النص 🔗
WEBVTT
00:00:00.640 --> 00:00:04.400
Hello everyone and welcome to another episode of Code and the Coding Coders to Code It.
00:00:04.559 --> 00:00:07.759
I'm your host, Drew Bragg, and I'm joined today by Mike Kudamarsh.
00:00:07.839 --> 00:00:13.279
Mike, for anyone who's not familiar with you, can you do a brief introduction for the audience?
00:00:13.839 --> 00:00:15.919
Hi everyone, this is Mike.
00:00:16.160 --> 00:00:21.920
I am a software engineer who has been obsessed with writing code.
00:00:22.079 --> 00:00:27.440
So I'm on this coders podcast, I guess, to talk about how I really like writing code.
00:00:27.600 --> 00:00:33.679
I've been doing it essentially for as long as I can remember, as soon as I started using a computer.
00:00:33.920 --> 00:00:37.520
I've worked on a bunch of cool stuff that you probably know about.
00:00:37.679 --> 00:00:42.479
I was one of the first engineers at producthunt.com.
00:00:42.719 --> 00:00:44.320
That was super fun.
00:00:44.640 --> 00:00:47.679
After that, I got the chance to work at GitHub.
00:00:47.920 --> 00:00:51.200
I was on the team that invented GitHub actions.
00:00:51.439 --> 00:00:53.679
So you might love or hate that.
00:00:53.920 --> 00:00:56.320
It's part partially my fault.
00:00:56.560 --> 00:00:58.479
Or you like it either way.
00:00:58.719 --> 00:01:07.359
And then five years ago, I left GitHub to work on a startup with a bunch of my best friends called Planet Scale.
00:01:07.840 --> 00:01:09.680
It's a database company.
00:01:09.760 --> 00:01:15.519
We do MySQL and Postgres for a bunch of really big and small, cool companies.
00:01:15.760 --> 00:01:17.840
And I've been doing that for five years.
00:01:18.000 --> 00:01:18.799
It's a lot of fun.
00:01:18.959 --> 00:01:20.400
So I'm going to keep doing it.
00:01:20.640 --> 00:01:24.159
I didn't realize you had been at Planet Scale for five years now.
00:01:24.319 --> 00:01:25.439
That sounds insane.
00:01:25.840 --> 00:01:26.959
Sounds like such a long time.
00:01:27.200 --> 00:01:28.480
It's gone really fast.
00:01:28.799 --> 00:01:30.480
Yeah, you've got quite a history.
00:01:30.640 --> 00:01:34.799
So for anyone new to the show, the way this is going to work is I'm going to ask Mike three questions.
00:01:34.879 --> 00:01:39.040
I'm going to ask him what he's working on, which sounds like we're going to get into some cool stuff.
00:01:39.200 --> 00:01:41.120
What kind of blockers does he have?
00:01:41.200 --> 00:01:46.239
And if he doesn't have a current blocker, what's a recent one that he had and how did he go about solving it?
00:01:46.480 --> 00:01:53.599
And then the last question is just asking him to share something cool, new, or interesting that he's recently learned, discovered.
00:01:53.840 --> 00:01:58.000
Doesn't have to be coding related, but this is code and the code encoders who code it.
00:01:58.079 --> 00:01:59.760
So it absolutely can be.
00:02:00.079 --> 00:02:10.479
So, Mike, I want to ask you what you're working on, but it's kind of interesting that you left GitHub to go work in a startup.
00:02:10.800 --> 00:02:15.759
And I want to talk about that real quick before we get into like what you're doing at planet scale right now.
00:02:16.000 --> 00:02:16.879
Like, why?
00:02:17.120 --> 00:02:24.639
Why would you leave a place like GitHub, which I think even five years ago was holy shit, it'd be so cool to work at GitHub.
00:02:24.960 --> 00:02:31.759
Why would you leave GitHub to go work at a startup, which is a little riskier, or at least the perception is it's a little riskier?
00:02:32.560 --> 00:02:34.159
Yeah, for sure.
00:02:34.400 --> 00:02:38.000
So my career has been, it felt like just very lucky.
00:02:38.159 --> 00:02:43.759
I keep getting to work at these companies that I never thought like I belonged to work at.
00:02:44.080 --> 00:02:51.520
And then I'd go there and get to work on these huge successful products that everyone knows.
00:02:51.680 --> 00:02:55.759
So yeah, GitHub was like absolutely, I was like, oh, how am I even here?
00:02:55.840 --> 00:03:01.039
You know, you're like opening, you're opening pull requests and you see the people who are reviewing them.
00:03:01.120 --> 00:03:03.680
It's like all these like Ruby legends.
00:03:03.840 --> 00:03:04.159
Yeah.
00:03:04.319 --> 00:03:07.840
And they're reviewing my code and like, oh, we're co-workers together.
00:03:08.080 --> 00:03:09.520
That's crazy.
00:03:09.759 --> 00:03:10.800
Yeah, yeah.
00:03:11.120 --> 00:03:13.919
So dream job material for sure.
00:03:14.159 --> 00:03:22.240
And even just like as a fan of the internet, getting to see the inside of GitHub was really fun.
00:03:22.719 --> 00:03:34.639
Because I remember when I started, I would really just go through the old pull requests and go find like, oh, this is when they invented this thing that the entire internet uses.
00:03:34.719 --> 00:03:43.599
And like, oh, here's the conversation about how they made the pull request page faster that time and you know, 2010, right?
00:03:43.680 --> 00:03:45.520
It's just like really cool stuff.
00:03:45.759 --> 00:03:47.759
I was having a great time at GitHub.
00:03:48.000 --> 00:03:48.879
And you know what?
00:03:49.120 --> 00:03:51.520
Had this not come along, I'd probably still be there.
00:03:51.680 --> 00:03:51.840
Okay.
00:03:52.159 --> 00:03:55.039
Because like it's a really good company.
00:03:55.280 --> 00:04:00.400
I was there before the Microsoft acquisition, and then we were acquired by Microsoft.
00:04:00.719 --> 00:04:04.719
Working for a public company, you make a ton of money, also.
00:04:04.800 --> 00:04:06.000
Yeah, that's a big factor.
00:04:06.159 --> 00:04:13.120
Like working for a company that has public stock, you get RSUs every month, which is amazing.
00:04:13.280 --> 00:04:17.920
You'll see people who stay at like Google or Facebook forever, and why not?
00:04:18.160 --> 00:04:21.600
You're making a ton of cash and it's a great job, it's
Load full transcript
comfortable.
00:04:21.920 --> 00:04:31.920
But the thing, at least for me, when you're in a company that big, you start to feel a little bit like a cog or like I don't know, I like my work to matter.
00:04:32.000 --> 00:04:39.279
I want to like really matter in a way that if I'm doing a bad job, it's not helping the company, right?
00:04:39.360 --> 00:04:44.480
I want to feel like every day I have to like show up and do my best because I just like I like the challenge.
00:04:44.800 --> 00:05:04.639
Then, well, important detail here when I was at GitHub, there was this team that got spun up by Sam Lambert, who is now the CEO of Planet Scale, and that team's goal was to make GitHub actions and make it really good.
00:05:05.120 --> 00:05:10.959
I remember at that time I was kind of like this mid-engineer at GitHub.
00:05:11.199 --> 00:05:16.160
I didn't feel like I was important, but I was shipping and getting a lot of things done.
00:05:16.399 --> 00:05:21.360
And they were like, we need you on this team because we want to build this thing out really fast.
00:05:21.839 --> 00:05:28.639
And then I got to work with all these, it was like the most important people at GitHub at the time, and we were working on this together.
00:05:28.959 --> 00:05:30.959
So that was like a really cool experience.
00:05:31.199 --> 00:05:35.600
I also got really close with that group of people because we worked hard on this.
00:05:35.680 --> 00:05:43.279
It was like seven days a week total death march for months, and we just all cared so much about like making Hub Actions good.
00:05:43.600 --> 00:05:52.319
I think it was a couple years later, two or three years later, after GitHub Actions launches, we had that had like that great experience with that group of people.
00:05:52.800 --> 00:06:02.959
Sam had left GitHub and then he was working on planet scale, and he called me and he's like, Oh, we're like getting the gang back together.
00:06:03.040 --> 00:06:06.480
Like it's the same people, it's all your best friends.
00:06:06.639 --> 00:06:07.199
Yeah.
00:06:07.439 --> 00:06:09.920
And I remember feeling like literally sick to my stomach.
00:06:10.000 --> 00:06:11.759
So I was like, oh, the dream job.
00:06:12.000 --> 00:06:15.120
Oh, I'm gonna leave the money of like the it was a lot.
00:06:15.279 --> 00:06:22.160
I was very nervous about it at the time, and it felt like such a big decision, but then like I did it, and I think it was like one month in.
00:06:22.319 --> 00:06:28.879
I was like, Oh, having so much fun like doing the startup thing, like yeah, getting to like build, and I'm not a cog anymore.
00:06:29.040 --> 00:06:37.680
The money, oh, that doesn't matter anymore because I I'm just like so psyched every morning to do it, and it's with all these people that I love working with, and that's why I'm still doing it.
00:06:37.759 --> 00:06:40.879
So it's the same people, we're still together hacking every day.
00:06:41.199 --> 00:06:42.079
That's so cool.
00:06:42.319 --> 00:06:45.920
And would you consider Planet Scale a startup anymore?
00:06:46.000 --> 00:06:48.480
I mean, it's it feels pretty big.
00:06:48.879 --> 00:06:50.560
Yeah, I mean, I don't think so, right?
00:06:50.639 --> 00:06:57.360
Like we've been like this for years now, where like we're responsible for some of the largest databases in the world.
00:06:57.519 --> 00:07:02.639
It's a very serious responsibility now, and it's really not so much a startup.
00:07:02.720 --> 00:07:03.759
Well, we're still small.
00:07:03.920 --> 00:07:08.240
I think we just we've been growing a lot recently, but I think we're still only 90 people.
00:07:08.399 --> 00:07:10.399
It's almost entirely engineers.
00:07:10.560 --> 00:07:11.439
It's pretty unique.
00:07:11.519 --> 00:07:16.879
It's like a very flat org of a bunch of engineers who are all very good.
00:07:17.439 --> 00:07:37.680
What was the biggest change that you had to adapt to going from a place like GitHub where I think GitHub started very much the same way, like super flat, no managers, everyone's an engineer basically, but it definitely grew into a much more mature company, pretty big, like you said, like you were starting to feel a little bit like a cog.
00:07:37.759 --> 00:07:43.759
And then you go over to planet scale, where it's a lot smaller, a lot scrappier because you're in startup mode.
00:07:43.920 --> 00:07:48.079
But what was the biggest shift for you that you had to deal with?
00:07:48.560 --> 00:08:03.600
I think something that's helped me a lot, and this helped me a lot at GitHub, where I think people they like look for structure and rules, or they ask for permission to do things.
00:08:03.839 --> 00:08:15.040
And I remember at GitHub, I'd frequently be nervous about like some project, or I'd see some gap somewhere in the business and no one was working on it.
00:08:15.199 --> 00:08:20.879
I remember I had in my notes app, I just I had it written in all caps, and I just wrote like why not me in all caps.
00:08:21.040 --> 00:08:26.160
And all the time I'd like to look at that and then I'd tell myself, and then no matter how scared I was, I'd just go do it anyway.
00:08:26.399 --> 00:08:33.279
It just like put me in situations where um I felt nervous, but then it just kept like working out, right?
00:08:33.360 --> 00:08:35.279
Like that was like a good thing to do.
00:08:35.840 --> 00:08:44.000
So going into planet scale, I felt like I was pretty prepared because there were literally no guardrails at all.
00:08:44.159 --> 00:08:51.759
It was and the the biggest guardrail of all is just having like a profitable business with customers, right?
00:08:51.919 --> 00:08:52.240
Yeah.
00:08:52.559 --> 00:08:53.840
When I started, we didn't have it.
00:08:53.919 --> 00:08:55.600
It was literally like Rails new.
00:08:55.840 --> 00:08:58.320
We don't have like database customers.
00:08:58.480 --> 00:09:05.840
And it's like figuring out what to build and just making mistakes.
00:09:06.159 --> 00:09:07.200
We did some crazy things.
00:09:07.279 --> 00:09:13.360
Like if I got to go back and do it all over again, we could probably condense five years into like a year and a half, right?
00:09:13.440 --> 00:09:18.639
Like so many things, things we did wrong, and just like figuring out and stumbling around.
00:09:18.799 --> 00:09:20.080
Like there's this concept.
00:09:20.240 --> 00:09:27.279
If you use plan to scale, we have this flow around schema changes, which we call the deploy request.
00:09:27.440 --> 00:09:41.200
And it's just it's like this really sick UI that shows like what schema is going to change, you can monitor it as it goes out into production, you can watch like the tables swap over, you can check metrics, you can revert it, all this kind of stuff.
00:09:41.440 --> 00:09:45.919
Now that's been five years and we've had that, it just seems like an obvious thing.
00:09:46.240 --> 00:09:50.480
But back then it didn't exist, and we were just like stumbling around in the dark.
00:09:50.559 --> 00:09:54.399
Like I remember the meeting where like, what the hell do we even call this thing?
00:09:54.480 --> 00:09:56.080
It's like a I don't know.
00:09:56.559 --> 00:10:02.639
And we just someone said deploy request, and now it's like a major thing that all of these tens of thousands of people use.
00:10:02.799 --> 00:10:05.039
So now there's structure and it's a lot easier.
00:10:05.279 --> 00:10:14.879
But back then it was very much like, why not me be the guy comes up with this, reposes it, and uh it's always scary, like you're just nervous all the time.
00:10:15.120 --> 00:10:15.679
Sure, sure.
00:10:15.919 --> 00:10:20.639
With more consequences because the business could fail, and then you're yeah, I guess what's worse can happen?
00:10:20.720 --> 00:10:22.399
I bet GitHub for my job back.
00:10:22.720 --> 00:10:24.159
I mean that's not that bad.
00:11:45.389 --> 00:11:49.309
No, I can imagine that it's a great place to be in.
00:11:49.469 --> 00:11:54.350
I get to go work on this cool thing with people I like, and if it doesn't work out, I'll just go back to GitHub.
00:11:54.829 --> 00:11:55.149
Darn.
00:11:55.469 --> 00:11:56.269
Yeah, terrible.
00:11:56.829 --> 00:12:02.029
Woe is yeah so now you're at Planet Scale and you've been there for a while.
00:12:02.189 --> 00:12:03.709
It's a profitable business.
00:12:03.870 --> 00:12:06.509
So what are you like, what are you working on there?
00:12:06.669 --> 00:12:10.429
What's something that like big project that you're working on?
00:12:10.589 --> 00:12:12.189
Are you working on a lot of little things?
00:12:12.429 --> 00:12:15.309
Is it something kind of more mundane now?
00:12:15.549 --> 00:12:17.469
What are you working on at Planet Scale?
00:12:17.949 --> 00:12:24.750
So my job is I lead the team that is responsible for pretty much everything a user sees.
00:12:24.909 --> 00:12:28.509
So dashboards, APIs, CLIs.
00:12:28.669 --> 00:12:31.230
It's 2026, so MCP servers.
00:12:31.469 --> 00:12:35.949
Anything that anything you interact with, that's my team.
00:12:36.269 --> 00:12:43.709
So I do a lot less of direct writing code and project work than I used to.
00:12:43.949 --> 00:12:47.230
Usually the big projects, those those go to people on the team.
00:12:47.389 --> 00:12:54.909
I'm still writing a ton of code though, but the code I'm writing is not like I'm not like taking the the big coolest project on the team.
00:12:54.990 --> 00:12:58.429
It's more like, how do I help the team work faster?
00:12:58.509 --> 00:13:00.189
How do I make CI faster?
00:13:00.350 --> 00:13:02.669
How do I make sure we're more reliable?
00:13:02.829 --> 00:13:03.949
Less bugs.
00:13:04.269 --> 00:13:06.509
Can I prototype out something weird?
00:13:06.669 --> 00:13:09.149
That kind of work, yeah, which is really fun.
00:13:09.389 --> 00:13:20.990
The big thing for this year, I'm curious if you felt this too, is uh just like the AI tooling has gotten so good that it's almost like a new job.
00:13:21.309 --> 00:13:22.990
Yes, it's different, right?
00:13:23.069 --> 00:13:24.029
Do you feel that too?
00:13:24.269 --> 00:13:25.230
Yeah, for sure.
00:13:25.389 --> 00:13:31.789
I'll admit, like, I have been historically extremely bearish on AI when it first came out.
00:13:31.870 --> 00:13:34.509
I was like, yeah, it's a glorified search engine.
00:13:34.669 --> 00:13:38.829
My Google foo has gotten good enough where none of this matters, it's not impressive.
00:13:38.909 --> 00:13:42.509
And then it started picking up, and then it started getting good at writing code.
00:13:42.669 --> 00:13:45.949
And I was still like, yeah, it's it's all right at writing code.
00:13:46.029 --> 00:13:47.870
It's not writing good code though.
00:13:48.110 --> 00:13:54.350
But yeah, in the past, definitely in the past six months, I've started to use it a lot more.
00:13:54.429 --> 00:14:01.389
I've started to work with it enough and understand it enough that I have it like more tailored to the way that I want to work.
00:14:01.469 --> 00:14:05.230
And now it's starting to actually be pretty damn useful in my day-to-day.
00:14:05.389 --> 00:14:07.469
I still really like writing code.
00:14:07.709 --> 00:14:15.789
So like I do have it write code when it's like boilerplate or something boring that I'm like, I know how like I've done that a hundred times already.
00:14:15.870 --> 00:14:17.870
I don't feel like writing it, you go into it.
00:14:18.189 --> 00:14:24.509
But yeah, it sort of feels like JavaScript frameworks back in the day where like there's something new every day.
00:14:24.669 --> 00:14:28.269
Like I wake up every day and there's something to read about something going on.
00:14:28.509 --> 00:14:39.309
You can go as deep as you want into it, because I don't know how machine learning works, let alone how LLMs and every layer of this stack now works.
00:14:39.549 --> 00:14:41.149
So there's a ton to learn.
00:14:41.389 -->
00:14:43.069
A lot of the job is the same.
00:14:43.149 --> 00:14:52.269
We're figuring out how to build the things that we want to build, but it's dramatically different on how we do it now.
00:14:52.509 --> 00:14:55.149
Is there an AI policy at Planet Scale?
00:14:55.309 --> 00:15:00.909
Do you guys just kind of use whatever, or is there a very structured way of how you guys are approaching it?
00:15:01.870 --> 00:15:05.789
So we have access to basically everything.
00:15:06.029 --> 00:15:10.669
I guess a lot of people use clawed code, a lot of cursor.
00:15:11.549 --> 00:15:18.990
And I would say the policy is hey, use whatever tool you want, and here's an unlimited budget.
00:15:19.309 --> 00:15:25.949
Go figure out the, you know, how can you automate your job or you know, make yourself more productive.
00:15:26.110 --> 00:15:29.230
Yeah, or like how can you do things better for our customers?
00:15:29.469 --> 00:15:31.549
Yeah, I would say there's really no limits.
00:15:31.789 --> 00:15:46.750
I remember like when we first started getting into these tools, our CEO at the time was like, yeah, if you want to do your personal projects and have that built the company, just so you're getting used to these tools, like, you know, go for it, hack on whatever because it's gonna help you.
00:15:46.909 --> 00:15:51.709
I'm pretty sure 100% of the company now, everyone has something that they're doing.
00:15:52.029 --> 00:16:00.589
For myself, even in just the past maybe three months, I kind of went from I've always been a Vim like TMUX guy.
00:16:00.750 --> 00:16:12.429
I went from just using, you know, autocomplete stuff like co-pilot, and then I started using Claude Code and Cursor Agent in a TMUX pane to sort of help me.
00:16:12.509 --> 00:16:15.069
But then I was still like jumping into Vim a lot.
00:16:15.389 --> 00:16:18.829
And then most recently I got really into cursor.
00:16:18.909 --> 00:16:20.829
Now I'm like actually I'm running VS Code.
00:16:20.909 --> 00:16:33.230
That's a thing I would never do, but I found like the productivity gain has been so good for me that most of the stuff I'm working on, I'm actually in cursor and I'm just using I'm using Vim a lot less.
00:16:33.469 --> 00:16:34.829
My discovery in the past two weeks.
00:16:34.909 --> 00:16:36.350
So I gotta let you in on this one.
00:16:36.509 --> 00:16:37.709
This stuff's wild.
00:16:37.949 --> 00:16:48.990
So I was thinking the other day, I tweeted about this a little bit, and I almost don't want to tweet about it, and I almost don't want to tell people about it because I feel like I'm at some advantage.
00:16:49.309 --> 00:16:53.949
I don't want the competitors to know because it's so good, but I'm just gonna leak it out to you.
00:16:54.029 --> 00:16:56.269
I'm just gonna tell you what to do and to try.
00:16:56.509 --> 00:16:58.750
So we talked a little bit about this.
00:16:58.990 --> 00:17:06.589
There's all these things that like every development team does that you
wish you could do every day, right?
00:17:06.750 --> 00:17:07.789
But like you can't.
00:17:07.870 --> 00:17:12.190
You're doing projects or dealing with bugs, there's like all this stuff going on.
00:17:12.430 --> 00:17:17.549
So for me, in like my role, I care a lot about app performance, of course.
00:17:17.630 --> 00:17:18.910
Like, is our app fast?
00:17:19.069 --> 00:17:21.549
That's important, and that's a hard one to maintain.
00:17:21.789 --> 00:17:25.630
I care a lot about our test suite being fast.
00:17:26.190 --> 00:17:28.670
So the whole team has great velocity.
00:17:28.829 --> 00:17:30.190
That's important to me.
00:17:30.509 --> 00:17:34.430
I care about like we use sentry for error tracking.
00:17:34.589 --> 00:17:37.789
When you log into sentry, I don't want it to be a garbage fire, right?
00:17:37.870 --> 00:17:38.829
Like I, you know, sure.
00:17:39.069 --> 00:17:42.589
I want people to look at it and be like, oh, I know what's going on and I know it's wrong.
00:17:42.750 --> 00:17:44.190
That's important to me.
00:17:44.350 --> 00:17:48.750
So I've been doing just been writing prompts basically in the cursor.
00:17:48.910 --> 00:18:06.430
And if anyone's not familiar with cursor, you can set up something called plan mode where you essentially give it a prompt, and then what it does is it researches your code base and in sort of any data that you give it, and then it writes a markdown doc with a plan of what it could do.
00:18:06.750 --> 00:18:10.829
A couple months ago at PlanScale, we launched an MCP server.
00:18:11.150 --> 00:18:13.470
And I worked on that a lot.
00:18:13.710 --> 00:18:22.350
Essentially, what that is, is it's just giving API access to an LLM so that they can pull data and have access to it.
00:18:22.589 --> 00:18:30.910
Where this gets super cool is so if you think about this, you're in plan mode and it has the context of your code base.
00:18:31.310 --> 00:18:34.509
But then also let's feed in data from somewhere else.
00:18:34.830 --> 00:18:37.230
So PlantScale has an MCP server.
00:18:37.470 --> 00:18:54.269
We added a tool to it where the LLM can examine your query patterns in your production database, like what's slow, which ones are missing an index, which ones using a ton of bandwidth.
00:18:54.509 --> 00:19:03.950
And just like as an experiment, because we were launching an MCP server and I'm doing this stuff, a prompt like, hey, go use the Planet Scale MCP.
00:19:04.269 --> 00:19:05.870
The code base is right here.
00:19:06.029 --> 00:19:06.430
Yeah.
00:19:07.070 --> 00:19:10.350
Find like find where that query is bad.
00:19:10.670 --> 00:19:11.790
Let's fix some stuff.
00:19:11.870 --> 00:19:13.150
Okay, let's do it.
00:19:13.550 --> 00:19:17.710
And it was finding things that were really valuable.
00:19:18.029 --> 00:19:27.150
And I was then opening a pull request, and we were like making schema changes or fixing N plus ones or whatever and making your app faster.
00:19:27.310 --> 00:19:28.190
That's cool.
00:19:28.350 --> 00:19:35.310
Here's like the big secret thing or the new discovery in the past couple of weeks is that's cool, but it still involves me.
00:19:35.470 --> 00:19:37.630
You know, like I'm at my computer.
00:19:38.269 --> 00:19:41.550
Cursor came out with, they've had this for a while.
00:19:41.630 --> 00:19:51.390
So they have a thing called cloud agents, where essentially you can run your Devit environment in a sandbox, and then they can do the plan mode, they have access to your code and all this stuff.
00:19:51.630 --> 00:19:59.550
They launched something called Cursor Automations, where you can it maybe it sounds complicated, it's easy.
00:19:59.710 --> 00:20:02.110
Go try it, it's great, it's so easy.
00:20:02.670 --> 00:20:07.390
There's a UI, and as a person who builds dev tools, I'm impressed with what they did here.
00:20:07.550 --> 00:20:08.269
They did a good job.
00:20:08.910 --> 00:20:13.950
So you can attach to this cloud agent MCP servers.
00:20:14.350 --> 00:20:18.830
So for example, PlantScale MCP, okay, now I've got my query data, right?
00:20:18.990 --> 00:20:19.630
That's awesome.
00:20:20.269 --> 00:20:22.910
Sentry has a very impressive MCP.
00:20:23.470 --> 00:20:29.310
Connecting that, now you've got your errors, you've got your performance data if you have that in Sentry.
00:20:29.790 --> 00:20:37.550
You could do data dog, you could do uh if you do build kite for your CI or GitHub Actions, there's MCPs for that.
00:20:38.590 --> 00:20:40.430
So you have all of that data.
00:20:40.590 --> 00:20:44.670
You can start writing prompts that are like, I call it my bot army.
00:20:44.750 --> 00:20:46.509
I have a whole bunch of these running now.
00:20:47.230 --> 00:20:49.070
This is my cheat code, everyone.
00:20:49.310 --> 00:21:02.990
So, for example, one of them that I like the most is we've had this idea of the like self-improving database where you can set up a cron, so say every day at 7 a.m.
00:21:03.470 --> 00:21:14.350
This thing goes and looks at your most expensive queries in production, like the ones that are reading a ton of data but not returning much data.
00:21:14.590 --> 00:21:15.470
Everyone has these.
00:21:15.630 --> 00:21:19.310
It's like, oh, this query read a million rows, but it returned one row.
00:21:19.550 --> 00:21:20.269
What a waste.
00:21:20.430 --> 00:21:22.830
And it was slow, and it's probably easy to fix, right?
00:21:23.230 --> 00:21:28.830
Turns out when the LLM has this data plus it has your code base, it's really good at fixing it itself.
00:21:28.990 --> 00:21:29.310
Yeah.
00:21:29.630 --> 00:21:39.950
So these cursor automations, they can use this data, make the fix, open the pull request on your behalf.
00:21:40.509 --> 00:21:47.070
Then you wake up and you look at it, and I was like, oh 90% of these are good.
00:21:47.550 --> 00:21:48.670
That's impressive.
00:21:48.910 --> 00:21:49.230
Yes.
00:21:49.470 --> 00:21:51.950
That you get 90% good too.
00:21:52.190 --> 00:21:59.230
Because it's still I still feel like we're running into 50% of the code that the AI writes is good code.
00:21:59.550 --> 00:22:02.590
It's like really solid.
00:22:02.670 --> 00:22:11.710
But I guess because you're you're scoping it down so much, you're giving it enough data to work with and a very narrow request, hey, make this improvement.
00:22:12.029 --> 00:22:14.269
There's not too much thinking or reasoning.
00:22:14.350 --> 00:22:22.830
So I guess it makes sense that it would be able to tackle something like that and take it off your plate or allow you to focus on the other stuff.
00:22:22.990 --> 00:22:26.830
Like that stuff matters, but it always gets swept under the under the rug.
00:22:26.990 --> 00:22:29.550
That's tech debt, that's never the priority.
00:22:29.870 --> 00:22:30.269
Yep.
00:22:30.509 --> 00:22:32.350
Now give it to the bot army.
00:22:32.430 --> 00:22:33.790
And is exactly what you said.
00:22:34.110 --> 00:22:49.630
Someone said this to me that management often feels like you have no idea what you're doing, but you are incrementally seeing behavior you don't like and then trying to correct it, and it's just an endless cycle.
00:22:49.950 --> 00:22:51.070
That's what people.
00:22:51.710 --> 00:23:04.029
It's very similar feeling with, I'll call it my bot army, very similar feeling where I started with these sort of simple scoped prompts of what I want them to do and accomplish.
00:23:04.670 --> 00:23:07.870
And I just kept watching it and seeing what would happen.
00:23:08.110 --> 00:23:13.070
And then when the result wasn't good, I wrote some really you know dumb code occasionally.
00:23:13.390 --> 00:23:16.190
I just add a line, hey, don't do this.
00:23:16.509 --> 00:23:17.790
And then it stopped.
00:23:18.029 --> 00:23:20.430
So just iterating through and it kept getting better.
00:23:21.070 --> 00:23:24.509
And at this point, like we have bots that are improving the database.
00:23:24.750 --> 00:23:31.710
We have one of my favorite bots is just watching sentry for errors and then it fixes them.
00:23:32.990 --> 00:23:33.950
What a concept.
00:23:34.110 --> 00:23:34.750
What a concept.
00:23:34.910 --> 00:23:36.590
An error comes through and you just fix it.
00:23:36.750 --> 00:23:38.110
Like that's brilliant.
00:23:38.350 --> 00:23:42.029
How have we gotten this far in software without doing that?
00:23:42.670 --> 00:23:43.070
Yeah.
00:23:43.310 --> 00:23:48.029
The thing that's interesting is it fixes ones that I might just not fix.
00:23:48.190 --> 00:23:48.430
Yeah.
00:23:48.590 --> 00:23:53.070
You'd see it and it's like, oh, well, this is happening to a user on a weird edge case.
00:23:53.150 --> 00:23:56.590
And I have so many other worst problems to deal with today.
00:23:56.830 --> 00:24:02.190
But if the bot's doing it basically for free, it's billing tokens and it will do it.
00:24:02.269 --> 00:24:04.190
I only have to review the pull request and merge it.
00:24:04.350
--> 00:24:04.670
Sweet.
00:24:04.830 --> 00:24:06.910
And then it's it's out of century.
00:24:06.990 --> 00:24:11.870
And I was looking at a graph of our total issues in Century.
00:24:12.029 --> 00:24:14.990
When I had turned the bot on, there was just like a sudden drop.
00:24:15.310 --> 00:24:17.070
It started to get lower.
00:24:17.390 --> 00:24:24.990
So I don't know if we'll get to the point where it's like we have this pure century page where you look at it and there's only like five important things.
00:24:25.150 --> 00:24:28.190
That would be the dream, but it feels possible with this audit.
00:24:28.990 --> 00:24:33.710
It's sort of like having a couple of extra mid to junior level engineers on the team.
00:24:33.950 --> 00:24:37.950
Still a little bit of hand holding involved, still gotta course correct them.
00:24:38.029 --> 00:24:39.710
You can't just blindly trust them.
00:24:39.950 --> 00:24:47.630
But if you give them a scoped enough task, they can just go off and do it and take one thing off your plate.
00:24:47.790 --> 00:24:50.670
So exactly how it feels.
00:24:50.990 --> 00:24:56.430
So something that you said earlier on, like Planet Scale basically started as Rails new.
00:24:56.590 --> 00:24:56.910
Yep.
00:24:57.070 --> 00:24:59.950
Is Planet Scale still a Rails monolith?
00:25:00.670 --> 00:25:06.990
So the way our stack works is in pretty much all infrastructure companies, there's two things.
00:25:07.070 --> 00:25:10.350
There's the control plane, and then there's the data plane.
00:25:10.590 --> 00:25:19.470
The control plane, that's what users interact with, like your dashboard, your API, CLI handles billing off, all that kind of stuff.
00:25:19.790 --> 00:25:24.190
The data plane would be that's where the databases are deployed.
00:25:24.350 --> 00:25:25.950
So two separate areas.
00:25:26.269 --> 00:25:34.190
For us, our company is the majority of everything is Ruby, TypeScript, or Go.
00:25:34.750 --> 00:25:36.110
There's a ton of Go.
00:25:36.990 --> 00:25:41.710
The control plane for us is two apps.
00:25:41.870 --> 00:25:47.550
We have a front-end Next.js React app, and that's the UI.
00:25:48.110 --> 00:25:54.990
Then the backend for the UI is a big Rails app.
00:25:55.070 --> 00:26:49.510
And it's essentially, we started it, I think it was in API mode, and it does a bunch of REST endpoints, and that's what the UI talks to.
00:26:49.830 --> 00:26:52.470
There's a ton of other cool stuff in there.
00:26:52.630 --> 00:26:55.910
It's a bunch of background jobs, running sidekick.
00:26:56.070 --> 00:27:03.030
That's what's keeping everyone's databases going, creating them, rolling out schema changes, all of that stuff.
00:27:03.350 --> 00:27:10.310
We have some kind of traditional like HTML ERB admin pages going to.
00:27:10.710 --> 00:27:17.750
So yeah, it's all like the fun parts of Rails, and then all that's talking to the data plane, which and that's just like a bunch of go.
00:27:17.990 --> 00:27:29.430
So that's awesome that even like a new and big company like Planet Scale just spun up like five-ish years ago as a Rails app when everyone still will tell you Rails is dying.
00:27:29.590 --> 00:27:32.230
It's like, no, it's not, it's still awesome.
00:27:32.550 --> 00:27:42.870
And in a way, it's like it's probably not even gonna matter in a few more years, just because the AIs are getting progressively better at writing code.
00:27:42.950 --> 00:27:48.150
So it's gonna come to a point where it's like it almost doesn't make sense to write your own code anymore.
00:27:48.310 --> 00:27:50.390
Like you do that for fun, not for work.
00:27:50.790 --> 00:27:53.750
And then, like, then what does it matter what you use?
00:27:53.910 --> 00:28:01.190
You're just gonna pick the most token efficient language and the one that the LLMs can operate on the best.
00:28:02.310 --> 00:28:04.070
I think about that frequently too.
00:28:04.230 --> 00:28:07.750
I was like, who's buying vent frameworks anymore?
00:28:07.910 --> 00:28:10.870
Like this era is seems to be ending.
00:28:11.190 --> 00:28:19.590
I mean, Rails' biggest advantage when it came out was the convention over configuration was like the speed amplifier.
00:28:19.830 --> 00:28:22.390
You kind of get that with AI now too.
00:28:22.470 --> 00:28:40.150
It's like less convention over configuration and more of like you give the guard rails to the AI and then it just can go off and do it, which actually I think is an argument for Rails in the AI era of like Rails has already defined to the AI the criteria of where everything needs to go.
00:28:40.310 --> 00:28:49.030
It doesn't have to think, you don't have to worry about it randomly deciding like this one model is gonna go in this random directory over here, or oh no, where should I put this thing?
00:28:49.190 --> 00:28:57.590
Like it's a very clearly defined convention for where everything should go, which solves some of the problems with AI.
00:28:57.910 --> 00:29:03.430
You know, the Ruby side, or allegedly Ruby's really token efficient and fast for AI to write.
00:29:03.590 --> 00:29:06.470
So maybe Ruby and Rails is gonna win that.
00:29:06.710 --> 00:29:16.630
But yeah, it is such a weird feeling for like to go through this whole career and like you get very used to it, and you're like, Yeah, this is what I'm gonna do.
00:29:16.870 --> 00:29:24.790
It'll change, it always changes, but like I've got the core of it down, and now it's suddenly like I feel like I don't know what I'm doing again.
00:29:25.030 --> 00:29:34.150
I can do the code writing and I can do the planning all on my own still, but I'm so much slower if I don't use AI just because like of course you are.
00:29:34.470 --> 00:29:36.390
Like I love it for research mode.
00:29:36.470 --> 00:29:46.710
Yeah, I can do so much more research, I can do so much more back and forth with it and like talking to it instead of just being like, build me an X, Y, or Z.
00:29:46.790 --> 00:29:49.350
I
tell it, like, I'm going to try and accomplish this.
00:29:49.430 --> 00:29:55.990
Here are some like preliminary thoughts, and like have it challenge me or have it go off and be like, Well, this is a part of the code base you're gonna be in.
00:29:56.150 --> 00:29:57.430
Have you considered X?
00:29:57.910 --> 00:30:17.270
Such an interesting tool, and it's so crazy how fast we I feel like we went from AI is this very niche scientific community thing to AI as a chat bot to like holy shit, your job is completely different because of this thing, and it's just pattern matching.
00:30:17.350 --> 00:30:26.790
I mean, it's very robust and fast pattern matching on an insanely huge data set, but it's just pattern matching, it's wild, so crazy.
00:30:27.430 --> 00:30:33.110
It's gonna be tough to like imagine like you're just graduating right now and you're getting into this.
00:30:33.430 --> 00:30:50.390
Because on one hand, it could be an advantage because sometimes my mindset is I struggle with going back and thinking the ways I used to do things, where I'll look at a problem and I go, that's expensive to solve, so I'm not gonna solve it.
00:30:50.630 --> 00:31:00.950
But if you don't have that, those years of of pain, and your only idea is like, oh, AI tool can do this for me, you just jump right in and you can just get more done.
00:31:01.110 --> 00:31:11.670
I mean, I think for us who've been doing it for so long, we just have to keep reminding ourselves anytime you kind of have that like nope attitude of, I know that's gonna be painful because I did that five years ago.
00:31:11.830 --> 00:31:14.710
Well, it's not painful anymore, so why don't we just jump in and try it?
00:31:14.870 --> 00:31:15.190
Yeah.
00:31:15.510 --> 00:31:20.070
How do you guys handle AI contributions at planet scale?
00:31:20.230 --> 00:31:27.830
Is it a AI can write the code, the commit, the pull request, and then a human has to review it?
00:31:27.990 --> 00:31:30.470
Or are you having AI review AI's code?
00:31:30.550 --> 00:31:49.590
Is it like, I know Amazon just because of their insane outages recently, were basically like a senior needs to sign off on all junior and mid-level code that was written by AI, which seems like a really sane take to begin with, but again, I'm pretty bearish on AI, so what do I know?
00:31:49.830 --> 00:32:01.430
Do you have any hard, fast rules for deploying AI code, or is it kind of like you have a enough trust in your engineering team that it's like, listen, we believe you when you say this is good?
00:32:01.750 --> 00:32:15.270
I would say the advantage of working at Planet Scale is essentially every engineer at Planet Scale would be like a staff or principal or distinguished engineer somewhere else.
00:32:15.430 --> 00:32:17.510
Like everyone is just very senior.
00:32:17.830 --> 00:32:22.070
So, like how we've approached this has a it's evolved really quickly.
00:32:22.390 --> 00:32:31.430
Even maybe in the past month or so, we've been using Bugbot a lot, which is cursor's code review tool.
00:32:31.750 --> 00:32:41.590
It's very good and it's gotten much better quickly, and it catches things that humans miss all the time.
00:32:41.910 --> 00:32:43.190
Quite impressive.
00:32:43.510 --> 00:32:48.710
So the stage where we're at now is a human always has to sign off, right?
00:32:48.950 --> 00:32:50.310
Review it and sign off.
00:32:50.550 --> 00:32:58.550
But we have bug bot running on every pull request, and it's finding like the majority of the things.
00:32:58.790 --> 00:33:14.470
There's like different levels to reviewing things where the bug bot will be really good at hey, you forgot this edge case, or this is going to cause an issue with this other system you didn't think about, or if a user enters this weird thing, like that's gonna happen.
00:33:14.630 --> 00:33:18.710
And there are things like I would just miss and maybe find in production a month later.
00:33:18.950 --> 00:33:20.630
So it's very valuable.
00:33:21.270 --> 00:33:26.790
The stuff that it's not good at is really just like the human thing.
00:33:26.870 --> 00:33:30.390
Still, like, should we even be building this?
00:33:30.630 --> 00:33:32.230
Should we even be shipping this?
00:33:32.710 --> 00:33:36.230
Or like, does this code even belong in this system?
00:33:36.630 --> 00:33:42.550
It could have been done so much easier somewhere else, but BugBot doesn't have that context.
00:33:42.790 --> 00:33:43.430
Right, right.
00:33:43.670 --> 00:33:49.670
So yeah, it's a mix, but I I do like having an AI tool review.
00:33:49.910 --> 00:34:01.910
We found it so valuable that we actually set, we said in GitHub recently that it's a we want it to run before we double anything because it's been finding so many things, and it just makes it easier for us to review.
00:34:02.150 --> 00:34:08.789
The other side of that is are humans writing code or is AI doing it?
00:34:09.110 --> 00:34:14.789
We sort of like write things about our culture internally, and we have this one document that talks about AI usage.
00:34:15.110 --> 00:34:22.710
And the model we've followed is like if your name's on the commit, like you are responsible for that.
00:34:22.950 --> 00:34:26.950
No matter you prompted it as some other tool, like this is still your code.
00:34:27.110 --> 00:34:29.829
You're yeah, you're the one proposing it.
00:34:30.070 --> 00:34:45.030
So long as people have that mindset that I need to understand what I'm putting in production and I'm responsible for it, even though it was easy, you know, they feel the weight of it and they're careful about getting that change out there, even though it felt easy.
00:34:45.110 --> 00:34:47.829
That's just sort of the balance they have to strike.
00:34:48.150 --> 00:34:58.150
So, one thing that has come up a lot, just when I'm talking to colleagues, talking to people outside of work, talking to my CTO at our one-on-one's whatever.
00:34:58.309 --> 00:35:01.190
And I'm gonna put
you on the spot for your answer.
00:35:01.510 --> 00:35:03.909
Does code quality matter anymore?
00:35:04.070 --> 00:35:08.869
Code quality has always been a subjective thing, like it's never been really easy or clear to define.
00:35:08.950 --> 00:35:13.510
So it's always been a weird thing to be like, it matters when you can't even define it.
00:35:13.750 --> 00:35:24.309
But like if you had a definition with the fact that AI can move so fast over code when needed, does code quality matter anymore?
00:35:24.710 --> 00:35:25.349
Yes.
00:35:25.829 --> 00:35:29.110
That's my answer, but I feel like I'm in the minority.
00:35:29.269 --> 00:35:30.309
So please explain.
00:35:30.710 --> 00:35:31.590
Interesting.
00:35:31.909 --> 00:35:45.430
I think I've always been the type of developer I am has always been, I am almost a hundred percent focused on the customer and the outcome and the result of the code.
00:35:45.670 --> 00:35:58.710
I will do insane things to make a customer happy and let the code base suffer or take shortcuts, except for when those shortcuts also then impact a customer.
00:35:58.950 --> 00:35:59.269
Right, right.
00:35:59.590 --> 00:36:04.869
Or they hurt the team velocity, which is also impacting the customer and the business.
00:36:05.510 --> 00:36:20.309
So thinking about models looking at your code and if code quality matters, I think it absolutely does, because just like a human, they are reading it too.
00:36:20.550 --> 00:36:23.670
And you'll notice this, right, when you're building things.
00:36:23.909 --> 00:36:33.590
If you give an LLM a total mess of a spaghetti class, it's just gonna be more challenging.
00:36:33.670 --> 00:36:41.750
Like we talked about this earlier, where it's like scoped problems, it's gonna do a better job at versus a big unscoped mess.
00:36:42.070 --> 00:36:55.029
So really it's just thinking about the same as hey, if my coworkers are able to read this, I'm writing code and I want my co-workers to be able to read and understand this six months months from now, the complexity there, I think it's the same thing.
00:36:55.190 --> 00:36:57.349
Like the models will benefit from it too.
00:36:57.590 --> 00:36:58.630
I can agree with that.
00:36:58.949 --> 00:37:00.630
There's a lot of slop going around.
00:37:00.710 --> 00:37:14.389
And there seems to be a lot of people that think that because the AIs are so fast, because the models can absorb all of this context and keep it all in their window and iterate on it so quickly, it doesn't matter.
00:37:14.710 --> 00:37:19.989
We're in that stage now where everyone just throwing money at it, but that can only last so long.
00:37:20.150 --> 00:37:22.230
Eventually, budgets are gonna tighten again.
00:37:22.309 --> 00:37:29.590
And the first thing that's gonna tighten is hey, instead of dropping thousands of dollars on this model a month, like you have a token budget.
00:37:29.989 --> 00:37:33.909
I am curious about will tokens get so expensive at some point.
00:37:34.230 --> 00:37:34.550
Yeah.
00:37:34.789 --> 00:38:00.309
We had our dev meeting a few hours before this recording, and among the things getting discussed, I brought up, I think really what the future of AI looks like when you're looking at some of these open source models that are very quickly closing the gap in their skill with the proprietary ones, you know, the codexes and the opus and what have you.
00:38:00.550 --> 00:38:03.670
They're getting really close already.
00:38:04.150 --> 00:38:06.389
And they're getting way more efficient.
00:38:06.550 --> 00:38:08.550
I don't know if you use LLM fit.
00:38:08.789 --> 00:38:13.989
It's a command line tool to like look at what open source models could run on your current hardware.
00:38:14.150 --> 00:38:16.710
It's insane what I can run on my MacBook Pro.
00:38:16.869 --> 00:38:19.670
If I had a beefier machine, even more.
00:38:20.070 --> 00:38:27.909
But like just on my MacBook Pro, I can run some pretty decent models locally for free, free in quotes.
00:38:27.989 --> 00:38:29.349
Like obviously I'm still powering it.
00:38:29.429 --> 00:38:33.829
Now my computer's got to be on and all of those things, but I'm not paying an API cost.
00:38:34.150 --> 00:38:37.670
And because it's on my machine, I can now tailor its training.
00:38:37.750 --> 00:38:51.190
Like I can just go in and tweak its training completely where I don't even need to like give it those agent MD file guardrails or skills or commands because its training can be exactly what I want it to be.
00:38:51.429 --> 00:38:57.829
I don't think it's quite to the level where that is feasible for the work and velocity that we're looking for.
00:38:57.989 --> 00:39:01.029
I think we're still going to be using Opus and Codex for a while.
00:39:01.269 --> 00:39:12.469
But I think there's a not too distant future where AI usage is what are you running on your machine, like on-prem, and what does that training look like?
00:39:12.949 --> 00:39:18.869
I'm curious what it's like at a much larger companies, you know, 50,000 people.
00:39:19.110 --> 00:39:20.710
What are they they doing?
00:39:21.029 --> 00:39:25.110
Because you can see, yeah, the the token cost must be outrageous.
00:39:25.509 --> 00:39:28.309
But it's a completely different world at those larger companies, too.
00:39:28.389 --> 00:39:41.269
Because I've heard from people who work at those very large types of companies, very enterprisey, that they have like a token budget, but not in the budget that we're talking about, where they have like a minimum token usage that they're supposed to use.
00:39:41.429 --> 00:39:42.789
Like you're not using this.
00:39:42.949 --> 00:39:47.029
Okay, it's like honestly, when I heard it, I'm like, you're out of your freaking mind.
00:39:47.509 --> 00:40:04.550
Didn't we move on from lines of code being the metric two decades, three decades ago, where it was like first it was lines of code written, and then it was like, oh, that's a fucking horrible metric because more code does not equal better at all.
00:40:04.789 --> 00:40:12.150
It just feels like telling someone they have to burn through X amount of tokens in a week is just why?
00:40:12.469 --> 00:40:12.949
How?
00:40:13.349 --> 00:40:13.989
What?
00:40:14.230 --> 00:40:24.710
Just game it, easy, totally easy to game, stupid simple to game, but just some of the approaches people are taking to this whole thing is weird.
00:40:24.869 --> 00:40:32.150
And then I'll talk to people who aren't in software, and I talk about AI, and they're just like, Yeah, we don't use it.
00:40:32.230 --> 00:40:33.029
Why would we like?
00:40:33.110 --> 00:40:38.869
I asked Match GPT for help with like what should I have for lunch because I'm having an energy dip.
00:40:39.029 --> 00:40:39.429
What a big one.
00:40:39.590 --> 00:40:45.989
We don't use it at work, like we get and I'm just like, oh right, we're in a very, very special world when it comes to AI.
00:40:46.070 --> 00:40:57.190
There's entire industries that are like, AI is still this like novelty little it's the paperclip in Microsoft Word, where it's just like it's there, it doesn't you don't use it for much.
00:40:57.349 --> 00:41:01.029
I wonder when that's gonna change, because that's gonna change soon.
00:41:01.349 --> 00:41:03.509
They gotta hear about the bot army.
00:41:03.750 --> 00:41:04.710
The bot army.
00:41:05.029 --> 00:41:07.110
They hadn't heard about the open claw yet.
00:41:07.269 --> 00:41:08.309
That's the problem.
00:41:08.789 --> 00:41:10.070
Yeah, that's the one.
00:41:10.309 --> 00:41:11.429
Oh boy, yeah.
00:41:11.590 --> 00:41:14.469
That's what I mean when I'm like, it's changing so fast.
00:41:14.630 --> 00:41:34.230
And because it used to be when a new trend popped up and people like ran to catch up, you'd get a lot of little things popping up, but now you get a lot of medium to big size things popping up because the AI can write your random thought faster than you can go.
00:41:34.469 --> 00:41:35.269
Should we?
00:41:35.750 --> 00:41:39.110
Does the world need another Pomodoro timer?
00:41:39.670 --> 00:41:41.190
Doesn't matter, it's getting it.
00:41:41.269 --> 00:41:48.150
And obviously Pomodoro timer is small, but stuff can come out so fast, faster than people can realize.
00:41:48.389 --> 00:41:51.670
Yeah, this isn't a good use of anyone's time.
00:41:51.989 --> 00:41:56.150
It makes it really difficult to anticipate like what is your job gonna look like in five years?
00:41:56.309 --> 00:42:13.029
And like every day I'm like, I don't know, I might not have a job in five years, or I could be doing something completely different in five years, or I'll be doing the same exact thing because LLMs are gonna hit a limit where it's like, yep, they're really good at generating some random code and they still suck at the planning.
00:42:13.509 --> 00:42:16.070
It's the same as being a software engineer forever.
00:42:16.150 --> 00:42:21.509
Like the ones who
are successful are the people who are curious and willing to learn and try new things.
00:42:21.750 --> 00:42:25.269
In five years, I think I'll still be curious and wanting to try new things.
00:42:25.590 --> 00:42:25.829
Right.
00:42:25.909 --> 00:42:35.989
So, like yeah, like whatever the thing is, and if it like for me, I care more about like what I build and the customer than the I enjoy writing code, of course, right?
00:42:36.070 --> 00:42:39.029
But like that's not the thing that's most important to me.
00:42:39.349 --> 00:42:50.070
So if my job evolves into I'm not touching code anymore, but I'm still building things for people, like these cool developer tools.
00:42:50.150 --> 00:42:52.630
Like, I'm I'm still gonna enjoy that a lot.
00:42:52.949 --> 00:42:53.670
For sure.
00:42:53.909 --> 00:42:57.829
So the next segment of the show is when I ask you about blockers.
00:42:58.550 --> 00:43:00.789
So do you have a current blocker?
00:43:00.949 --> 00:43:07.349
Is there something at work that you're currently like scratching your head on, or have you had one recently that you can talk about?
00:43:07.750 --> 00:43:22.389
Literally what you were just talking about, I find very interesting, which is you're talking about Pomodoro timers and like okay, the the idea, the the cost to like get an idea out into the world is nothing, right?
00:43:22.550 --> 00:43:23.590
Yeah, yeah.
00:43:24.230 --> 00:43:40.150
So now I think all of us who are embracing writing code with AI are experiencing this at work because previously, if you had an had an idea, it's just like what you said.
00:43:40.550 --> 00:43:41.989
Does the world need that?
00:43:42.150 --> 00:43:43.750
Does the customer need that?
00:43:43.909 --> 00:43:56.389
Is this like a thing that we want to release and support and all the you know other side effects that come from putting something out in the world?
00:43:56.710 --> 00:44:12.550
And when the barrier is so low to make that happen, I think just a problem I've seen is okay, we're throwing out ideas so quick, we need to then deal with you know, like supporting them.
00:44:12.710 --> 00:44:18.230
Like we're building so fast, but we're not going through the should we as much.
00:44:18.789 --> 00:44:28.469
So there's a lot of like bad ideas, and it's a little more difficult, or it's just like a little more challenging when you go, that's a bad idea, but it was easy to do.
00:44:28.710 --> 00:44:32.949
Before, that's a bad idea, and it's gonna take us two weeks, so let's really not do it.
00:44:33.190 --> 00:44:33.670
Right, right.
00:44:33.750 --> 00:44:33.989
Yeah.
00:44:34.230 --> 00:44:38.230
So those discussions are much more interesting than they used to be.
00:44:38.309 --> 00:44:44.230
And just like calibrating, like I've just felt that just with my coworkers, like we're calibrating about it.
00:44:44.389 --> 00:44:49.269
And like myself should, you know, that would be an easy idea, but maybe I just shouldn't do that.
00:44:49.509
--> 00:44:49.829
Yeah.
00:44:49.989 --> 00:44:51.750
It's tough to tell.
00:44:52.070 --> 00:45:07.989
You used to be able to kick things down the lane a little bit by saying, Oh, that's gonna take a lot of time, so we're not gonna let it creep into our scope, or we don't have good consensus on this, and it's a resource hog, so we're just not gonna do it yet.
00:45:08.230 --> 00:45:14.230
And now it's like, yeah, you know, like you said, you can get anything out there really quickly.
00:45:14.710 --> 00:45:17.190
So, how do you decide what to build now?
00:45:17.509 --> 00:45:26.710
Because there used to be extra criteria on top of like, oh, find your product market fit if you're like trying to come up with a SaaS or your own little app business.
00:45:26.869 --> 00:45:28.710
You got to find that product market fit.
00:45:28.789 --> 00:45:37.829
And it's like maybe you do to like sustain a business, but like to start one now is like, yeah, open up AI and like prompt it.
00:45:37.989 --> 00:45:41.909
It's insane to see how many apps are flooding areas now.
00:45:42.230 --> 00:45:54.309
So you've got that, the Evergreen app, but then also what do we put into our large app that's been, you know, Planet Scale's relatively young at at five, six, whatever years.
00:45:54.550 --> 00:45:57.349
Podia is getting there, it's over a decade old.
00:45:57.509 --> 00:46:00.469
I've worked on apps that were really old.
00:46:00.869 --> 00:46:04.150
Those are not as easy to just cram shit into.
00:46:05.110 --> 00:46:10.389
It's a lot easier with AI now, but it makes the whole system more brittle.
00:46:10.630 --> 00:46:11.829
I'm not at that level.
00:46:11.989 --> 00:46:13.750
I'm not that decision maker.
00:46:13.989 --> 00:46:20.150
I push back on like scope because I don't want to do dumb stuff that they're just going to rip out later.
00:46:20.389 --> 00:46:24.389
I can't fathom what it must be like now to be like, we can build anything.
00:46:24.550 --> 00:46:25.349
What do we build?
00:46:25.590 --> 00:46:31.909
That feels like such a huge leap to decide what gets built.
00:46:32.230 --> 00:46:42.550
One thing I keep like coming back to is really for any team, what I've seen work, and then also what I've seen hurt them.
00:46:42.710 --> 00:46:46.469
What hurts a team is when they have too many things going on.
00:46:46.789 --> 00:46:46.949
Right.
00:46:47.029 --> 00:46:49.110
And they're responsible for too many things at once.
00:46:49.590 --> 00:46:57.989
It's easier today for teams to get into that state than it used to be, because it's so much easier to just start putting things out there.
00:46:58.469 --> 00:47:09.029
Something that our team likes to do is we're really just setting the limit on the number of things that are big and in progress at a time.
00:47:09.349 --> 00:47:10.789
You have six engineers.
00:47:11.029 --> 00:47:13.509
Okay, we're committing to like two things.
00:47:13.750 --> 00:47:16.869
Let's get them out of the way, and then we
can do two more things.
00:47:17.110 --> 00:47:23.190
We're able to get through those two things way faster than we used to, but please let's just still focus on only two things at once.
00:47:23.670 --> 00:47:34.550
Because if you're you know focusing on six or eight, people are swapping context all day, they're unhappy and grumpy, and then you start multiplying that by teams, and it just gets out of control.
00:47:34.789 --> 00:47:43.750
And you're like, hey, I get that you could, but what if we finish that thing we already started, get it done, and then just focus on one thing and your your life's gonna be really nice.
00:47:44.150 --> 00:47:45.269
I feel attacked.
00:47:45.349 --> 00:47:46.789
What do you mean, finish a thing?
00:47:46.949 --> 00:47:49.029
You mean 80% isn't done?
00:47:49.429 --> 00:47:52.150
It's the last 10% that take take the most time.
00:47:52.389 --> 00:47:52.710
Yes.
00:47:52.949 --> 00:47:53.909
That has not changed.
00:47:54.150 --> 00:47:59.269
That's not, yeah, that's nothing different, especially uh AIs get worse as their context window grows.
00:47:59.349 --> 00:48:00.309
So they have the same problem.
00:48:00.389 --> 00:48:09.590
You get 90% of the way done, their context window is 90% full, and they're useless, and you're just like, well, I guess I'm done because the AI can't figure it out anymore.
00:48:09.989 --> 00:48:11.429
They're just like us.
00:48:11.909 --> 00:48:32.150
It is amusing at times to be in a like a planning mode where you're deliberately doing a very back and forth like conversation style with one of the LLMs because uh especially like I think Opus does this the most, it does sort of feel like talking to a person.
00:48:32.389 --> 00:48:33.110
Yeah, it does.
00:48:33.349 --> 00:48:37.829
And I guess that's because they were so heavily trained on Reddit, it sounds like talking to a bunch of Redditors.
00:48:38.150 --> 00:48:39.590
That's actually not a good thing.
00:48:39.989 --> 00:48:47.110
It is funny to like have to occasionally remind yourself like this is not a human being that I'm talking to.
00:48:47.349 --> 00:48:57.670
Once I close this session window, when I open a new session, it will not know anything we just talked about, unless I explicitly say, like, write this down and remember it.
00:48:57.829 --> 00:49:01.989
It's a weird feeling at times to just like remind yourself.
00:49:02.070 --> 00:49:10.550
Like, I think because I spent so much time with growing up, having the internet introduced as I was growing up, right?
00:49:10.630 --> 00:49:22.630
And like having to remind myself at times when I was a kid, teenager, I guess, you're typing into a computer and the words are showing up on a computer, but the person receiving these words is a person.
00:49:22.869 --> 00:49:36.869
Like there's a human being on the other side of this AOL instant messenger or IRC channel reminding myself that now it's the reverse where I'm like, I have to remind myself, like I'm talking to a computer, it doesn't care.
00:49:37.269 --> 00:49:39.349
Do you still find yourself being polite though?
00:49:39.509 --> 00:49:39.750
I do.
00:49:40.389 --> 00:49:41.829
I'm always saying please and thank you.
00:49:42.150 --> 00:49:44.869
Yeah, I always say please and thank you, or like, good job.
00:49:45.029 --> 00:49:49.670
Because you know what when the AI uprising does happen, maybe they will spare me.
00:49:49.829 --> 00:49:50.789
I don't know, probably not.
00:49:50.949 --> 00:49:53.429
But that also that's a good habit, right?
00:49:53.590 --> 00:50:00.869
If I get used to being like you're a fucking idiot to the LOM, I'm gonna bleed into talking to my coworkers, and I don't want that.
00:50:01.269 --> 00:50:02.630
I want to treat them with respect.
00:50:02.710 --> 00:50:05.909
And if I just default to respect, then everything's fine.
00:50:05.989 --> 00:50:07.909
I don't have to remember who I'm talking to.
00:50:08.150 --> 00:50:09.590
But good life rule.
00:50:09.909 --> 00:50:11.029
Yeah, sure.
00:50:11.750 --> 00:50:22.630
So you're using cursor, which I used like when AI coding started to become a thing, but I haven't used it in a while, so now you've motivated me to try and check it out again.
00:50:22.789 --> 00:50:27.190
It's I do use I use open code, so we'll see.
00:50:27.829 --> 00:50:30.949
Both cursor and open code run on planet scale.
00:50:31.429 --> 00:50:32.070
Nice.
00:50:32.550 --> 00:50:33.509
We love it.
00:50:34.150 --> 00:50:35.989
We use the tools and then we help them build the tools.
00:50:36.150 --> 00:50:36.469
Yeah, yeah.
00:50:37.590 --> 00:50:38.150
So cool.
00:50:38.309 --> 00:50:41.670
I remember for a while Planet Scale was just MySQL.
00:50:42.070 --> 00:50:42.469
Yep.
00:50:42.710 --> 00:50:54.789
Which was like a bummer because I was using MySQL and I was so excited to get back to Postgres when I changed jobs and started working at Podia, and then PlanetScale picked up Postgres.
00:50:55.509 --> 00:51:06.070
So was that just a someone was very comfortable with MySQL and decided they wanted to start with MySQL and Postgres later because Postgres is pretty awesome and why not?
00:51:06.309 --> 00:51:17.110
Or was it more deliberate of like, well, there's already managed Postgres hosting, let's do something like MySQL that maybe did have it but not as big.
00:51:17.429 --> 00:51:18.869
Third option, ready for this.
00:51:19.190 --> 00:51:21.829
Or third option, totally unexpected.
00:51:22.150 --> 00:51:25.349
At GitHub, GitHub runs on MySQL.
00:51:25.590 --> 00:51:28.789
So a lot of us that is our expertise.
00:51:29.269 --> 00:51:38.469
And GitHub uses this open source project that was born out of YouTube called Vitesse.
00:51:39.190 --> 00:51:50.949
Vitesse is the scaling layer that YouTube used to scale out MySQL so that it could handle all of the traffic of YouTube.
00:51:52.230 --> 00:51:55.429
The team there open sourced it.
00:51:56.150 --> 00:51:59.509
A bunch of other big companies started
to use it.
00:51:59.909 --> 00:52:10.869
And the test really became one of the best ways for these sort of hyperscaler companies to scale out their relational databases.
00:52:11.429 --> 00:52:15.829
When I was at GitHub, I got to see how this thing works.
00:52:16.630 --> 00:52:19.909
It is so cool.
00:52:22.309 --> 00:52:25.829
There's so many things I could tell you about why it's so cool, but I'll give you one.
00:52:25.989 --> 00:52:26.550
All right.
00:52:27.029 --> 00:52:38.469
So Vitesse allows you to do sharding, which is essentially uh say you have one big table and every company has this.
00:52:38.630 --> 00:52:41.829
I have a big table and it's so big, it's kind of a problem.
00:52:41.989 --> 00:52:44.550
It's annoying to query, it's annoying to backup.
00:52:44.789 --> 00:52:49.110
If I do a schema change on it, it's annoying to run a schema change.
00:52:49.190 --> 00:52:50.309
It's just all pain.
00:52:50.389 --> 00:52:51.909
I think pretty much every company has this.
00:52:51.989 --> 00:52:53.269
You can probably think of one in your head.
00:52:53.349 --> 00:52:54.550
You're like, yeah, easy.
00:52:54.630 --> 00:52:55.269
Yep, I got you.
00:52:55.989 --> 00:52:57.110
Yeah, everyone has that.
00:52:57.349 --> 00:52:59.750
Okay, what if you split that table in half?
00:52:59.909 --> 00:53:01.190
Is it still big?
00:53:01.429 --> 00:53:01.989
Maybe.
00:53:02.150 --> 00:53:03.750
What if you split it into four?
00:53:03.989 --> 00:53:05.190
Is it still big?
00:53:05.509 --> 00:53:08.389
You get to the point where it's no longer a big table.
00:53:08.550 --> 00:53:10.230
You have a bunch of small tables.
00:53:10.389 --> 00:53:14.789
And when you have a bunch of small tables, the big table problems they go away, right?
00:53:14.869 --> 00:53:16.710
Like the backups are easier.
00:53:16.869 --> 00:53:18.550
Often people don't think about this.
00:53:18.789 --> 00:53:21.349
I think about backups all the time because that's what we do.
00:53:21.509 --> 00:53:22.789
We keep people's apps online.
00:53:23.750 --> 00:53:26.949
But yeah, if you have a small table, it's easy to back up and it's easy to restore.
00:53:27.110 --> 00:53:30.869
That's really important if you have downtime or an outage.
00:53:31.190 --> 00:53:43.509
If you have, when you run these really large databases, and this is a thing that people often don't experience, and I get to experience it every day, is hardware fails all the time.
00:53:43.989 --> 00:53:48.789
And if you're running sort of a small app on one node, you're like, oh, hardware doesn't fail.
00:53:49.029 --> 00:53:53.829
But if you're running an app with a thousand nodes, they're failing like every day.
00:53:53.989 --> 00:53:55.269
Like things just go wrong.
00:53:55.590 --> 00:54:07.590
So if your table is now spread over six servers, for example, and one of them fails, well, one sixth of your customers are now upset, but the rest are still going.
00:54:07.750 --> 00:54:07.909
Right.
00:54:07.989 --> 00:54:09.829
So splitting things out is really nice.
00:54:10.070 --> 00:54:16.630
So what the test does is it allows you to pick a sharding key for a table.
00:54:16.869 --> 00:54:23.509
So for example, if like you're thinking of github.com, you know, and you go to github.com slash, and then there's the organization name.
00:54:23.750 --> 00:54:28.550
A great shard key for GitHub is organization, because that's how the data is split up.
00:54:28.789 --> 00:54:36.389
So then what the test does is all of the data for X organization is co-located on one shard.
00:54:36.630 --> 00:54:39.190
The different organization is on a different shard.
00:54:39.429 --> 00:54:49.029
So if you're querying and doing joins for a certain organization, all the data's there on the same machine and it's fast and easy to access.
00:54:49.429 --> 00:54:56.309
I think the thing for I just remember doing this for the first time and being like, this is like the coolest thing I've ever seen in my life.
00:54:56.550 --> 00:54:58.949
Is I was using Vitesse.
00:54:59.190 --> 00:55:02.309
I took a table and I sharded it, right?
00:55:02.469 --> 00:55:05.670
So now it's split up and there's a shard key.
00:55:05.829 --> 00:55:07.829
And then I started writing data to it.
00:55:07.989 --> 00:55:09.989
I was just like insert, whatever.
00:55:10.309 --> 00:55:15.429
Then I logged into the individual servers and I was just doing like select all.
00:55:15.590 --> 00:55:17.110
And I'm like watching it fill up.
00:55:17.349 --> 00:55:20.710
Like, oh, here are half of my records, and here are the other half.
00:55:20.869 --> 00:55:23.829
And I can see them getting distributed over multiple machines.
00:55:23.909 --> 00:55:25.750
Like, this is so sick.
00:55:26.469 --> 00:55:35.670
And then if I just connect to Vitesse, which has something called a VT gate, it's essentially a load balancer for your database.
00:55:35.989 --> 00:55:42.389
And I say, you know, hey Vitesse, give me all the records in the mic organization on GitHub.
00:55:42.550 --> 00:55:47.509
It's going to take that query, bring it to the correct shard where the data is.
00:55:47.909 --> 00:55:53.190
It's like a hash, like it maps to what server the data is on, and it just returns it to you.
00:55:53.349 --> 00:55:59.670
So it looks like a single database, but really I've got 12 or a thousand databases running there.
00:55:59.829 --> 00:56:01.590
This is just one thing Vitesse does.
00:56:01.670 --> 00:56:03.349
There's so many other cool things.
00:56:03.670 --> 00:56:09.909
So to get back to where we were was PlanSkill started as a Vitesse company.
00:56:09.989 --> 00:56:13.829
How can we bring and do hosted Vitess?
00:56:14.630 --> 00:56:16.230
And we did that.
00:56:16.789 --> 00:56:18.949
And it became very successful.
00:56:19.349 --> 00:56:31.750
The thing that we're, and I would say our best at was these operational problems around hosting databases, which is we talked about it a bit.
00:56:31.829 --> 00:56:33.829
Like, I care about backups a lot.
00:56:34.150 --> 00:56:36.630
People don't care about backups, but I care about backups.
00:56:37.029 --> 00:56:38.230
Like restoring stuff.
00:56:38.389 --> 00:56:41.989
And if you've ever dropped your production database, you care about backups.
00:56:42.389 --> 00:56:42.710
Yeah.
00:56:42.869 --> 00:56:45.829
And I'm responsible for tens of thousands of those every day.
00:56:45.989 --> 00:56:46.789
So I care about it.
00:56:47.269 --> 00:56:54.789
Your caring is slightly different than the average Joe's carrying of like, I remember when I dropped production on my one app.
00:56:54.949 --> 00:56:57.269
I now all three of my apps have backups.
00:56:57.429 --> 00:57:02.710
You're you have a backup scaling problem, one might say.
00:57:03.110 --> 00:57:03.750
Yes.
00:57:04.070 --> 00:57:04.389
Yeah.
00:57:04.630 --> 00:57:13.829
So we became really good at keeping people's databases up at huge scale, and then also making them very fast.
00:57:14.630 --> 00:57:19.670
Not only very fast, but we are able to do faster and cheaper than Amazon.
00:57:19.989 --> 00:57:23.110
When you hear that, companies are like, oh, I can bring my database to you.
00:57:23.190 --> 00:57:24.710
It's both faster and cheaper.
00:57:24.869 --> 00:57:25.829
And it's going to stay up.
00:57:25.909 --> 00:57:26.949
Oh, I like all that stuff.
00:57:27.509 --> 00:57:36.389
And I would assume easier to use because anyone who's used AWS for anything knows how convoluted their whole system is.
00:57:36.869 --> 00:57:37.269
Right.
00:57:37.429 --> 00:57:41.349
And our designer is the guy who designed the pull request page.
00:57:41.590 --> 00:57:44.230
And yeah, so our DUI is quite good.
00:57:44.389 --> 00:57:44.630
Yeah.
00:57:44.710 --> 00:57:45.829
So we have that too.
00:57:46.070 --> 00:57:47.269
A bunch of things.
00:57:47.509 --> 00:57:50.469
So we were at this stage where people just kept going.
00:57:50.710 --> 00:57:55.829
The momentum is in startups and young companies is behind Postgres.
00:57:56.230 --> 00:57:58.070
All these companies are starting on Postgres.
00:57:58.550 --> 00:58:04.710
And then they were also all having problems with their existing hosts.
00:58:04.869 --> 00:58:07.750
They're like, we can't scale this thing.
00:58:07.829 --> 00:58:08.869
It's going down.
00:58:09.110 --> 00:58:11.269
We've outgrown AWS.
00:58:11.989 --> 00:58:15.670
And some of them had MySQL databases on us, some of them didn't.
00:58:15.829 --> 00:58:19.110
And they would just keep coming to us and go, Could you please?
00:58:19.269 --> 00:58:20.949
Like, we can't move to MySQL.
00:58:21.029 --> 00:58:23.110
Could you just do a Postgres for us?
00:58:23.349 --> 00:58:24.309
Please.
00:58:24.630 --> 00:58:25.269
For a long time.
00:58:25.429 --> 00:58:28.389
We're like, no, like we were really good at this of the test thing.
00:58:28.630 --> 00:58:33.750
And then um it was last year we talked about it.
00:58:33.829 --> 00:58:36.309
And it was
like, let's try it.
00:58:36.550 --> 00:58:38.710
Let's just see how far we can get.
00:58:38.949 --> 00:58:40.389
I think I mentioned this before.
00:58:40.550 --> 00:58:48.550
Like our company is really unique where it's basically all engineers and they're all like they'd be like the smartest person at some other company.
00:58:48.710 --> 00:58:53.110
It's really, it's a weird, it's a weird place to work because everyone's like universally good.
00:58:53.190 --> 00:58:54.070
It's very strange.
00:58:54.309 --> 00:58:54.550
Yeah.
00:58:54.630 --> 00:59:00.710
So we met up in San Francisco, a good amount of us, and then other people remote.
00:59:00.789 --> 00:59:04.150
And we were just like, all right, let's spend a week and see what happens.
00:59:04.230 --> 00:59:06.309
Like, can we make a Postgres run?
00:59:06.630 --> 00:59:13.509
And it was Tuesday evening, the second day of our offsite, and we had it running.
00:59:14.710 --> 00:59:17.190
Wow, there's uh potential here.
00:59:17.429 --> 00:59:21.989
And by the end of the week, we actually had like an interesting product, so we just kept going.
00:59:22.230 --> 00:59:26.550
Yeah, by the end of that summer, we had like a really good Postgres product.
00:59:27.190 --> 00:59:27.829
Nice.
00:59:28.309 --> 00:59:30.150
I love working remotely.
00:59:30.389 --> 00:59:32.869
I love like the flexibility that offers.
00:59:32.949 --> 00:59:40.469
I love that I have my very unique space to think and like I can walk the dog and do some thinking out there.
00:59:40.550 --> 00:59:46.389
And like I do legitimately think that remote work makes me individually a better engineer.
00:59:46.710 --> 00:59:52.710
But there is something super cool about getting together and putting heads together on a problem.
00:59:52.949 --> 00:59:58.550
How much do you think it was that you were all in the same place and like it was just like a hey, what do you think about this?
00:59:58.710 --> 01:00:03.989
Oh, I don't know, that looks cool, versus like, hey, can you hop on a Zoom real quick to chat about this?
01:00:04.070 --> 01:00:06.150
Or like here, you know, slacking someone.
01:00:06.230 --> 01:00:12.230
How much do you think that made a difference in getting this thing going up so fast?
01:00:12.789 --> 01:00:20.949
I think probably the the biggest benefit was just we were almost just like isolated, and this is the thing you're focusing on.
01:00:21.110 --> 01:00:21.269
Right?
01:00:21.349 --> 01:00:30.070
Because when you're all all still remote, there's just like the other, I could go to another Slack window, and there's that other project I'm working on, but it's just dedicated focus.
01:00:30.230 --> 01:00:32.230
And you could just feel the energy.
01:00:32.630 --> 01:00:40.309
I remember going into that week expecting, oh, we're gonna like grind so hard, writing tons of code to make this happen.
01:00:40.710 --> 01:00:42.710
But my coworkers are so smart.
01:00:42.789 --> 01:00:45.349
I remember it's like, I'm not even trying hard.
01:00:45.429 --> 01:00:48.309
We're just it's just oh it already works.
01:00:48.550 --> 01:00:49.590
Thank you, everyone.
01:00:49.670 --> 01:00:51.029
You're so smart.
01:00:52.550 --> 01:00:53.509
That's awesome.
01:00:53.670 --> 01:00:54.949
That is so cool.
01:00:55.269 --> 01:01:08.309
And it's working like feature parity-wise now, or are there still areas that need to improve to get the Postgres side of the business as awesome as the MySQL side?
01:01:08.630 --> 01:01:19.909
The fun part is for me, it was well, I had already done a database SaaS because we launched it for the tests, and then here's my second experience.
01:01:20.070 --> 01:01:22.949
I get to write another database SaaS for Postgres.
01:01:23.269 --> 01:01:27.029
Yeah, and I was just like, wow, I'm so much smarter than five years ago.
01:01:27.190 --> 01:01:28.789
Like it's so much easier.
01:01:29.029 --> 01:01:34.630
The thing about our company is no one wants to leave because like we're all having so much fun together.
01:01:34.789 --> 01:01:35.110
Yeah, yeah.
01:01:35.349 --> 01:01:44.630
So it's like essentially a lot of the same people, and we've all had this shared experience of building it on like building a MySQL product.
01:01:44.789 --> 01:01:49.110
So then us building Postgres was we just had all those learnings.
01:01:49.349 --> 01:01:53.269
Like there's a lot that's different about the communities and like what people want.
01:01:53.509 --> 01:02:06.469
So we didn't go out to build the same product, but we like took the things that we learned from it and then just directed that at what do Postgres users really want, which is different.
01:02:06.550 --> 01:02:08.550
And then also, like, what are the challenges of Postgres?
01:02:08.710 --> 01:02:13.190
For example, um managing connections on Postgres, that's a whole thing, right?
01:02:13.269 --> 01:02:17.029
Anyone who runs Postgres goes, I know everything about connections because you have to.
01:02:17.269 --> 01:02:19.909
For Vitesse users, that's just handled for you.
01:02:20.070 --> 01:02:23.029
It's not even like like 10,000 connections, who cares?
01:02:23.110 --> 01:02:25.190
That's you know, it just works, right?
01:02:25.590 --> 01:02:25.989
Vitesse.
01:02:26.469 --> 01:02:27.190
That's super powerful.
01:02:27.590 --> 01:02:27.750
Sure.
01:02:27.829 --> 01:02:30.230
Yeah, tell me, Vitesse is very nice.
01:02:30.469 --> 01:02:40.550
Yeah, so like for Vitesse, we didn't need to spend that much time on worrying about how do people tune their PG bouncers or their max connection pools or any of that.
01:02:40.789 --> 01:02:47.269
For Postgres, that's been months of thought and effort to make that experience good.
01:02:47.509 --> 01:02:50.150
So it's different.
01:02:50.869 --> 01:03:04.070
The thing that's coming up that I'm most excited about is we have the team that works on Vitesse has been working on building a Vitesse for Postgres.
01:03:04.789 --> 01:03:08.550
So like that
cool stuff we talked about that's really only Vitesse only.
01:03:09.190 --> 01:03:15.590
We have those same people, and they're getting to do their second time, except they're applying it now to Postgres.
01:03:16.550 --> 01:03:20.070
So that cool stuff I talked about with sharding.
01:03:20.309 --> 01:03:22.710
Oh yeah, it's like we'll bring it to Postgres.
01:03:23.029 --> 01:03:25.029
There's a lot of like companies who need that.
01:03:25.269 --> 01:03:27.190
Yeah, that's gonna be fun when we ship it.
01:03:27.429 --> 01:03:34.230
Yeah, that sounds so cool, especially that's being able to use Postgres at scale because that was that's kind of always been the argument.
01:03:34.309 --> 01:03:39.750
Like MySQL is faster, it handles scale better than Postgres does.
01:03:39.909 --> 01:03:50.230
Well, I guess it's been an argument, but to kind of solve that, and it's like, yeah, you just use whatever you want, we'll handle making sure that you can scale with either.
01:03:50.389 --> 01:03:51.829
That's pretty freaking awesome.
01:03:52.070 --> 01:03:58.630
Yeah, I feel like I'm always embedded in the arguments of like Postgres or MySQL, and now running them both.
01:03:58.949 --> 01:04:08.309
My experience is, and I've run them both in production for years across different companies, and Postgres for the developer.
01:04:08.389 --> 01:04:15.670
So, like a person like you and I, who's writing code for an app, like a Rails app and using Postgres, the experience is like so good.
01:04:15.829 --> 01:04:19.989
Just the types are cool, the extensions, it's just yeah, it's great.
01:04:20.070 --> 01:04:24.710
It's like a great developer's database, and that's why you see people so excited about it.
01:04:24.869 --> 01:04:37.349
And then when you hear things where people say, My SQL is maybe faster or better to scale, it's not always necessarily true in ways that you'd expect.
01:04:37.590 --> 01:04:41.110
The ways that it's easier is it's easier to operate.
01:04:41.509 --> 01:04:45.190
And the average developer doesn't know about this or care about it.
01:04:45.269 --> 01:04:50.869
They don't think about like how ease we talked about this before, how easy is it to backup and restore my database?
01:04:51.029 --> 01:04:51.349
Yeah, yeah.
01:04:51.590 --> 01:04:57.909
Well, the reason why it's easier on MySQL is because there's all these old giant companies who have built tons of tooling around it.
01:04:58.070 --> 01:05:02.150
And there's just been a lot of like hours put into making it good.
01:05:02.309 --> 01:05:09.429
And then Postgres just doesn't have the same amount of big companies who have put in the same amount of time.
01:05:09.829 --> 01:05:11.989
So that's an area where it's weaker.
01:05:12.070 --> 01:05:14.150
And like we run now, we run them both.
01:05:14.309 --> 01:05:15.509
That's just something we found.
01:05:15.590 --> 01:05:20.070
Okay, it's it's more difficult to operate this than it was MySQL.
01:05:20.309 --> 01:05:33.269
And the thing that I think is going to be really great for Postgres with our company focusing on this, is we have a bunch of people who have seen various different databases operating at huge scale.
01:05:33.509 --> 01:05:39.829
And now also we are incentivized to make running Postgres at huge scale really good.
01:05:40.309 --> 01:05:40.630
Nice.
01:05:40.789 --> 01:05:46.550
So we can take those things that we learned from like running MySQL at huge scale to apply it to Postgres.
01:05:47.029 --> 01:05:48.389
That's going to be like the fun thing.
01:05:48.469 --> 01:06:00.389
It's not just, oh, we're hosting a database for you, but like how can we make it better and bring over this these places that are a bit of weaknesses for Postgres and we can shore them up and make them better.
01:06:00.710 --> 01:06:19.029
That's so cool, especially to be able to close that gap and give you some of that the same you can choose your own tool and expect to get the same benefits of each tool, at least from a backup user developer, more than user usage way.
01:06:19.750 --> 01:06:20.789
That is super cool.
01:06:21.029 --> 01:06:29.509
So the last question: what is something other than what you just talked about, which is super cool and arguably new and interesting?
01:06:29.670 --> 01:06:32.869
But what is something cool, new, or interesting you want to share?
01:06:33.269 --> 01:06:35.349
Have you used spokenly?
01:06:36.230 --> 01:06:37.829
Spokenly, no, no.
01:06:38.070 --> 01:06:39.429
Have you heard of Whisper?
01:06:39.909 --> 01:06:41.110
Yeah, I use Whisper.
01:06:41.190 --> 01:06:41.349
Yep.
01:06:41.509 --> 01:06:42.869
Oh, you use Whisper, okay.
01:06:43.190 --> 01:06:47.429
Some of my coworkers have been saying that they just talk to their computers all day.
01:06:47.590 --> 01:06:50.550
And I was like, guys, well weird.
01:06:51.750 --> 01:06:52.710
Then I tried it.
01:06:53.110 --> 01:06:55.909
Then I tried it, and I really liked it.
01:06:56.230 --> 01:06:58.550
So I tried Whisper at first.
01:06:58.630 --> 01:06:59.750
I felt like it was a little slow.
01:06:59.909 --> 01:07:01.190
Like I didn't like the delay.
01:07:01.429 --> 01:07:11.190
So then I found this app called Spokenly, which you should totally try because it can actually run a local model on your laptop.
01:07:11.509 --> 01:07:11.670
Okay.
01:07:12.070 --> 01:07:13.110
So it's quicker.
01:07:13.269 --> 01:07:16.710
I don't know if the results are as good, but they seem really good.
01:07:17.190 --> 01:07:19.190
But it's basically instant.
01:07:19.429 --> 01:07:20.789
So I've been using that.
01:07:20.949 --> 01:07:23.829
I know they have a Mac app, I think they have an iPhone app.
01:07:23.989 --> 01:07:25.269
I haven't gone that far.
01:07:25.509 --> 01:07:31.670
But yeah, I'm just all day like dictating to Slack and to cursor and all this.
01:07:31.829 --> 01:07:32.789
And I'm like, oh, it's so easy.
01:07:32.869 --> 01:07:34.150
I don't have to type.
01:07:34.469 --> 01:07:34.869
Yeah.
01:07:35.110 --> 01:07:38.630
I've historically not been great at typing.
01:07:38.789 --> 01:07:42.789
It's always been like I see people using the split keyboards, and I'm like, oh, that's so cool.
01:07:42.869 --> 01:07:43.590
I wish I could do that.
01:07:43.670 --> 01:07:46.869
But I'm like, these two fingers on each hand do the typing.
01:07:46.949 --> 01:07:48.949
The other ones just kind of sit there and do nothing.
01:07:49.110 --> 01:07:50.869
It's just not a good typer.
01:07:51.110 --> 01:07:53.349
So like anytime I'm like, oh, cool.
01:07:53.429 --> 01:07:56.469
And my brain goes way faster than my hands can keep up.
01:07:56.630 --> 01:07:56.869
Yep.
01:07:57.110 --> 01:08:01.269
Not necessarily in a good way, but it is so nice to be able to like dictate.
01:08:01.509 --> 01:08:03.590
But yeah, Whisper works.
01:08:03.829 --> 01:08:04.550
You might like something.
01:08:04.869 --> 01:08:05.670
I don't mean yeah.
01:08:05.750 --> 01:08:06.789
I'm gonna give it a try.
01:08:06.869 --> 01:08:07.110
Yeah.
01:08:07.349 --> 01:08:08.710
There is that delay.
01:08:09.029 --> 01:08:14.309
And it is the one thing like I use Android as my phone.
01:08:14.550 --> 01:08:21.269
And when I'm dictating to my phone, I see what I'm saying moments after I say it.
01:08:21.349 --> 01:08:21.590
Yeah.
01:08:21.750 --> 01:08:27.909
But with Whisper, I have to like stop the dictation for it to paste in what I said.
01:08:28.309 --> 01:08:31.509
And again, we go back to like my brain is going faster than my hands can move.
01:08:31.590 --> 01:08:35.429
My brain is going faster than my brain remembers that I just said something.
01:08:35.590 --> 01:08:39.590
So like I'll say the same thing three times within the same paragraph.
01:08:39.670 --> 01:08:43.509
And it's like, no, you idiot, stop talking and you'll realize this.
01:08:43.670 --> 01:08:47.109
So dictation's great for like getting stuff out of my head really fast.
01:08:47.189 --> 01:08:50.789
It's so bad when I have to go back and read it, edit it.
01:08:51.189 --> 01:08:51.909
Oh no.
01:08:52.149 --> 01:09:00.630
So I don't know if spokenly will solve that issue, but if spokenly can show me what I'm saying even moments after I'm saying it, it might save me some headache.
01:09:00.789 --> 01:09:01.989
I will be giving that a try.
01:09:02.150 --> 01:09:02.949
So that's a good one.
01:09:03.110 --> 01:09:03.670
I like that.
01:09:03.989 --> 01:09:06.310
It might save you, you know, a thousand milliseconds here.
01:09:06.630 --> 01:09:07.670
Just the trip over the internet.
01:09:07.909 --> 01:09:08.230
Yeah.
01:09:08.390 --> 01:09:13.590
I also like the idea of like my voice not going to some company on the internet all the time.
01:09:13.750 --> 01:09:14.150
Yeah.
01:09:14.310 --> 01:09:15.270
I like the idea.
01:09:15.590 --> 01:09:26.470
Like we talked about, where I think there's a version of the future where our LLM usage is potentially local first or has a massive local component.
01:09:26.630 --> 01:09:39.510
I like that idea too, because then I'm not worried about what information is going in and which one of these billionaire run tech companies are getting my particular data, I say, as a Google user.
01:09:39.829 --> 01:09:41.029
But such is life.
01:09:41.270 --> 01:09:44.550
If Linux proper ever comes out with a phone, I'll use that.
01:09:44.789 --> 01:09:50.470
But for now I've resigned that Google's gonna know more about me than anyone ever should, but it's fine.
01:09:50.710 --> 01:09:59.989
I think that it is cool to be able to keep things local, especially when it comes to something like this, where it's like, this is me talking.
01:10:00.150 --> 01:10:03.430
This isn't even just, hey, look at my works code base.
01:10:03.590 --> 01:10:05.750
Like, oh, that'd be nice to keep on-prem.
01:10:05.909 --> 01:10:11.270
Not that I think that they would ever do anything with like Podia's code base, but still.
01:10:11.590 --> 01:10:12.470
Yeah, cool.
01:10:12.630 --> 01:10:14.470
Good answer accepted.
01:10:14.630 --> 01:10:17.110
I dig it, I'm gonna be checking it out for sure.
01:10:17.430 --> 01:10:20.310
Well, thank you, Mike, for coming on.
01:10:20.390 --> 01:10:26.710
I really appreciate you giving up some of your time on Friday to come and talk about all the cool things that you're working on at Planet Scale.
01:10:27.029 --> 01:10:29.829
And you guys are definitely working on some cool stuff.
01:10:30.149 --> 01:10:32.869
You got me a little like excited to see what you guys do next.
01:10:32.949 --> 01:10:35.430
So yeah, nice.
01:10:35.670 --> 01:10:38.710
Yeah, it's always fun for me to talk about it because it's five years in.
01:10:38.869 --> 01:10:41.909
I'm still excited every day, and I'm proud of what we built.
01:10:41.989 --> 01:10:42.310
Yeah.
01:10:42.550 --> 01:10:43.750
Yeah, very cool.
01:10:44.069 --> 01:10:48.789
Where can people find you on the internet if they want to follow what you're working on a little more closely?
01:10:49.189 --> 01:10:56.550
On Twitter, although it's called XMS 4Cs, M S C C C C.
01:10:57.029 --> 01:10:58.149
That's where I am.
01:10:58.390 --> 01:11:03.750
Visit planetscale.com if you want to see what I spend all my waking hours carrying off.
01:11:04.149 --> 01:11:04.310
Okay.
01:11:04.550 --> 01:11:05.750
Twitter or Planet Scale.
01:11:05.909 --> 01:11:06.229
Got it.
01:11:06.310 --> 01:11:06.789
Yeah, cool.
01:11:06.869 --> 01:11:08.149
Well, again, thanks, man.
01:11:08.229 --> 01:11:10.229
I do really appreciate the time.
01:11:10.390 --> 01:11:15.430
And you're welcome back anytime you have something cool going on at Planet Scale that you want to talk about.
01:11:15.750 --> 01:11:19.109
Maybe uh the bot army v2 when that comes out.
01:11:19.989 --> 01:11:23.269
Yeah, bot army v2, army of botanists.
01:11:23.510 --> 01:11:23.909
Cool.
01:11:24.069 --> 01:11:24.390
All right.
01:11:24.550 --> 01:11:27.189
Listeners, I will see you in the next one.
01:11:27.430 --> 01:11:28.069
Bye.