Yeah, and ah during our daily stand-up, we like... Sit down and everyone talks about the tickets they're working on. Um, Which is you know I think we've. We... we talk about.
00:43:19.10
Chris Morrell
that It's nice. Because... Um, you get to see a little glimpse into all the different pockets of the Code base. stuff that you're not necessarily working on.
00:43:31.23
Chris Morrell
i ah I like to have everyone like pretty well aware of. how the whole system works and i mean i've said this before but like another piece of it is This is a...
00:43:42.45
Chris Morrell
many hundreds of thousands of lines of code base that's existed for decades. It. covers like you know. in home inspection like reporting and uh like digital signatures and like you know, file uploads and like it covers The application is very big.
00:44:03.29
Bogdan Kharchenko
illing courses and so on and so forth.
00:44:06.23
Chris Morrell
Right. It's. It's not the type of thing where a person can just be like, oh yeah, I kind of like get. I get how this whole thing works by just glancing at it. You know, I see Nathan like.
00:44:17.91
Ben Carr
maybe Maybe we...
00:44:17.83
Chris Morrell
Making a face over there.
00:44:18.54
Nathan Mayo
but
00:44:19.97
Ben Carr
Maybe worth mentioning just because they brought it up on the podcast. Bye. my quick count. Our code base is one-fifth the size of AMP's code base.
00:44:27.59
Bogdan Kharchenko
what No way.
00:44:28.32
Ben Carr
by their own.
00:44:28.46
Nathan Mayo
Jeez.
00:44:28.47
Chris Morrell
That's crazy. yeah
00:44:29.43
Ben Carr
metrics.
00:44:29.62
Bogdan Kharchenko
That's wild.
00:44:30.48
Ben Carr
Which...
00:44:30.66
Bogdan Kharchenko
Cheers.
00:44:31.39
Ben Carr
Feels bad to me. I don't know why.
00:44:33.69
Nathan Mayo
Yeah.
00:44:34.40
Bogdan Kharchenko
and I think only Amp has only been around for like a year and a half from what I heard else on the podcast.
00:44:37.87
Ben Carr
Yeah. And meanwhile, this nearly 30 year old.
00:44:39.21
Bogdan Kharchenko
It was kind of wild. Right.
00:44:41.09
Ben Carr
Monolith is... Yeah.
00:44:42.75
Bogdan Kharchenko
Right.
00:44:43.23
Chris Morrell
And arguably, we do a lot more than Anthem.
00:44:43.28
Bogdan Kharchenko
That's wild.
00:44:46.08
Chris Morrell
was.
00:44:46.60
Bogdan Kharchenko
come
00:44:46.71
Chris Morrell
Right. Like we have. many, many full-blown product categories like built into our application.
00:44:54.76
Bogdan Kharchenko
h
00:44:54.93
Skyler Katz
I think lines of code is like not a good metric. But...
00:44:58.52
Chris Morrell
right
00:44:58.58
Skyler Katz
I do. It's not a good metric anymore. Um.
00:45:04.63
Skyler Katz
I think. Bye. How do you prevent shipping? Slop. and not slop that's just like front end slop that people can see versus like the code base.
00:45:17.40
Skyler Katz
And like. from the From the podcast, it seems like Amp's like, well, we'll just fix the... poor performance with more code if we need to and we don't care that Like.
00:45:29.12
Skyler Katz
It might have been good from the beginning. and Maybe that's good. I think part of that is like they're a year and a half old and they're trying to like. grow versus we have what like 95 of the product market at the moment and so like We can take that time.
00:45:44.13
Skyler Katz
Or not, I don't know. like
00:45:45.75
Chris Morrell
Right. They're very they're very different like moments in the product's time.
00:45:47.14
Skyler Katz
It's just very different.
00:45:49.79
Chris Morrell
like timelines for sure.
00:45:52.09
Skyler Katz
Yeah.
00:45:52.45
Chris Morrell
Yeah.
00:45:53.79
Nathan Mayo
Well, but I do think.
00:45:53.96
Chris Morrell
But it it is true.
00:45:54.45
Nathan Mayo
I. I do think there's a shift in... you know models a year ago versus models now. Like being shipping slop.
00:46:04.80
Nathan Mayo
uh was much more detrimental I think in the past. with this new project that we've been working on for the higher exam.
00:46:14.60
Nathan Mayo
mean I wouldn't say we're shipping. Slop. but there is it's certainly much faster and Um... The agent is able to debug.
00:46:26.07
Nathan Mayo
any issues in development as you're going so there's like a case for well Do we refine, like... Bogdan and I have sat down and reviewed some PRs and where we're like, well, do we Do we need to refine this? or is it good enough for what we're wanting to do right now and Most of the time we just shipped it And.
00:46:46.01
Nathan Mayo
Uh... it has it's worked Fantastic. So I don't know.
00:46:50.16
Chris Morrell
Well it comes back to Ben's point. earlier of like You know, the the caring about. tabs versus spaces or where your curly braces are positioned.
00:46:58.26
Nathan Mayo
Right.
00:47:01.72
Chris Morrell
you know, the. They'll like. All of those decisions essentially came down to... and like How easy is this to understand? How easy is it to work in? How is it how easy is it to delete? right? That's That's generally what I would say, like all those syntax decisions.
00:47:18.44
Chris Morrell
come down to... Those. three things mostly right or like modify and delete Right. Like. If we are consistent. Then everyone can be on the same page because you're not having to like Parse.
00:47:33.94
Chris Morrell
wildly different. approaches to the same. same types of problems, right? You see the same pattern. you know generally what it is right so that's like syntax and sort of stylistic decisions.
00:47:45.100
Chris Morrell
And then it's like. When we wrote a lot of the code. I mean, the the thing i'll I'll out myself here, like I still. care about tabs versus spaces because when i go into a code based at spaces And I use my preferred IDE.
00:48:01.64
Chris Morrell
And I like. Hit backspace. It. goes back one space instead of one tab indent. And I'm like, What the hell? like This is 2026.
00:48:14.65
Chris Morrell
I don't care about tabs versus spaces like from a. you know Um...
00:48:20.24
Bogdan Kharchenko
Ideology.
00:48:21.26
Chris Morrell
Yeah, ideological perspective, but I do want Backspace to go back one tab index when I'm like editing code.
00:48:21.47
Bogdan Kharchenko
and
00:48:24.52
Bogdan Kharchenko
Yeah.
00:48:27.30
Chris Morrell
but like
00:48:27.78
Bogdan Kharchenko
Yeah.
00:48:29.22
Chris Morrell
you know even there it's like when's the last time That. came up for me. i you know, it just doesn't.
00:48:34.24
Bogdan Kharchenko
When was the last time you used an IDE, Chris? It's 2026.
00:48:37.17
Chris Morrell
ah I mean... I still do.
00:48:39.68
Bogdan Kharchenko
ah
00:48:40.25
Chris Morrell
I still do um But I'm not writing a whole lot of the code.
00:48:42.35
Bogdan Kharchenko
Yeah, same.
00:48:43.90
Chris Morrell
It's mostly going in, understanding the code better. through. my decades of experience using an IDE. It's much more a code reading tool. or a code experimentation tool.
00:48:56.43
Chris Morrell
Um, than it is like a code authoring tool tool now for me. Um, Hmm. But yeah. And like, and the and then the other piece is like. Yeah, we choose architectural decisions because...
00:49:09.06
Chris Morrell
It makes it easier to edit or delete or change in the future. Um, Because, yeah, if I co-locate. a bunch of related things.
00:49:19.80
Chris Morrell
Then if I ever change my mind about those things, I just have one place to look, right? Like that's the whole. kind of argument for verbs. that we've talked about, right? And it's like. I don't know does any of that matter anymore? If that, if the.
00:49:33.28
Chris Morrell
Agent. Doesn't care. Do I need to care anymore? like as long as i can understand like if i look at a pr and It used to be if I look at a PR and I don't like the code.
00:49:48.44
Chris Morrell
I, my like. decision would be like well i'm gonna have to live with this so like it matters that i don't like it But now if I look at it. pr and i don't like the code but i understand like it's clear what it does It's just doing it in a way that like.
00:50:03.40
Chris Morrell
Feels... Kind of. weird for some reason to me i'm just like well That's fine. I still understand it. Um... I still... like can reason about it and can.
00:50:14.24
Chris Morrell
like instruct an agent on how to fix something if I run into an issue with it. like So it's fine, you know, and. So. to a degree. I think this whole. premise of like what even is slop is like I like the way that the amp folks talk about it. It's like...
00:50:31.49
Chris Morrell
Well, you know. if it's bad, like a good developer knows if it's bad. and We just don't. Like, I trust that you guys aren't doing that, you know, like. Um,
00:50:40.69
Skyler Katz
yeah
00:50:41.83
Chris Morrell
And again, it kind of comes back to trust, right?
00:50:44.92
Chris Morrell
Excuse me.
00:50:45.67
Bogdan Kharchenko
Yeah.
00:50:45.84
Skyler Katz
mean I will say like, I'm very.
00:50:46.17
Bogdan Kharchenko
good Good.
00:50:48.89
Skyler Katz
the They're conversations about like their. process. were very compelling.
00:50:55.60
Chris Morrell
Yeah.
00:50:55.67
Skyler Katz
Of like. They weep. They meet on Mondays. And they like all write down in a Google doc, like, here's the thing that we are going to work on. and there's like more to it i think than that but I'm not sure that...
00:51:10.53
Skyler Katz
Like, I love meeting every day. to like get to see people because I work from home. And... Like.
00:51:16.84
Chris Morrell
Yeah.
00:51:17.49
Skyler Katz
Bye, wife. does not work from home. So I'm like by myself all day. So it's great to like. see people. um But like going through linear every day and being like yep still working on this thing or I'm working on that thing like it doesn't feel.
00:51:31.100
Skyler Katz
particularly useful. And now that the team is so many people, it's like, well, there's. an hour and we we've gone through like what tickets we're working on so i don't know like um I don't know that I...
00:51:46.33
Skyler Katz
Yeah, there were compelling things about their process. um
00:51:50.58
Chris Morrell
Yeah.
00:51:51.66
Bogdan Kharchenko
I agree. I think the process thing is something that you know we're going to have to shake up. somehow right uh you know it's like whether it's like Should we plan for a sprint or should we plan for you know a two-week sprint or ah a one-month sprint? Or is it like...
00:52:07.32
Bogdan Kharchenko
just a kanban board like we just do or something you know because sometimes you know the reason why we've kind of done this like even sprint planning or even the daily um
00:52:09.55
Chris Morrell
Right.
00:52:18.14
Bogdan Kharchenko
stand-ups is to basically make sure that things are like you know, on track, right? That's like within that cycle uh within that sprint and like does that even matter like you could probably knock out this big feature uh like when it's gonna get knocked out when it gets when it's gonna get knocked out You know?
00:52:37.91
Bogdan Kharchenko
um So I do definitely want to explore that like what does that look like and I know that amp obviously just you know is operating at a different level than i think most probably like 90 % of. ah companies out there you know i think many companies are in this like sprint world and like how do you even Uh... you know.
00:52:59.33
Bogdan Kharchenko
teach an old dog new tricks you know to to like maybe see the world a different way because you know like i don't know sometimes i feel like i'm like man am i like just uh like an unk now comparing to some of these am guys like these am guys like don't even have daily stand-ups you know and i'm just like this uh dinosaur cobalt
00:53:14.78
Nathan Mayo
but
00:53:21.38
Bogdan Kharchenko
developer you know ah
00:53:23.06
Ben Carr
I mean, the funny part is they describe themselves as capital A Agile. So it's like teaching a new dog old tricks. If anything.
00:53:28.60
Chris Morrell
Yeah.
00:53:30.27
Bogdan Kharchenko
Yeah.
00:53:31.16
Skyler Katz
app Yeah, I mean, I think that like. Part of it. Part of it that seems... Like. A thing about Cycles that was nice.
00:53:40.58
Skyler Katz
Is that. It was like, all right, every two weeks, like we're celebrating that. We've like accomplished this goal. But now... like
00:53:50.66
Skyler Katz
The goals are being accomplished sooner or... or And later. Like they're not. Like it's mid, it's like mid last cycle. Chris was like, all right, Skylar, like rebuild the education system.
00:54:03.46
Skyler Katz
the e-learning platform, like rebuild it. end. Like I didn't create. tickets for it? like Claude created tickets as I was building out the different branches of work.
00:54:14.50
Skyler Katz
And like. I finished, I started mid cycle and I'm finishing mid cycle. And it's like, well, the celebration happened yesterday when we launched the thing. And now it's the middle of the cycle. Like, I don't know what I'm supposed to work on. Like, we don't even, who cares that like every two weeks linear is like switching over to get. tickets off of the
00:54:32.15
Chris Morrell
Right.
00:54:32.32
Skyler Katz
chart. And we're meeting once a month to do a retro. And I think that. It's nice. like What's gone well this month? So like, I don't know that I even think in cycles anymore because we also don't.
00:54:43.62
Skyler Katz
do release notes anymore like we just like uh the ceremony around two weeks is gone. um
00:54:51.38
Chris Morrell
Yeah. There's maybe something to be said for like... continuing to have the concept in place as like a Oh, we know that this is a thing that we need to get to in the next couple of weeks. like let's Let's give it a home. you know, but that like.
00:55:07.34
Chris Morrell
That's... I mean, already, I think that that's just becoming more of the like. Um, exceptional case rather than the normal case right it used to be that we planned out a cycle and then we worked we worked down the cycle and it was like okay if you finish everything on the cycle then there's like a backlog that you can pick up from but like we're working on the cycle And now it's starting to feel like it's more like.
00:55:30.54
Chris Morrell
Well, we generally know what projects everyone's working on. and like As it. is useful will put things on the cycle to like give them a home. But like.
00:55:42.29
Chris Morrell
Maybe we should. We should be like leaning into that even more in that like. We're never shooting to like fill up the cycle with all the work we expect to do in the two weeks. We're only really planning to like.
00:55:54.46
Chris Morrell
put things on the cycle as a way of saying like, Oh, this small thing that someone needs to get to. like should be done in the next two weeks or not or should be done. next month sometime or not. You know, it's like more just.
00:56:07.71
Chris Morrell
allocating for the the like smaller asks because like you know, even though we can all be building like much bigger features with much more autonomy.
00:56:19.19
Chris Morrell
There's also lots of... little things that just come up. from running a product for you know, real people for a long time that like, need to have a little bit of a home and it's almost like the cycle becomes the home for that stuff More than the actual like feature product development work you know
00:56:36.35
Skyler Katz
Yeah. I mean, when we when I first started here. We did a lot. More. Bye. Something would come in on Slack from somebody and we would. just like pivot to do that stuff. And then that felt bad because it was like, how do we move big features forward? So we like went more into planning.
00:56:56.51
Skyler Katz
And now... It almost feels like. We... like want to go back to be more reactive. Two.
00:57:05.98
Skyler Katz
what customers or staff like wants. And. AI makes that like. Easy to... to ship those things and then i do wonder though like what does happen to the big features Um,
00:57:20.38
Skyler Katz
But who knows? like I think you could probably do both. Um...
00:57:23.73
Chris Morrell
Right.
00:57:24.100
Skyler Katz
But it's like finding that. Time to like. put forward what the big feature is and What is a big feature in a... 25-year-old company. I don't know.
00:57:36.85
Chris Morrell
Right. Yeah, I definitely feel like.
00:57:38.100
Skyler Katz
I mean, lots of lots of ideas. Everyone has lots of ideas.
00:57:42.04
Chris Morrell
Yeah. Well, I mean, I've been saying this for the last, what, like three months of planning. It's like we get into planning and I'm like. I don't know. Does this work for like, does this process work for us anymore? What we're like, how we approached planning before.
00:57:56.65
Chris Morrell
And like. It almost does feel like. yeah those those uh monthly planning sessions maybe become just like a Like.
00:58:08.12
Chris Morrell
What are the high level? sort of directional things that we want to make sure. Like are being accounted for Right. Because, ah you know, I think that's another thing in it.
00:58:18.71
Chris Morrell
in a organization like ours. It's just a little bit harder to have the... the type of tight focus that like they're talking about in the in that podcast right like we have Just...
00:58:33.11
Chris Morrell
You know, like. Some people... care about proctoring and other people care about The like. exam system but in a totally different you know like it's just i i don't know it's Um,
00:58:47.09
Skyler Katz
I mean, the stuff that they make fun of in that in the podcast about like. You have your ah RFC and you have your like release. and Stuff like we don't. have any of that.
00:58:56.36
Chris Morrell
We just never have done that.
00:58:57.50
Ben Carr
yeah
00:58:57.51
Bogdan Kharchenko
Yeah.
00:58:57.70
Chris Morrell
Yeah.
00:58:57.86
Skyler Katz
But like. we have people on staff who like interact with various state agencies to get approvals for education and we have people that are like writing courses and people that are like Bye.
00:59:09.36
Skyler Katz
That's stuff that like. It's just... different. It's a different type of employee.
00:59:13.84
Chris Morrell
Yeah.
00:59:16.92
Skyler Katz
And. They have different needs
00:59:20.98
Chris Morrell
Yeah. Yeah. So I think that... Us. having some sort of process in place that is about like stepping back and looking at the big picture and like understanding the tension between all the like different sort of poles of the different teams Um,
00:59:38.25
Chris Morrell
you know, what does member services need versus the education team versus special programs versus, you know, like all these things. Um, you know are vying for they might not be vying for our capacity to write code anymore but they're still vying for attention right Um,
00:59:56.41
Chris Morrell
And so... Like, I think that that's the biggest thing for me is it's like. that human component. is like... even more uh the like precious resource And figuring out how to like.
01:00:13.69
Chris Morrell
Um, account for that. More. and like worry and make our process less about the code right like The. as a dev team it used to be that a lot of a lot of our process was out about the code And...
01:00:28.79
Chris Morrell
And in a lot of ways that meant that. you know I was kind of doing a lot of the product decision. direction. Choices.
01:00:39.82
Chris Morrell
Um, But I don't think. I don't think that ever worked great for anyone. You know, it was just the reality of the situation, but it's like. Am I making a better decision than like...
01:00:53.16
Chris Morrell
If... you know. bogdan and lisa sit down and actually have a conversation no like Maybe there is a component that I know about that. is important but i can chime in there Um,
01:01:08.25
Chris Morrell
So yeah. It's so then like. there's the question of how does everyone sort of stay aligned and like aware of choices that are being made that may impact something else. and you know.
01:01:21.50
Ben Carr
I mean, to that exact scenario, if Bogdan and Lisa had a conversation about a new feature
01:01:22.01
Skyler Katz
Good night.
01:01:22.37
Chris Morrell
Amen.
01:01:26.33
Ben Carr
the ideal place for you to chime in with some opinions. would not have been the PR.
01:01:31.84
Bogdan Kharchenko
No.
01:01:32.05
Chris Morrell
ah Right.
01:01:32.41
Ben Carr
You know?
01:01:32.66
Chris Morrell
Yes.
01:01:33.28
Nathan Mayo
Hmm.
01:01:33.11
Bogdan Kharchenko
Yeah.
01:01:33.65
Ben Carr
So I think we've... Like we've never been terribly precious. As I say, never. I've been here four months. like In the four months, which seemed like a tremendously long amount of time. and also No time at all. ah it doesn't seem like we've been very precious about the the ceremony stuff which i think in the podcast is a lot of the
01:01:52.36
Chris Morrell
Right.
01:01:55.24
Ben Carr
The struggle that they describe being. uh kind of sympathetic to is these people that have these rigid processes where like If it's in its brain, it has to get done.
01:02:01.11
Chris Morrell
Right.
01:02:02.06
Ben Carr
If it doesn't get done. You're going to have a conversation with your manager. Uh... Whatever. And like every cycle it's been.
01:02:07.33
Chris Morrell
Right.
01:02:08.77
Ben Carr
you know we'll probably expect a couple of things are going to roll over the next cycle and that's fine. um And then things come up and we just add them into the cycle. and like you know the cycle sort of takes shape while we're in flight and that's already been the case
01:02:19.29
Chris Morrell
Yeah.
01:02:21.22
Ben Carr
But I do wonder... Like... You seeing the implementation in a PR of the thing that Bogdan and Lisa talked about. might spawn some like Hey, I have some ideas for how we should adjust this in the future and i'm wondering where in this new world or in the ant process even Like where would that have happened? would have been like,
01:02:40.18
Ben Carr
Bogdan and Lisa have the conversation, and then... Bogdan says... Hey, this is what happened in the conversation would have been Is it like you digging through? Chat transcripts. for their conversation to like try and get that and and inject yourself like um i don't know where that that type of thing fits.
01:02:58.83
Chris Morrell
Well, I mean, I think that to some degree it's like. being able to pass. around a video of like a fairly far baked. feature Mike.
01:03:10.70
Chris Morrell
Is. I mean, I think that that's one of the most exciting things of all of this is like. The. The ease at getting. ah a video or a screenshot or like a.
01:03:21.20
Chris Morrell
you know, a URL to go to. to look at a thing. you know or like we've been we pass clawed artifacts around it's like i feel like The cool thing there is, yeah.
01:03:33.16
Chris Morrell
a couple of people can have a conversation. come to a pretty solid understanding of like where you think you're going in direction And either hand off like. Uh...
01:03:44.02
Chris Morrell
you know, clawed artifact that's like a visual summary of sort of the key decisions there.
01:03:52.06
Ben Carr
Yeah.
01:03:52.07
Chris Morrell
Or a video of like the first draft of the feature. And like. at that point in time if i come in and say oh i actually think we should do x y and z like you know kind of going back to the cost thing.
01:04:04.76
Ben Carr
yeah
01:04:05.17
Chris Morrell
At that moment in the conversation, it's like... The cost of. incorporating or not. incorporating those comments is like so low.
01:04:16.38
Chris Morrell
Right. Because like the cost. to bogdan in this scenario of like taking on my comments and adding it is like... send that off to Claude and be like... Go do what Chris said, right?
01:04:29.09
Ben Carr
Yeah.
01:04:29.35
Chris Morrell
And then sort of like, you know, there's there's additional cost of like managing the impact of that and all that. But on the flip side, there's also like. The cost to me of Bogdan saying like, oh, for a reason A, B and C, like we're not going to do that right now.
01:04:43.72
Chris Morrell
Um, is like, oh, well. if I care enough. I can just. have Claude, do it. afterwards for me you know like
01:04:52.00
Ben Carr
Yeah.
01:04:52.43
Chris Morrell
Um,
01:04:53.64
Ben Carr
I think the place.
01:04:54.05
Chris Morrell
So both sides of it.
01:04:54.09
Ben Carr
like the cave
01:04:55.54
Chris Morrell
Yeah.
01:04:56.30
Ben Carr
Yeah, it kind of goes back to... ah The place I think that conversation would happen today for us would be in stand-up. just because it's a dedicated.
01:05:02.70
Chris Morrell
Mm hmm.
01:05:04.80
Ben Carr
time where we've all individually agreed to be present for each other.
01:05:09.35
Chris Morrell
they're Right.
01:05:09.90
Ben Carr
Like that kind of is the... The biggest benefit, I think, like if there's a thing.
01:05:13.74
Chris Morrell
Yeah.
01:05:14.57
Ben Carr
you you know, Bogdan could have posted a video in Slack. But like... There's not really the expectation. That. Everyone should respond to this thing now, please.
01:05:25.18
Chris Morrell
All right.
01:05:26.14
Ben Carr
So if there is that standing like going back to the process question, if there is something that's standing where we all agree to be there. then that's the perfect place for getting feedback. from people that might not have given it asynchronously.
01:05:39.30
Chris Morrell
Right.
01:05:40.76
Bogdan Kharchenko
i
01:05:40.72
Skyler Katz
You know, I...
01:05:41.13
Chris Morrell
I also think, oh, go ahead, Skylar.
01:05:43.63
Skyler Katz
Well, this changes the topic a little bit, so. if you want to respond.
01:05:46.07
Chris Morrell
Okay, well, I'll say one other thing, which is like... I also... think that like There's a component of... Um, Like we are now in an. era where it's a lot easier to say like You know.
01:06:02.14
Chris Morrell
Use it or lose it. Kind of like. I'm going to post the thing if I don't get feedback. like We're going to do it. And like. Before. it was a little less realistic to kind of have that attitude because it's like you know. so Shifting gears costs a lot.
01:06:19.72
Chris Morrell
But like. Yeah, nowadays... if if there was like if we started to set a little bit more of an expectation maybe we set up like a new channel that's like feedback and you post...
01:06:31.50
Chris Morrell
screenshots and videos there and like Either people go in and give feedback or they like give up the right to have had that feedback. And like. If then later something you see it and you're like, man, I don't like the way it is. You just go and.
01:06:46.98
Chris Morrell
Like I like that attitude that they talk about in the amp. that that podcast of like. If you don't like it, just... Go in and change it after the fact, you know, and like. To your point.
01:06:57.50
Chris Morrell
If it's like a major customer facing feature, you don't want to like. you know, give them whiplash.
01:07:02.54
Ben Carr
Yeah.
01:07:03.05
Chris Morrell
But in a lot of cases, I think that that's like. Not the reality you know. ah Sorry. So what were you going to say, Skylar?
01:07:11.93
Skyler Katz
Yeah, well, I think so i think like traditionally i would say that we're a developer. Like. A developer-led company.
01:07:22.29
Skyler Katz
We have these meetings every week with. the various departments and they come to us and they say like, can you please help us with this thing? And we say like, yes or no. And I actually wonder if.
01:07:35.43
Skyler Katz
Our daily process doesn't need to change, but actually these like. once a week meetings where they say the leaders are like saying this is what's important at amp and what are we going to work on this week and it's actually like Maybe it becomes like the education team is leading their.
01:07:53.72
Skyler Katz
initiatives and we're Like. going to work on those things that they are saying. are important as opposed to us saying like, well, we think that this is important for you.
01:08:04.88
Skyler Katz
And they're like, I guess it's cool.
01:08:05.20
Chris Morrell
Yeah.
01:08:06.20
Skyler Katz
Thanks for like doing this thing. But actually. We need something else. And I think there's a mixture of like. We sometimes have to lead like. Chris, you're the CEO. like sometimes you're gonna leak lead things but i do wonder if it's actually our like Wednesday calls.
01:08:23.28
Skyler Katz
that maybe need to change in their format.
01:08:25.26
Chris Morrell
Mmm. Mhm.
01:08:28.36
Skyler Katz
So like. Be more reactive to the things there requesting as opposed to like yeah I don't know. Like.
01:08:36.18
Chris Morrell
Yeah.
01:08:36.55
Skyler Katz
I don't like to give up my own power.
01:08:37.05
Chris Morrell
That's interesting.
01:08:38.45
Skyler Katz
But I do wonder if like that's the part that. changes a little bit.
01:08:43.14
Chris Morrell
You see, I I don't. My initial reaction is like. No. um just because I think that there's so much there's like so There's so much that goes into the like like the product.
01:08:57.57
Chris Morrell
decision layer, right? like and And in a lot of ways, like. In your capacity on those meetings, like you're more acting as like a product. person than a developer almost you know where it's like you know the team comes and says oh
01:09:10.24
Skyler Katz
Yeah.
01:09:12.85
Chris Morrell
I want X. And after. 20 minutes of conversation you realize oh like pain point y over here is actually what they're getting at And like. the solution to pain point Y is totally different than X, right?
01:09:26.73
Chris Morrell
Um, And so I don't know that. Yeah. it it That makes. Perfect sense.
01:09:35.75
Skyler Katz
I mean to push back a little bit. I do wonder like. If. If code is cheap and we can change things. Like. If pain point X is what they're saying is their problem and we... do it. And it.
01:09:47.14
Skyler Katz
Turns out to not solve the problem. And then we change it. Like, is that bad?
01:09:53.10
Chris Morrell
Sure. Well. It brings me to...
01:09:54.96
Skyler Katz
I think so. i think so like I do think it's bad, but I am like willing to try it out and see. like i don't know
01:10:02.90
Chris Morrell
it It brings me to a really interesting question, which I don't know how long we want to get into this because I know we have to stop relatively soon. but um I had someone. DM me yesterday about.
01:10:13.40
Chris Morrell
Um, like oh, what if we used Claude this way? And it was essentially like... I want to use Claude. To. click around the website to do a bunch of tedious tasks that I'm doing.
01:10:28.43
Chris Morrell
And I'm like, Well, this is obviously not. a good use of like this tool. Because it's like. If that. sequence of tasks is common and tedious, there should just be a feature that makes it.
01:10:44.09
Chris Morrell
Not tedious if you do it. Often. But... There's a part of me that's like. Do we just let anyone? Open a PR. Like. Do we?
01:10:56.44
Chris Morrell
basically just... introduce a way for anyone in the company. To say. This is a thing. This is like a problem that I have. Help me find a solution.
01:11:09.16
Chris Morrell
And then that just like implements a feature. And we... so then we can decide what to do with that. Um, With the understanding that like we'll probably say no to a lot of them.
01:11:21.49
Chris Morrell
But like. Is it? There's like a whole other layer here of... Does it have to be developers who are like making that? first pass, right? Because I think it's like...
01:11:35.07
Chris Morrell
The second, the it's it's getting that first pass to prod. That takes. our expertise. um But like being able to iterate on an idea.
01:11:45.78
Chris Morrell
And like come to us with. Hey. I like this. Like I can. I can click around this and I like it. is kind of interesting. you know?
01:11:56.66
Nathan Mayo
I think that's a good use case for... a tool like AMP. where you can see like the original intent
01:12:04.47
Chris Morrell
Mm hmm.
01:12:04.75
Nathan Mayo
of. person who's building that first pass like that would be a good
01:12:07.87
Bogdan Kharchenko
That thread, yeah
01:12:09.01
Nathan Mayo
a good place to see
01:12:09.19
Chris Morrell
Yeah.
01:12:10.88
Nathan Mayo
um I don't know. if That's just what's running through my mind right now.
01:12:14.09
Skyler Katz
we We can't even.
01:12:14.27
Nathan Mayo
Bye.
01:12:14.49
Chris Morrell
Skylar looks terrified.
01:12:16.18
Skyler Katz
Well, it's not that I'm terrified. It's like We... Our time to merge on agreed upon work. is already too long. And then if we add in like.
01:12:28.45
Skyler Katz
I mean... In reality, like this person, whoever they are that asked this thing is going to do it one time and never do it again. So it doesn't like, sure, we should just let anyone do it because no one's going to do it. But like.
01:12:39.84
Skyler Katz
Um, I think if we can. If we can figure out... how we decide like what of our PRs need a human review versus can just be merged.
01:12:50.50
Bogdan Kharchenko
That's a good question.
01:12:51.08
Skyler Katz
then maybe it becomes more realistic to like let anybody in the company.
01:12:51.79
Chris Morrell
Yeah.
01:12:56.99
Skyler Katz
open a PR. for a human review or not i don't know
01:13:01.13
Chris Morrell
right
01:13:01.41
Bogdan Kharchenko
I will say this kind of goes back to like this agency question because there's multiple PRs that I see, even like Skylar, for example, was working on this multiple feature thing. There's like stacked PRs.
01:13:12.21
Bogdan Kharchenko
And like, I know Skylar like looked at this thing. you know upside down you know and inside out basically and then I get a feature a code review thing and I'm like oh yeah well this kind of makes sense and I review it and it's like
01:13:27.67
Bogdan Kharchenko
Well, like I kind of trust that this work is going to get shipped and then the work gets shipped. And then, you know, there's maybe some sort of update that needs to happen a few days later because now now it's in production. We need to change something. And like, I'm okay with it, you know, but it's like, I feel like it is.
01:13:42.82
Bogdan Kharchenko
a little bit overwhelming sometimes where it's like, oh yeah, there's like this big feature. And and you know, these stack PRs are not 10 lines each, right? There's like 500 lines in every stack and it's like seven of them.
01:13:51.96
Chris Morrell
ah right
01:13:54.51
Skyler Katz
Look, I assign it to Bogdan when I want it to get merged quickly, and I assign it to Ben when I want it to get looked at, you know?
01:13:58.49
Nathan Mayo
do
01:14:01.28
Chris Morrell
ah
01:14:01.90
Bogdan Kharchenko
Hey, man, I feel like, you know, ah my position is like, yeah, let's just ship it and let's just see what the user. ah likes and doesn't like about it because it's very hard like to some degree I do agree with some of the methodology AMP has it's like you've got to get it out there you know otherwise you're just working in this like little bubble of yours and you don't know what the actual um response is going to be from the people.
01:14:25.63
Bogdan Kharchenko
Um,
01:14:25.83
Chris Morrell
Well, and it it makes me think of like. Back in the day... Like. You know. I just fricking edited files directly in prod.
01:14:33.43
Bogdan Kharchenko
on the server Yeah.
01:14:35.06
Chris Morrell
And when I did that. I was really careful. Right. I was in there being like.
01:14:39.06
Bogdan Kharchenko
Mm-hmm.
01:14:40.88
Chris Morrell
I'm... I gotta be sure. that this change is going to do what i want and not introduce some some bug Because like as as soon as I hit save, this is live, right?
01:14:54.27
Chris Morrell
And I. And there is a degree of like, I don't want to call it complacency, but I know that I. We'll well. you know two other people are going to review this pr So like.
01:15:07.61
Chris Morrell
you know If there's really an issue, someone will. Someone will catch it, you know Um, And I think there are some PRs where... that's really Good. Right. There's some code changes where you're like, I want a couple sets of eyes on this because this is like dangerous in some way. Right. We're touching off.
01:15:23.33
Chris Morrell
We're touching money. you know, like. there's there's security implication or like yeah Um, But I... think that one thing that came out of that.
01:15:34.89
Chris Morrell
like listening to that that. that really appeals to me is, and I think like we've been doing a little bit more of this and it's. I feel like it's been awesome. um of like you know with this education thing that's got her did or like with the event stuff that you did a while ago bobbed in or like That where it's like, nah, you like.
01:15:53.51
Chris Morrell
You own it entirely. And like, if something. explodes like... you know, it's going to explode in your face. So like. Get it right. But I trust you to get it right then, you know.
01:16:04.71
Chris Morrell
um And if you really need someone's input, like ask for it. but like If you don't... I trust you to do it you know what I mean like moving a little bit more to that being our default model.
01:16:17.44
Chris Morrell
Um, And in that scenario, like You know, I think from a process perspective, it's hard for us to like. just stop doing PRs right now, but like. In that scenario, just opening a PR, looking at the automated review, making sure it passes tests and then hitting merge like without. it ever.
01:16:34.58
Chris Morrell
Like. passing another person's desk for you know Seems fine to me. And we're already doing more of that. ah you know, now. But I'm...
01:16:45.24
Chris Morrell
I'm feeling like that. that should be a little bit more of our our default. you know, and and like a little bit more of the like burden comes back to the person who's writing the code of like Do I? Like.
01:17:00.24
Chris Morrell
Are there things in here that like I'm not sure if this is the right call? Right. Now I need to go out and ask, right? Is there. something in here that I'm like pretty comfortable with, but like. hmm i don't know like could this have some other implication okay now i should like ask right like Um,
01:17:19.27
Chris Morrell
I...
01:17:19.17
Bogdan Kharchenko
I will say, you know, like just having a space to talk about that, like I feel like the the daily stand-ups, the way we do it right now, are kind of like cumbersome where we're just kind of looking at tickets. And maybe that needs to change.
01:17:30.15
Bogdan Kharchenko
But it's also kind of very informative. It's like somebody's working through a feature. You raise the PR. You're like looking over your code. And like that's the place to raise this. and kind of like have this discussion i think it i think that process may not be perfect now, but I do like the fact that we do get together, even though sometimes it feels like it's a waste of time.
01:17:52.44
Bogdan Kharchenko
I feel like we make up for this waste of time with being able to iterate quickly using AI tools. um With just better kind of like, you know, making sure everybody's like aware of kind of like what the changes are are going to happen.
01:18:08.76
Bogdan Kharchenko
So I do like. just you know maybe that process exists but it's kind of like yeah you know. uh here's this thing this is how it works uh i'm very confident about it i'm gonna ship it and you know Maybe that's like a one way to like segue into this.
01:18:25.61
Bogdan Kharchenko
Uh... you know, not needing to ask for review. um you know, a ceremony type of thing. It's like, well, here, I just talked to you about it. Here's just kind of what's going out. and you know and you have to like trust that person providing you that information and then It just gets shipped.
01:18:44.91
Chris Morrell
I feel like um we could be gut we could go for a while longer, but we have ah we have a stop.
01:18:51.04
Bogdan Kharchenko
All right, her stop, yeah.
01:18:52.12
Chris Morrell
So I'm wondering if we... Just say. To be continued here. Because I do kind of think.
01:18:57.12
Bogdan Kharchenko
I like that.
01:18:58.10
Nathan Mayo
Hmm.
01:18:59.36
Chris Morrell
Um...
01:19:00.86
Ben Carr
We didn't solve anything.
01:19:01.04
Chris Morrell
I'm I'm.
01:19:02.69
Bogdan Kharchenko
Yeah.
01:19:02.62
Chris Morrell
Well, no, I kind of feel like. I... really like where this conversation landed and i feel kind of like inspired to say do we propose a couple of like real Like.
01:19:15.69
Chris Morrell
process changes that we try for the next couple of months. And I think it might be interesting to do that. Like. publicly and then report back on you know how it went. Um,
01:19:27.11
Chris Morrell
So maybe. maybe there will be a part two of this or maybe part two will be hey we did that and now it's been a couple months and Here's the update. I'm not sure what will happen, but.
01:19:37.07
Skyler Katz
we all of our customers have churned we're like everything's on fire
01:19:37.03
Chris Morrell
Um, Yeah, yeah. We're shutting the company down. we you know
01:19:41.83
Bogdan Kharchenko
Part two is all the vibe coders are now submitting.
01:19:45.46
Chris Morrell
ah
01:19:45.16
Bogdan Kharchenko
PRs and we're overwhelmed with tickets.
01:19:47.22
Chris Morrell
Yeah.
01:19:48.48
Bogdan Kharchenko
um
01:19:50.56
Chris Morrell
So we'll we'll leave we'll leave the listener hanging there.
01:19:50.75
Bogdan Kharchenko
I don't know
01:19:53.55
Chris Morrell
Any any last thoughts before we close up?
01:19:57.87
Bogdan Kharchenko
I'm very excited. for us to have this conversation I feel like you know just having the time to like just vocalize some of these things I feel like a lot of people have all types of feelings in like the software world of like, oh, my job is about to be gone.
01:20:11.83
Bogdan Kharchenko
But then on the flip side, we have a lot more programmers getting minted you know ah with the vibe coders and stuff. And you know I think there's a bunch of these challenges happening.
01:20:22.08
Bogdan Kharchenko
in all types of organizations. And we don't know which way to do it. And I can't say AMP has it right, because I think they're also discovering. And I think a lot of teams have to discover for themselves to see what fits best.
01:20:32.38
Chris Morrell
Yeah.
01:20:35.28
Bogdan Kharchenko
But I suppose we're going to. find out. how this plays out.
01:20:40.09
Chris Morrell
Nice.
01:20:40.39
Nathan Mayo
well and what fits best now.
01:20:41.14
Ben Carr
expense
01:20:42.85
Nathan Mayo
Could completely change in the next couple months.
01:20:44.12
Bogdan Kharchenko
It's also true.
01:20:45.37
Chris Morrell
it's true it's true it's a moving target yeah
01:20:45.53
Bogdan Kharchenko
It's a moving target.
01:20:46.16
Nathan Mayo
So.
01:20:46.50
Bogdan Kharchenko
Yeah.
01:20:47.30
Nathan Mayo
Like. everything is moot yeah It's moving so fast
01:20:49.86
Chris Morrell
Oh man.
01:20:50.36
Bogdan Kharchenko
So. Okay. Well, awesome.
01:20:52.01
Chris Morrell
All right.
01:20:51.86
Bogdan Kharchenko
Thank you for having us.
01:20:52.33
Chris Morrell
Well, uh, keep your Keep your eyes out for part two. We'll see.
01:20:59.49
Bogdan Kharchenko
See you guys.
01:21:00.78
Nathan Mayo
so guys
01:21:00.96
Skyler Katz
See ya.
01:21:13.63
Bogdan Kharchenko
Ting ting.
