YouTube Based DevRel

Greg Baugues: Hello, everybody. My name is Greg. Oh, man. It's so I'm just now looking out at all of y'all. It's amazing to see so many friendly faces, and thank you all for coming here.

So let me share a little bit about my story. I had the privilege of joining Twilio's developer evangelism team in 2014. Rob Spector, who's sitting right here, hired me. I got to serve at Twilio, for nine years. And a couple days after my nine year anniversary, I got laid off.

That was about two years ago. Incredible. And Twilio was incredibly generous, and it was an incredible time. And I had this opportunity to try to, instead of, like, just hopping back in the next thing, take some time and figure out, okay. I just did nine years of DevRel.

Growing a YouTube Channel from Scratch

What do I want the next chapter to look like? And I decided at the beginning of this year that I was going to start a YouTube channel and really focus on a personal YouTube channel, and I set a goal of trying to grow that YouTube channel from zero to 10,000 subscribers by the 2025. As of this morning, I'm right around 8,500 subscribers, so I'm on pace probably to to hit the goal in the next few weeks. And what I wanted to share, is just some of the lessons learned there. I am so incredibly not an expert on YouTube.

I am an expert on starting from scratch in YouTube and trying to get to 8,500 subscribers at least because that's been my my primary focus. So this talk is for anyone who's looking to get started either with your personal brand or starting up a YouTube channel. And in fact, I am so confident in this. I recently just changed my my title on LinkedIn to vibe coding YouTuber. I don't know if you all saw the post by Thomas Tasek.

It's like, all my AI skeptic friends are nuts. And there's this line in there where he's talking about, he says, you know, the common objections. I don't understand the code. And he's like, are you a vibe coding YouTuber? I'm like, actually, I think that might actually be what I am today.

Why YouTube is Essential for DevRel

It's been hard to figure out titles recently. So I wanna break this up into two parts. First is why YouTube? Like, what is the opportunity here to do DevRel over YouTube? And then second, I wanna talk about how you do YouTube, how do you get started on this, and just some lessons and tips I've learned along the way.

So first, why YouTube? Let me talk a little bit about the DevRel playbook that we were doing at Twilio when I started, let's say, eleven years ago. And I think I'll hit this pretty fast because I think this theme is gonna be echoed a lot over the, these next couple days, I would broadly describe the playbook that we ran, on the evangelism team when I joined, as being two primary tactics. The first was events. We traveled to a lot a lot of events.

We spoke at. We sponsored a lot of events. You know, I think, one, just setting aside corporate budgets and zero interest rates and COVID and travel and all that stuff, just personally, I just don't wanna travel that much anymore. I have two girls. They're young.

I don't have a ton of time with them. Like, they actually like spending time with me. I wanna, like, soak that up while I have it. I I don't mind getting on a plane occasionally, but I just don't wanna do it like I used to. The second tactic, when we weren't on the road, we would write a lot of technical content.

And the idea here is that we wanted to write blog posts so that when a developer ran into a problem they have an IDE open and they ran into a problem, they would Google that problem, and they would land on the Twilio blog, and we would help them solve that problem. That was one big chunk of the portfolio of content that we would write. I think it's safe to say that today, when developers run into problems, they tend to solve that problem in the IDE via AI. And they're not going and searching. They're not gonna end up on your blog.

And so over the last couple years, I've been thinking like, okay. So we had these two big pillars. Like, lot of my career, my DevRel career was spent doing events and writing technical content. If I'm not gonna do that, what do I do? Right?

And so I spent a lot of time building stuff for the last couple years. It was really great to be, leaving right at the same time all the AI stuff came up, and I just started building stuff with AI. It was the most fun I'd had coding in a long time. And I wasn't sure if I was gonna, like, start something. And what I found, I think this is the nature of both me and my ADD and then also just AI, is I'd work on something for a couple weeks and then I'd be, like, really excited about it and then I'd move on to the next thing and then the next thing.

And I felt like after a couple years of just spending time with AI, I didn't have a lot to show for it. I thought maybe I'd start like a SaaS or something. And then I really like DevRel. I like doing this. I was like, well, what if I made videos?

Like, what if videos was the output? What if I spent a couple weeks working on a thing and I made a video about it and the video was the product? And then I could go on to the next thing and feel good because I had this thing to show for it. And so I started doing AI con or I started doing consulting for some various AI firms. And our friend, Matt Mackay, built this great tool called plushcap.com.

And he tracks the stats of a bunch of dev tool companies, about 500, dev tool companies. And he has these leaderboards of all their content. And I was really interested in his leaderboard for YouTube because of the 500 companies he tracks, only 230 have more than a thousand subscribers. And of the 500 he tracks, only 80 have 10,000 or more subscribers. And this was interesting to me because I was getting paid by some AI companies to create content for their YouTube channel.

But a lot of them, the companies I work with, had fewer than 10,000 subscribers. And I thought, well, if I could grow my own channel to the place where I had more than 10,000 subscribers, then it would make more sense for them to publish content on my channel than on their own. And then now I'm building an asset as opposed to just, like, trading cash for for videos and whatnot. And so that's where I came up with this goal at the beginning of the year of I'm just gonna focus on YouTube. The asset is going to be the YouTube channel.

The thing the product's gonna be the YouTube channel, and we'll see if we can grow this. So let's talk a little bit about why you, specifically, as someone who's in DevRel, might wanna consider doing YouTube and why I think we have a bit of an unfair advantage compared to other YouTubers. Right? And because I think if you all have younger kids, these days, YouTuber is the most popular job. It's what all the kids it's like how I wanted to be a baseball player when I was a kid.

Right? Like, it is what the kids wanna be when they grow up. But being a YouTuber is so hard. If you were to look at, say, Peter McKinnon is one of my favorite YouTubers, he has multiple camera angles, all these cinematic shots. I spent a lot of time, like, setting other niches like woodworking.

These folks are are doing you know, they got whole builds. They've got all these different amazing, like, film setups and whatnot. This is probably the best developer YouTuber, out there. This is Jeff from Fireship. He has we'll talk about him more in a little bit, but he has about 5,000,000 subscribers.

This is his setup. This is what's required, like, right in front of him to be the best developer YouTuber. You don't need a whole bunch of fancy camera equipment. You basically need a computer and a screen. Jeff actually doesn't even put his face on it.

For folks who do, you need a single camera. You need some way to record your screen, and you're good. Like, that's how you can make content for this audience. Also, this audience happens to be like the audience that probably has the, you know, most disposable income. Right?

Like, it it is Also, in order to make content for this audience, you need to be a developer. Right? You need to know how to write code. And so I've always enjoyed DevRel because I don't feel like I'm a great developer. I don't feel like I'm a, like a world class public speaker.

But I'm pretty good at both, and that gets you in this realm. To be successful in YouTube and in DevRel YouTube, you need to be pretty good at code, pretty good at speaking, and you need to be, like, competent at video production. And that combination is just not something that a lot of people have. And so I think YouTube, like, you can take just a few months, learn enough video production, and be very successful in this realm, and all of a sudden, you're not facing a lot of competition. Like, it's not like, you know, other YouTubers, you wanna be really good at video games, you just need to learn how to play video games.

Right? So you all have the prerequisite skills here plus a little bit of video editing, and you can be a successful YouTuber, and do this job that so many folks wanna do. The last part I think is really interesting about DevRel YouTube, sort of the perennial question of this conference, how do you measure the impact? YouTube has metrics. Like, they have so many metrics.

You can tell exactly who's watching this. You can tell where they are. You can tell how long they engage with your content. You can tell when they dropped off your video. You can tell, all sorts of, you know, whatever data you want, this is probably the best, tactic in terms of measurement for Deborah.

Key Lessons for Starting YouTube Content

Alright. So let's talk about if you wanna get started on YouTube, how do you get started? First, you gotta study YouTube. And I think the mistake that I made was not appreciating that YouTube is its own thing. There are so for instance, I've been giving talks at technical conferences for a long time.

And I feel good at storytelling and speaking to developer audiences. I gave this talk of the business software a couple years ago, and I remember getting off stage and feeling like I crushed it. Right? Like, just felt like like the room was with me and everything. When the video came out and I went and watched the video online, I was like, man, this is terrible video content.

Like, it was slow to get started. The it just like, I was trying to, like, do things with the room that didn't come across online. And the there are specific tactics that work on YouTube videos that you need to learn. And so at the beginning of this year, I started off and I was like, I'm just going to study this game. And so there's a couple books that I would recommend.

You could take screenshots of the or take pictures of this. YouTube Secrets was great. There's another one called YouTube Formula. There is, of course, you know, there's a whole field of folks who are trying to sell you on how to do this professionally. So there is hype.

You gotta, like, sort through that stuff. But I think reading a couple books on, like, how professional YouTubers do this stuff, will get you very, very far. They'll also help you understand more about the opportunity and the growth of YouTube, and especially the growth amongst younger demographics. I have a playlist called Learn YouTube. There are lots of folks who are teaching how to learn YouTube.

I really like Jay Klaus. He's one of the guys that I study, and he brings a lot of YouTubers on his channel. I just keep a playlist. And when I'm working out or when I'm doing the dishes or whatnot, I'm just always listening to this. You can really learn some really great techniques.

And you what you wanna do as a developer is go learn from the professional YouTubers and then bring that into the work that you're doing. If there's one person I'd recommend, like, just going like, or even just if you just wanna rewatch one video just to get started, I really like this interview from between Jay Klaus and Patty Galloway. Patty Galloway is a consultant who has worked with MrBeast. Really great, just strategy in helping you figure out, like, how do you get started and what are the main things you wanna, think about. So let me show you one one of the first big lessons I had from YouTube, and that is around packaging.

Importance of Packaging Video Content

When I started doing work with clients, I would often have these mistakes. We come up with an idea for a video and say, okay. Yeah. We're gonna ship the video. I'll have it done on Friday.

I do find estimating to be very, difficult here because, videos about developer stuff has all of the estimation problems of software development plus all of the estimation problems of writing plus all of the estimation problems of video editing. And so what always happens to me is like, yeah, yeah, we'll have it done Friday. And then I get to like Friday, and I'm like, okay. Alright. I have the video done.

You know, maybe I dropped it on Slack. They previewed it. And so then I I go to YouTube and I upload the video and YouTube's like, what's the title? And I'm like, oh, yeah. Yeah.

I need a title, don't I? Well, it's a real it's a really good video. You know? So I was like, I don't know. Pydantic tutorial, Pydantic AI tutorial, whatever.

And it's like, you need a thumbnail. I'm like, oh, yeah. Yeah. I guess I need one of those too. Like, I'll I'll go to, I'll I'll go to Canva and I guess I'll I'll throw some words on the screen and use one of their templates.

And one of the first things that I really learned from, like, Patty Galloway was the importance of packaging. And he says that when most folks get started on YouTube, they think, I'm just gonna make a really great video, and if I make it, people will watch. And he says, actually, YouTube is a click and watch platform. You can make the best video in the world, but if nobody clicks on it, no one's going to see it. And so the title and the thumbnail are actually, he says most folks do it at the end.

He encourages his clients to spend, like, 30% of their time coming up with concepts and titles and thumbnails. And to not do any video recording, to not do any writing until you have a title and a thumbnail that you are confident in. Right? And so how do you come up with ideas for titles and thumbnails that you're confident in? And again, I would say go study YouTube.

So for instance, if you were to look at MrBeast, MrBeast has a common theme. You'll see these all across different channels. Right? X versus y. Right?

So you'll see it one versus a $100,000,000 car, $1 versus a $100,000,000 house, $1 versus $250,000,000 private island. And so for me, I started thinking like, okay, how do I adapt these themes that I know work other places into our realm? The first video I ever did where I had this idea and I had like a really clear picture of the concept before I did any work on the video was this one. It's a cursor versus clawed code. And and this came from just like studying, and then like I had been playing with both these tools.

And this concept came to me before I sat down to record anything at all. And this video I remember telling my wife. I was like I was like, I just had the idea for the video. This is gonna be my first video that gets 10,000 views. Alright?

And that video to date is at 75,000 views. And and so that is something I would not have come up for the concept of this video had it not been for going out outside of the developer world, looking to see what's working on the rest of YouTube, and then trying to bring those concepts in. Next, I wanna talk about time. Time is weird when it comes to YouTube, and I've been thinking about this a ton. So I wanna share this video here.

I'll say I landed on this concept of x versus y. Right? And so then I was like, I'll make another video like that. And so OpenAI came out with a codex. And I went I was gonna do the same thing.

And there's all these kinda, like, YouTube best practices that you wanna, like, open these curiosity loops where you pose a question, but then you don't, like, reveal the answer to the question at the end, so you maintain, like, retention. I don't really love it. And on this video, I did not like OpenAI's Codec CLI. And it so it felt disingenuous to make a video where I was dragging out the suspension. I'll see if I have tried both Cloud Code and OpenAI's Codecs out head to head on a couple different projects.

Respecting Viewer Time in Video Creation

I'll kind of go against YouTube best practices here and just spoil it. Codex fell pretty short and it was disappointing. So I I basically just like spoil it in the beginning and I was like, oh, this probably won't do well. It was really interesting to me to see the comments because the comments I got so many comments from folks saying, hey, thank you for respecting my time. And as a developer, I'm like, yeah, actually, that's what I want.

Like, I I want people to respect my time. So it's like there are these YouTube best practices, but then when do you violate those best practices for the sake of your audience? And I started just thinking like, okay, what would it look like if I just took respecting my viewers' time to the extreme? In fact, what would it look like if if I were to invest so much of my time in order to give them more of their time back? So Anthropic released this paper, Claude Code Best Practices for Agentic Coding.

And it was, just I mean, it had so many great tips for using Claude Code. And I had this idea of like, okay, what would a YouTube native version of this, of this piece of content look like? And so I made this video and and the video is called, I think it's 47 pro tips in nine minutes. And so I'll I'll play you just a short clip just so you can get a feel of what the video feels like. These, pro tips pretty quickly.

First tip, clog code is a CLI. So all the things that you're used to doing with other bash based CLIs, you can probably do a clog code. For instance, you can pass in command line arguments which will be run on startup. You can use dash p to run it in headless mode. You can chain it with other command line tools.

You can pipe data into it. You can run multiple instances of it at once. So this thing goes on for about nine minutes. And and my thought there is I'm gonna spend one to two sentences on each tip, and I'm just gonna deliver as much value as possible. This video probably was fifty hours of maybe sixty hours of work between learning everything, trying to recreate all the tips myself, taking screenshots of them, editing.

I probably filmed for well over an hour to get nine minutes of content and then came back over and put all of the the bits on there. And so in my mind, I'm like, man, I am taking sixty hours of my time, and I'm trying to compress that and give it as a gift to a developer of, like, here's sixty hours and nine minutes. Right? But the really cool thing that I I'm just having a hard time wrapping my head around is I sixty hours is a lot. I I took a lot of that time.

But now, the Internet has engaged with that piece of content for over six thousand hours. And that's a it'd been an interesting way for me to think about the this work is how do you take your time, invest a lot of your time to compress that down for a single developer, but you do that then in a way where so many developers engage with your content that, you know, I'm getting, let's just say, like, a 100 x delivering a 100 x of value to the turn. Like, the world has gotten a 100 x more time out of this piece of content than I invested into it. And that feels really cool. Like, it feels cool to create that much value for folks.

So I've been thinking a lot just about time. Like, how do I use this channel to respect time? How do I, like, both know the YouTube techniques but also just be really, really aware of how precious a developer's time is and respect it and use this channel to, to, to respect that and give them that. Let me talk a little bit about process because I do think the most complex part of YouTube or the most challenging part of YouTube is the time. It is so time intensive.

Managing Time and Process for YouTube

When you look at the process of doing a video, first thing you gotta build and learn, then you gotta write a script or an outline at least. You gotta record, you gotta edit, you gotta ship. Those first three those first two steps there are what you had to do to ship a blog post in the past. Right? So And I felt like at Twilio when I'd write a blog post, I was looking at twenty or thirty hours just to do the blog post.

And it's like, okay, congratulations. Now you have a script. Now the recording starts. Now the editing starts. Right?

And so the biggest challenge for me has been how do I ship more content? And I think the way you do that is you have to start looking at how do you compress or how do you delegate some of these steps. So for instance, that can look like, how is it possible to delegate the learning? So the Anthropic post was one of the was an example of this. I didn't have to come up with that content from scratch.

I could take content that already existed, and I could create the YouTube native version of that. Alright? Another example of that is I shipped my first, interview video last week. And so there's this package called CC usage, And I reached out to him and said, hey, do wanna hop on a call and we can talk about this thing that you built? Right?

And so there, I was talking to the expert. And I understand now why so many YouTubers basically just have podcasts where they bring guests on the show. They're delegating the learning process to the expert that they bring on. You can also, I am not personally doing this yet, but talking to folks who run programs, DevRel programs, you can hire editors. I don't feel like I'm at the place yet where I understand my own taste and where I I know exactly what I want to be able to delegate that.

But editing is a place like, this is what one of my recent edits looks like. This is probably twelve, thirteen hours of editing into this interview video. I need to figure out a way either through getting sharper with the tools or by delegating this out to figure out how I can shorten this up. But I also like this because in an age of AI slop, I feel like I put a lot into the videos to demonstrate that it's handcrafted, and I feel like folks are increasingly going to appreciate the things that they can tell a human created. Another technique on the editing, I think in YouTube, in the same way that the title and thumbnail is really important upfront, the beginning of your video needs to validate that they're getting what they thought they were getting when they clicked on the title and thumbnail.

And once you have a viewer and they're watching and they're invested in that first minute, you can do less editing and less polish on the back half once they've engaged. But it's really important to invest as much editing in that first minute as possible and make that as sharp and as crisp as possible. And you can see in this video that I just recently published that in the back half, really the back, like, you know, 90% of the video, there was a lot less editing. Alright. A little bit about psychology.

Overcoming Psychological Challenges in Content Creation

I I I have found psychology to be one of the most difficult parts of this journey. I am working alone. I very much miss working with a team. It is very easy to fall prey to procrastination and perfectionism. In fact, I tried starting a YouTube channel when I first left Twilio, and I just didn't ship consistently.

I couldn't figure out the style. I couldn't figure out how to get myself to do it. My dad's a pastor, and my dad would always say, my sermon is never finished. It's just eventually Sunday morning comes. And I I will say, I stood up on the stage a few hours ago, and I looked out, and I knew I was gonna be facing you all, and that was great motivation to finish this talk.

You don't get that same sort of push when you're staring at this setup. Alright. The Internet doesn't care if you ship today, or if you ship tomorrow or ship ever. There is so much more content for them to consume. And so so much of this challenge is figuring out how do you get yourself to ship without the pressures that you feel of giving a talk or whatnot.

I was inspired by a magician who has a YouTube channel with 7,000,000 followers named Chris Ramsey. And I read an article about him, and it basically just said when he was making this transition, he just put himself on a publishing schedule. And so my publishing schedule is my goal is to ship one video a week. Candidly, I don't get there. Typically, I fall more like every ten days, every every once every two weeks or so.

Sometimes I do. But having that schedule helps a lot. Like, for instance, this video I shipped last night, May Claude Code Meow with Hooks. I'm not, like, thrilled with that video. Like, I kinda wanted to spend, like, another ten hours editing that.

But I knew I was gonna be here for a couple days, and if I didn't get it out last night, I wasn't gonna get it. And so if I hadn't made that commitment to ship, I I would've let that go. And honestly, if I would've let that go another week, I probably never would've shipped it. And so that's what I has been most impactful for me in this journey is just to say, I'm gonna do once a week. If I miss it, that's okay.

I do not miss two weeks in a row. Like, I I just cannot let that happen, but you gotta figure out what cadence can you ship on. Another on the psychology, the comments can be pretty rough. This is actually the worst one I could find, because I just delete the bad ones. Like, I like I I like, as soon as it comes out, I'm like, is my space.

I'm not dealing with this. I let this go. I do think the bad comments, and this was actually kind. I'm not even saying this is a bad comment. This is helpful because I did have the the typo, and I appreciate the the criticism.

But I would just say, like, you gotta be on guard with the comments and just I'm just fine deleting them. I don't think you I always, whenever someone leaves a negative comment, go look on their channel. And it's funny. I almost never get bad comments from anyone who has shipped a video. So, you know, take that for what it is.

And then finally, it just can be lonely. And I'm really thankful to Matt and Ricky and Rob and Andrew Baker. And I have a number of group chats of folks who I just I send the videos to, I brainstorm ideas with, I talk about the stuff I'm working on, and I think it's just really, really hard. I think the loneliness and working in isolation is the hardest part about doing this. I know a lot of folks here are looking for jobs.

Candidly, I don't know what the next phase of my career looks like. I don't know if I'll end up doing this independently or if I'll I'll parlay this into doing this full time for a company. I will say I in the last couple years, I applied for a job at Anthropic, and I heard nothing back. And I'm not saying they should have hired me. I'm saying I was qualified for the position and I should have gotten a response.

And I was, like, kind of offended by that. And I was just, like, curious about the job. Was like, man, why didn't they write me back? And I remember seeing this interview with Casey Neistat, who's a YouTuber, and he had this very famous ad that he did for Nike. And he got paid, you know, a bunch of money for this stuff.

And he I saw this interview with him. Said, yeah, before we got paid to do that, we just made Nike ads for free. And I was like, man, if I want a job with Anthropic, like, it's kinda silly to, like, just be one of a thousand resumes. Why don't I just do the thing that I'd wanna do for them? And so a couple weeks later, I made this video.

This video has now gotten like a 115,000, views now. It might be the most viewed technical tutorial on Cloud Code. And, a few weeks after this, Anthropic reached out. And and, and, you know, I don't know what will happen here. Like, this this conversation is still ongoing, and I'm I'm truly not really sure, like, what I wanna do next.

But it's a huge difference if you're if you're job hunting from being one of a thousand resumes to just being like, hey, look, I made a thing about your product and a 100,000 people watched it. And if you want me to keep doing that, you could bring me on. And so I think this is an unfair advantage for any of you who are looking for your next DevRel opportunity. Finally, to wrap this up, are you too late? I was inspired by Jay Klaus.

He has a channel with Notion tutorials. And he says, in the tech world, every single week, new things are coming out. And there's new new content to be made every single week. All of the videos that I've made on this channel this year are about technologies that did not exist twelve months ago. Right?

And so so you are not too late. In fact, the opposite might be the bigger problem is that I'm not sure if any of these videos that I've created will still be relevant in twelve months. Alright? And so it does feel like a bit of a treadmill. And I I do think there's a bit I understand how folks get burned out.

Finding Your Unique Style on YouTube

But if you are looking at this and being like, oh, man, there's already so many folks doing this, I would say you're not too late. I would also say I think there's so much opportunity here for you to find your style. Jeff from Fireship, as I mentioned, this is one of his videos looks like.

Greg Baugues: Out of nowhere, Amazon released its own AI IDE named Kiro, yet another Versus Code fork that will compete for the hearts and minds of programmers alongside editors like Cursor, Windsor, Firebase Studio, and Copilot, just to name a few.

Greg Baugues: So Jeff does no face, highly, highly, highly edited, four or five minute videos, really funny, like sarcastic but not cynical, as Ricky put it earlier. Then you have, like, the Primogen, they and Theo, they do streaming, and then they cut their videos from there. Tina Huang, makes really great technical tutorials, has almost a million followers. This guy Nate b Jones, like, basically, like shoots from his webcam, stream of consciousness, has 50,000 followers. You can do interviews.

You can do technical tutorials. You can be funny. You can There is You could There's so much opportunity here and there are so few developers who have video production skills that there is like unlimited opportunity for you as a developer to go learn what's working in the rest of YouTube, learn a little bit of video production skills, and then come do it your way here. And the best advice that I read here is just from mister beast. And he said, you the way you're gonna get better at YouTube is just to ship a 100 videos and to get better each time.

Thank you all very much.