Do you have something that actually moves the needle? If you can really convince yourself that's the case, the second question is, well, why on earth doesn't anyone recognize that? Right? And that's also my advice if you're like, you're considering making content, if you're considering doing DevRel, like, make sure you're excited about the thing first. Because once you are, the rest just becomes a whole lot easier.

I'm joined today by Vincent from Marimo. Marimo is a next generation Python notebook and this episode is all about YouTube and how to grow your YouTube channel as a DevTool. Enjoy. Maybe before we get into what you've been doing on YouTube, you could just share with people a little bit about like the results like that you've had and how long you've been doing it and stuff like that so that they get an idea.

Yeah. So I mean, I think when we first spoke, I worked at Marimo for six months and we had half a million subscribers. Oh, sorry. Half sorry. Not subscribers.

Wait. Views. Very different number. So we had half a million views in half a year. I'm now here at, like, month seven, and we just reached, I think, six fifty k.

The way we kinda looked at it is it'd be nice to just have a good foothold on YouTube simply because it's a, like, the second biggest search engine out there these days. And two, some people really prefer to read, like get the documentation, that's how they learn. And other people actually prefer video these days. So Mhmm. For a lot of reasons, we figured, okay, there's a lot of good reasons to like invest in YouTube.

And the thing that really caused a lot of these views to happen is the Marimo team is quite small, but we make, two to three releases a week ish. Like, we really do make a lot of software releases. So there's not necessarily a shortage of little announcements that I can make during the week. That's like one of the things we really learned with these YouTube Shorts. If you make like five demo able but really small changes, that actually does quite well on YouTube Shorts.

So one of the reasons why we actually get so many views is mainly because we started doing YouTube Shorts. And every single YouTube short, I think we like, the minimum we get is, like, 2,000 views per video ish. You know, we have to be a little bit careful what the view is on YouTube. So if a video starts playing, that's already a view on a short. And if it starts replaying, like it starts at the beginning again, it gets counted twice.

So there are like, be careful with interpreting the number there. Like, number of views does not mean number of viewers. But even so, like, you can also look at the stats and say, oh, there's like 40 to 50% of the views are from people who've never watched anything from our channel before. So we can also pretty legitimately claim that we do reach like thousands of people every week with just the shorts, basically.

Yeah. When sorry. When you said features, could you give an example of like what one of the shorts might be?

Let me just go to our YouTube oh, actually, I know one of them. We we support python3.14 right now. It's like a thing you can announce. And then you can show a feature of Python three fourteen, and that works in a Marimo notebook, download the latest version, done. Another thing

Would people be how why would just like if it's because that would strike me as like, that's something that like you guys care about. And if someone uses Marimo, then they would care about the like

Yeah. Yeah. So that's not so that that would not be a video that would do amazing on the new viewer metric. Like I would just that that would definitely be like a typical video of like, we have a 5,000 subscribers at this point. Let's say that those are more of the hardcore Marimo users.

So those users are particularly interested in in learning about the new features, and, like, that's why we make some of those shorts for. Interesting. Something that did very well, this one got, like, 17,000 views. It's a bit of a silly example, but I made a two d slider this one time that works in Marimo. So for context, like, Marimo is a Python notebook, but we do have this very convenient mechanism to just add custom widgets.

But I have this one video that's, like, kinda silly. It's like, hey. If you wanna have if you ever wanna use a two d slider inside of a Python notebook, now you can. Look. It's a two d slider.

It works. Great. Another thing I made a while ago is I have these game pads. You can also use these game pads via the game pad API in the browser to communicate with Python. So you can actually trigger Python functions with a game pad now.

It was just like one of these, like, weird little demos. And you can imagine that those things do really well also to new people. It's not necessarily enough for someone to download Marimo right now. Right? But what we hope is that, of course, if you see one of those videos, you might not mind seeing video number two.

And after we're seeing 10 of these videos, at some point, you do kinda go like, gee, these people release a lot of software, and they have interesting features. I'm a Python person. I would like to give this a spin. But that's kind of what you hope for, of course. But it it it remains YouTube.

So it's it's also not necessarily a science. It's also just throw spaghetti at the wall and see what sticks. And it turns out that if you really release a lot of good software and you just make some videos of that, you know, not every video is a slam dunk, but once a week, if you make, like, five to seven videos a week. Some of them do really seem to stick.

Scaling DevTools is sponsored by WorkOS. WorkOS helps you get enterprise ready. That means they give you all the things that you need to start working with enterprises. Things like audit trails, skin provisioning, role based access control, and single sign on. Let's hear from Utpal from digger.dev, a dev tool using WorkOS.

We haven't had to think about OTH at all. I think support is great. We have a Slack connect channel with them. Issues, if any, there haven't been many. But anytime there have been issues, it's been addressed super quickly.

So Odd Trail, SSO, stuff like that, we don't think about that anymore. Generally, we don't think about that anymore. And you don't necessarily think about these enterprise features, but they still lead revenue. And it kinda is a no brainer in that sense.

WorkOS helps your dev tools start selling to enterprises much faster. And they're trusted by dev tools like Cursor, Fowl, and Vercel. If you use their user management off, you can get your first million monthly active users completely free. Do you care about subscribers much?

So the what you're getting at here is also like a question we I'm not really grappling with, but like, okay, we invest in YouTube. Right? Like, what's the number? What what number tells us we're doing good thing? Because we invest, we wanna see number go up.

And, you know, we we like it if we see subscribers because it's some sort of a proxy that a new person is more interested in Marimo than before. Like, that that that feels good. We also have that view count number of like, okay, this video got a lot of views. And then in hindsight, can we also confirm that those were new people? Okay.

Then this video did good. In the end, though, like, we have this questionnaire that pops up in a bunch of places, like when you join our Discord or there's even, like if you have feedback, we have a feedback form that you can leave us. The main thing that we just saw is that as time went on, we have this one question of, hey. How did you learn about Marimo? And some people just said one of Vincent's videos.

Really? Apparently, some of them know me on a first name basis, which is weird. But yeah. Like and the fact that, like, it's not really quantitative, but the fact that on a gut level, it does feel like roughly one in five people that fill in such a questionnaire do mention that they discovered us via something on YouTube. I mean, that does give us the impression that we're doing something right.

And like, we do get more of the feedback from the, let's say, not necessarily hardcore, but like people that have been using Remo for a few months that they like the Shorts because they also learn about the new features and that's sort of their being updated feed. And other people like, a thing that we really do is we try to make sure, okay, if you're an Airflow user, let's say, like, what can Remo do for you? And then I hear some sounds from the Airflow community of, like, oh, that's really sweet. I might wanna give a talk about that at AirflowConf. It's not called AirflowConf, but to to sort of paint the picture.

And similarly, there's, like, Django people, and then they say, oh, we have a Django conference. I might wanna talk at my meetup about this. There's, a little pattern I wanna repeat. So we we look at the stats, you know, every week, we do a gut check and that's, like, good to do. But the real impact it makes on me as a person is more just the qualitative aspect of it, like hearing stories of people discovering us via video.

Like, that's that's also stuff that feels meaningful, I would say.

Yeah. Yeah. It's a that that seems like the thing to kind of like focus on, especially when you're getting started and you're getting like a kind of like there's growth there as well. Right? Because I guess whatever you're getting now, if it's continuing to grow with the same inputs then presumably, you know, however many people are hearing about Marimo from YouTube now, in twelve months, there's probably gonna be a lot more with the same effort.

I have to be a little bit careful with the statistic because it does vary a little bit. But if I'm not mistaken, I think on average, we the download numbers grow between 56% a week at this point. And I we should do a check maybe if, like, that percentage was way lower before I joined, but I can at least claim that I I think we did do a check, but I think we we should redo the number. But something along those lines, like, we are definitely in the growth up. But one thing that's also just really cool to see is it's not just me making YouTube videos about Marimo at this point.

There's, like, other people, other, like, techie influencers that also, like, do a mention of Marimo here and there. So but that that's, of course, kind of what you do when you're doing DevRel. You try to spread good ideas, and hopefully, replicate them. And we're right now in that, like, really interesting phase where we see the the exponential sort of thing going up. People are asking questions about us more and more.

There's, like, companies now writing blog posts about how they're transitioning to more email for their back end stack and stuff like that. We got, like, four blog posts of we just hired someone who's doing more gross stuff, like, little bit more formal marketing. Mhmm. So the blog is also getting, like, more attention now. But yeah.

It's it's really cool to see that with a small team, you can still make a big oomph impact. But I think in the end, like, I can make as many videos as I like. The real blessing is that we just have a really nice velocity schedule that just makes a lot of these things so much easier.

More stuff to talk about and people are excited because it's yeah. Gonna be

Yeah. That that really does help. But the one thing I will say, when we started the YouTube thing, like, we had a bit of a tactic where we said, okay, education might be the first thing to pay attention to. Because there's all these things that like because like, if you're used to because a lot of people compare us to Jupyter, that's like an alternative Python notebook. If you're used to Jupyter, then Marimo is new, and there's, a lot of Jupyter habits that people have to unlearn in order to really appreciate Marimo.

So, like, the first month or two, maybe three, most of our videos were really just education education. Like existing features double down on that basically. And only after like four months or so did we give those shorts a spin, and boy, did we learn a lesson when we did. Really?

In hindsight, would you have done the educational ones?

I still think do all doing the educational ones is still a good idea. And you can also kinda notice that sometimes you have an educational video that hits a couple of nice keywords. Those will be videos that keep on getting traffic. Because very often there's like a curve on YouTube for a long term like, also, shorts have that even more strongly. Like, it goes up, and then it becomes flat.

Usually stops.

Yes. And like, one thing I've I've learned from my private channel, so I people who are watching, like, on display can see. I I like these weird ergonomic keyboards, and I'm one of the few people online that reviews them. So if you type in the name of this weird keyboard, odds are you're gonna see one of my videos on top. So review videos like this one, if you're the one guy reviewing them, oh, they're gonna get traffic, like, continuously.

So the the keyboard videos that I make, they typically just have a line straight up, and I just keep on getting views as time moves on. Some of the Marimo videos also have that if it's a very popular search term because Marimo isn't like Marimo Python and then like very specific Marimo feature. That doesn't get a whole lot of traffic. But when you have 20 videos that each get like 50 views a day, that does start to trickle in. And a couple of our videos have that property where it's a straight line up.

It's just most of them, they do flatten out after a bit. But the funny thing about those educational videos is they have a fairly high percentage, relatively speaking, of videos that do keep on attracting revenue of revenue attention because it's just a popular search term.

Do you look at search terms before you make videos? Or is it like kind of chance?

I so I do think about the YouTube thumbnail and title because you have to on YouTube these days. And you you know, you do your best to grab attention, but not be too click baity because there's too much of that already. You try to make a thumbnail that looks as if effort was put in, but also doesn't overpromise to anything ridiculous. Right? So to that extent, the the the main source of traffic on the short term, like when you just release a video, the SEO is only something that's gonna help you long term.

Like if the video has been out for a couple of months and if you have a popular search term, then you're gonna get views on it. Everything in the first week, which is 90% of the view count normally, that's all YouTube algorithm recommender stuff, basically.

So do you you would focus far more on the algorithm side than the search side, typically?

So so that that is what I do nowadays. But I have, like, in the when I just joined Marimo, I mean, we got to video number 50 really quick. So like, we really did a good sprint on like, okay, there's like all this video stuff missing, and people keep on asking us all these questions. We just wanna be able to send them a video real quick. Yeah.

So I think in the first two months, we had like 20 videos with like all the explainers that we're missing, and that was our our nice little backlog. And that allowed us to say like, okay, let's now do more topical things, more stuff that's like very popular library now. Let's see if we can make a video in the remote for that. But it's so I wouldn't ignore good educational content. That might be the thing I would still tackle first.

It's just that once you've got the frequently asked questions covered, okay, then you gotta do something else.

So it's almost like product. Like the first part is like product almost that you just kind of an extension of your docs, so you just wanna be able to send people.

Part of the story there was, okay, we get a lot of questions. It is on the docs, but people still somehow don't find it. Maybe if we have some videos, maybe less people will ask the question on the forum because it's also on YouTube, but maybe that's also a great way for people to learn about it. Of course, like after two months, those questions were pretty much answered. So then you try to move on to something else.

We try doing some topical things. That was a like, it did work really well a couple of times. But since then, we've really just figured, well, for now, let's just pick some specific communities and figure out if we can offer something special. So just to give a example, a video I made a while ago. I'm I'm also a Django developer.

Like, I used to do a lot of Django back in the day. It's like a Python web framework. Django comes with an admin interface that's pretty nice. But there are just some of these moments where, oh, I would love this admin interface to also have a slider. And it's really easy to build that in Marimo.

It's just not something as easy to configure with the Django admin interface. So, okay, I make a tutorial on how you can get Marimo integration to work with Django, and then some of the Django people can maybe could take inspiration from it. And that's, like, kind of like trying to get a nerd snipe to a specific community working. That's, like, a way you could describe the tactic. But that turns out to work very well because then the people within the Django community, some of them really like it.

They they start giving us feedback. And then I try to repeat that, but then maybe for the Airflow community, and it's definitely a different way to maybe approach growth. But it's something that I'm able to do because we have all those frequently asked questions already sort of done, basically. So the the the framework we've got now is we try to do, like, one main video every week where we try to, like, do good education for a specific community. On top of that, we do all sorts of YouTube shorts.

And if a quick topic appears that we feel we gotta jump on because it's a new, I don't know, LM thing that we have a specific opinion about, then we also try to allow some space for that. But I also still write a little bit of software, so it's not all just YouTube videos, but that's the main the main way we approach YouTube at least.

Yeah. Yeah. Yeah. So could you maybe like dig in a little bit more to like let's say like, either last week's videos or next week's videos. How did you decide specifically to pick the topics?

Just because I I don't know if I'm over indexed on this, but I feel like kind of picking topics is quite a hard thing to do on videos.

I never really had that problem. But the the maybe like a thing that also helps a bit. Like, if if you've seen one of my videos, you know what I'm talking about. When I make a video, nothing is scripted. Only the intro is scripted.

Everything else is just me life coding. And the reason why that works so well is because I have a drawing tablet. So I have a a pen and, like, a drawing the cheapest drawing tablet I was able to buy on, like, Dutch Amazon. It's $40. And you it's a Yuleon tablet.

It is a really good device. Like, I do recommend it to people. But there's an app on Mac called Screen Brush, and I just hit a shortcut, and then I'm able to draw over any code that I've got on display. So a lot of it just me, like, okay, writing something in a in a notebook and, okay, this kinda works now. Okay.

Go into record mode, look into the camera. Okay. I just coded this. This is what this does. Arrow, this is what this does.

So it's a lot of it is really just, okay, I I pick a topic and then I, of course, make sure the code kinda works beforehand, but everything else from there is just me life coding.

Interesting. So, yes. Because I feel like a lot of the time people go and write a tutorial, although go build like an example app. Maybe like build an example app, then write the tutorial on how they built that, and then go record it as like a secondary thing. I I mean, I know everyone has their different methods, but

I will do my prep, but Mhmm. But then, I guess you're talking about like maybe JavaScript people, I suppose, that sort of build a web app and then like do some sort of API integration.

Yeah. It's not like

yeah. So you can imagine that for a Python person I mean, I can I can integrate something in a notebook, but typically, a thing you do more in a Python notebook is like analytics? So, you know, I could find maybe a fun dataset if I'm if somehow I'm exposed to a fun dataset on socials, then that could lead as an inspiration to, like, maybe make a fun video about it. But in my case, it's a lot easier to just sort of say, well, what tools exist in the ecosystem? And, you know, everyone uses a Jupyter Notebook if they're doing anything with data in Python.

So there's so much, like, land for me to just pitch a tent and do something. Like, one thing that might be good to mention here is, like, the Jupyter Notebook, in the end is not a Python file. So it's this different different file format. The the Git story is, like, really, really different, and it's kind of meant to be a multilingual thing, which is also why the Python integration is not really direct. Remo in the end is a notebook, but it's also still a Python file.

So suddenly, we have a Python thing that is a notebook, but it's definitely still a Python file, which means that you can rethink the way a notebook is supposed to work. So suddenly, if you're a senior, like, Django dev, oh, maybe now I can use a notebook because, oh, it's just a Python file now. That's no longer weird in CI. We can maybe consider that. So basically, it's it's almost to me, it feels like every single package out there in Python, I can make a story around.

It's surprisingly easy given the topic.

And you'll have an interesting take on it because people are seeing it in a notebook rather than just

Well well, it's also well, it's

a It's more than that.

To me personally at least, and like the the best way to explain this is, the the Basecamp people had a quote about this a while ago. Like, marketing is basically the exchange of enthusiasm. Like, I've been using notebooks for my entire professional career, and I'm and I was, like, super excited when I saw Marimo. So it's also the thing that I'm, like, just genuinely excited about this whole I can rethink the way I write notebooks aspect of it. I honestly write command line apps in notebooks now, and there's good reasons to consider it.

So, like, I'm it really does help that I'm genuinely excited about it, and I'm also genuinely making all these widgets to see if they can they can work. So and that's also my advice if you're, like, you're considering making content, if you're considering doing DevRel, like, sure you're excited about the thing first. Because once you are, the rest just becomes a whole lot easier.

I think Swix wrote an article about this recently or like,

it was Swix in writes a lot of articles. It's my intention.

He's a machine.

But but he did he's writing about the, oh, maybe DevRel is back. And then Yes. Exactly. And on on the front, I think I I think I saw the same article. There was also this idea of like, don't as a company, you shouldn't look for a DevRel.

You should look for your DevRel. You need to find someone who's like technical and like can do the explaining thing, but is also like super excited. Like if that ingredient is missing, it's gonna it's not gonna work out, I fear.

Yeah. And I think he I think the distinction there was like, don't don't hire someone that's passionate about Python. Hire someone that's passionate about writing Python in notebooks. Or like

Yeah. Or or or whatever the company is doing. Just make sure that like the company is doing something someone is genuinely excited about. And I did kind of a test, that's usually what I do when I see a new tool. Like, if I expose myself to a tool and then like a week later I find myself making plugins for it, that's usually a sign.

And and with Marimo, I ended up making like three plugins in the first month, All of which were pretty substantial too that made it to like their gallery. So that was like my sign of like, okay, Vincent, something is happening here. You should take this maybe serious.

I definitely noticed a lot about that of like, yeah, just lean into what you're interested in and I don't know. I mean, that's the only reason scamming DevTools still exists, I think. It's just because if you're interested in it, you'd probably be making doing similar things if whether or not you are being paid to

I I like the idea of like, you should probably if you wanna learn, you gotta expose yourself to new things. But you should also pay attention if you then feel some sort of an attraction. Like, oh, this this is a cool thing. I wanna keep doing it. Like, okay, listen to that voice and then just chase it.

Let's shoot. Like, I've I've also I it's it's a weird thing to admit, but like, I've never gone through a normal recruitment pipeline in my life. Like, I've I've always been hired because of my blog or some video I made. Like that that's always been the way I've gotten hired in like every professional capacity so far.

Yeah. I mean, it's not surprising. I feel like DevRel maybe is like kind of more like that because it is quite public as well.

Yeah. I would say so. I mean, your resume is public. Yeah. Basically.

So it's it's kinda like, oh, what have you done, Vincent? Well, there's these these are three companies I worked at. These are three YouTube channels that I was you you can see how I've developed there. They're definitely better now. Like, those are things that you can just say by just pointing to a few things that are public.

Yeah. How are your videos different now than they were six months ago?

So I started making videos professionally, I suppose, during COVID. And the interesting thing during COVID was there was a lot of demand and not a lot of supply yet. So if you just started during COVID and your videos are just kind of okay, as in reasonable video and audio quality and you're you were explaining something quite clearly, it was pretty easy to get a lot of views. These days, though, it almost feels the other way around. Like, it feels like there's actually too much supply and not enough demand for the supply.

Like, I don't think we've got enough eyeballs on the planet to supply every video that is produced with like a thousand views. I just think we're not gonna be able to get there anymore. It's hard for me to give advice on this, because like one thing that I've done most of my career, I've always had a blog, I've always done tech talks, and these were usually really well received. So I've accumulated a bit of a reputation. And because of that, it's also easy for me to, like, have some followers quote unquote sort of cling on to me.

And when I do a new thing, they just that is usually quite interested. So I think the the first video I made a week after, we were already, like, between 500 and a thousand subscribers or something like that, which is I've been told relatively rare for a new YouTube channel. Like, usually it takes a while to get to your first

That's great.

But, yeah, for us, it was, like, relatively easy. Also because Marimo is, like, gaining steam, but also because, you know, a lot of people that follow me for my password were also just interested in it. Eventually though, you do wake up and you do kinda go, oh, the video style that I had before, like, without my face in it, okay. You can still do that, but you need you're gonna wanna do a whole lot of editing. Like, watch a popular YouTube video and just put it on mute and just pay attention to how many scene cuts there are, how often like a pop up appears, and oh my.

Like, if you really want to be high up there, then it does become a lot of the editing game. And Yeah. Okay. Like, I do prefer to be able to do all that stuff myself without too much crazy editing. So, okay.

I I will make sure that I've got like a really solid introduction, a good thumbnail, and like do all of those basics. But from there on, what I just really try to do is just make sure the editing is minimal for myself. I also program my keyboard for video editing. We've mentioned that in the pre call before as well. And everything else is just okay.

Am I still excited about it? Because if I'm excited about it and I'm if I'm able to sort of be proud of a video I made in terms of, like, information density, do people actually learn from this? Then that is a game that I can keep on playing for like years is also the impression. Because I don't wanna you don't wanna be a person who has to like spend four days editing a video and only gets gratification from like having views because somebody has just super stochastic. Right?

So my my main thing is like, okay. Just make sure we've got the velocity. Make sure we've got, like, a good video every week. Make sure every video has, like, a good lesson in it. And if we don't get 10,000 views in every video, that's also just kind of fine.

6% growth a week, if we can just maintain that, that's also just solid.

Yeah. That's very sensible way to do it. So just focusing on enjoying it, doing a process

I'm focusing on the marathon. That's kind of another way of putting it. Like, if I if you if you have every video be a sprint, part of me is also wondering, well, you need to do it for years, typically. So, you know, we as a team, we also kinda get together like, hey. This quarter, what's, like, the thing we wanna really get at?

The first quarter was, like, education. Second quarter was, like, experimentation, and now we're sort of doing specific community outreach, basically. I'm sure we're gonna pivot a few times as well. But for, like, for YouTube, my impression is that consistency is one element that really matters a lot, though. So, like, if you are in a position to really make, like, one long form video every week and, like, a bunch of shorts, YouTube does reward consistency is impression.

Yeah. Yeah. And another fun

game you can play is is just the game of, can you predict how many views a you a video will get? And then this is just sort of as a mental exercise. Because you also have to keep your eyes a bit open in terms of like, okay, am I missing out on something that could work well on YouTube? And like, okay, if this video didn't get that many views, why might that be? And also do the thumbnail AB testing and like experiment a little bit.

But it but still with all that in mind, it's a marathon, not a sprint.

Yeah. That's actually a really good I I think I might try that. Trying to predict like what it's gonna get. Because then, yeah, you kind of will know, I guess. Okay.

So I think this will get a lot and it will which means x. And because this like if it's for me, it's like because this person's famous or because this is a controversial take. And then you realize like that didn't do well. Okay. Well, my hypothesis maybe is not true.

And and then and then you have to answer the question like why might that be? Yeah. I've actually noticed on my so on my personal channel, I'm now in the AB group, so I'm able to not just AB test the thumbnail, but also the title. Oh, really? Yeah.

So that is something that they are gonna release onto the world at some point soon. So like be imagine way more clickbaity titles on the short term, and with that in mind Yeah. Than on my channel because YouTube in general. Yeah. But but, yeah, it's it's a lot of it is about expectations.

Like, can you ground yourself? Can you can you come up with a theory of how YouTube sort of works? That that that's a good sort of thought exercise, I would say.

Yeah. That's that's super interesting. Do you have any like, if you had to give one piece of advice to founders who are starting YouTube channels, DevTools founders, of course.

In the end, it still all comes down to excitement, if I'm really honest. And like, can you can you sort of expose your excitements in a way that like there's an exchange such that other people are like also eager to listen to it? I was talking to to, like, the founder of Moomoo the other day, and he did like, we were sort of talking about, like, what's what's hard about making a software business or, like, a dev tools business? Like, what's what's the hard part about it? Because there's this meme of, like, first time founders worry about implementation, and then second time founders worry about, like, distribution.

And we did agree that, like, distribution is a hard part of the puzzle. Like, that doesn't happen automatically. You can't say, let's build it, and they will come. It's not the way the world works anymore. But in the end, like, the hard part is also, do you have something that actually moves the needle?

That that's that's maybe also like a way to think about it. It's not so much about solving the problem, but, like, have do you actually have something that really can make someone's life like, professional life a whole lot better? And if you can really convince yourself that's the case, the second question is, well, why on earth doesn't anyone recognize that? Right? But you do need that first bit.

Like, you do need to really convince yourself that what you've got is actually, like, a great thing, And then you can sort of push your excitement and then sort of you need one before you can actually do the other. So I think promotional activities is the exchange of enthusiasm. But before you can be excited about something, you actually you do need something of value.

Which I guess, solve solving a problem that is actually burning rather than just like like something that really matters rather than something that's, oh, that's kinda cool.

And and also again, and a good and a good just check is just again the excitement thing. Like, if you find that it's hard for you to get excited about something, it's probably because it doesn't matter to you.

Yeah. Okay. Vincent, that was very very helpful. Thank you very much.

My pleasure.

Do you have any shout outs that you wanna give?

I guess there's a few. So one of them is actually surprisingly still Reddit. So for all of it's like, there's awkward Reddit subreddits and stuff, but that like one subreddit that I'm a fan of is again the ergonomic keyboard subreddit. And I do think it's really cool to see that there's still places where communities can kind of thrive and where communities also sort of protect themselves. So if you try to post marketing y stuff in the ergonomic keyboard channel, you're just gonna get revoked instantly.

And, like, a lot of these little Reddit channels still kinda do that. And the reason that I mentioned it is that sometimes I'm able to share a video on a subreddit that has, like, very strong moderation on it, but they still accept the video because it's relevant to their community. One like, a lot of marketeers kinda like to complain about Reddit, like, oh, it's so hard to get something accepted. In a way, that's true, but consider that a good thing and also consider it a point of pride if your thing does get through because it means you're doing stuff that's relevant to a community out there. So don't sleep on Reddit is like a thing I would maybe also say.

Be picky in which subreddits you pick, basically. But especially the smaller communities do tend to be more active these days on Reddit too. So that's like a thing maybe people should recognize. And also, I love me a good blog. It feels people are doing more slop stuff on blogs.

I I really like the fact that we still actually write them by hand. So like to any DevRel people or like founder people, say it like you mean it, don't do the slop thing. People can tell.

Yep. True. We we just did an episode with Ben from Planet Scale, and it was like the which is kind of the antidote to to slop. I don't know if you've seen

Oh yeah. Sloping place. Yeah. Yeah. No.

They're they're great. But you can tell. I mean, again, like, oh, these people care. And oh, they and and then their excitement is the thing that matters. And that's it's hard for me to share our Slack communication, but it's also nice that, like, at the company right now, like, we all care about the quality of Marimo.

And one of the things I also do is just if you're a dev help person, you're typically also trying to look for experimental excuses to use the software in ways that aren't necessarily meant to be used. And that there is something about being core developer where you always kinda stick between the lines sometimes because you want this thing to sort of work in a stable way. But I guess that's, like, another bit of advice. Like, if you're a DevRel person, you can listen to the community and you can give that feed. You can forward that feedback to the engineering crew.

You can also come up with the craziest demos yourself and see if you can break your own software. And then reflecting on that is something that engineers actually also just kind of appreciate. Yeah. You can also be a source of, like, experiments that your engineers typically don't have the time So that's also something I might recommend doing.

That's really cool.

And and buy ergonomic keyboards because they're way better for your wrists. But yeah. On that the but that end of speech on that front.

Amazing. Amazing. Yeah. And where could people learn about Marimo?

Marimo.io or the web yeah. Or the I think marimo underscore on YouTube, I think. But yeah. If you type Marimo it's it's also like a Japanese moss ball. If you but if you type Marimo and then Python, you will find the

A Japanese what?

Yeah. So it's it's I actually have some in the back of So it yeah. So Marimo is actually a moss is is a it's a species of moss that if you clump them together just the right way, it forms a ball and it survives underwater. And kind of the the cute the cute metaphor here is that, like, all these individual pieces of moss, the sum is actually greater than all these individual parts. So Marimo is named after a moss ball.

And it's also like a it's a great name because it sounds like a Pokemon, so it could be a tech thing. It abbreviates very well. The domain name was available. That's like all good reasons for like an amazing name. Thought it would

be something to do with sailing. Like

No. It's it's a it's a Alright. Well, it is a water it's a it is a water based moss ball. Like, it's submerged and all that. But no, it's it's a Python library, but it's also a moss ball.

It's

a Super fun.

Fun little anecdote. Yeah. Marimo.io. You can find us on YouTube. We also have a Discord channel.

Feel free to hook us up there. And yeah, ask me anything

if you find me. Cool. Well, thanks so much, Vincent. And thanks everyone for listening.