Why Claude Recommends Resend & Building in Public | Better Stack Podcast Ep. 15
BBetter Stack
컴퓨터/소프트웨어마케팅/광고창업/스타트업구직/면접AI/미래기술
Transcript
00:00:00Welcome to the Better Stack Podcast, where we have conversations about AI, software dev,
00:00:04and all kinds of new technology. I'm one of your hosts, Richard, and I'm joined by...
00:00:09I'm James.
00:00:10I'm Josh.
00:00:11And I'm Chris.
00:00:12Hey, Chris. So it's really good to see you. We've been trying to organize this podcast for a while,
00:00:17but we're glad to finally have you on. For those of you who have no idea who Chris is,
00:00:21I'm going to give him a chance to introduce myself. So who are you, Chris?
00:00:24Yeah. My name is Chris Penningson. I'm the DX engineer over at Resend. We're an email API for developers.
00:00:32I also have a YouTube channel, Coding in Public, that I've had for the last several years as well.
00:00:36So that's how I think we connected is through both those things and excited to chat about that
00:00:40and anything else you all want to talk about.
00:00:42And your cool video setup with your mic.
00:00:45That's thanks to Resend. Yeah.
00:00:47All right. Yeah. So we can get into that a bit later.
00:00:50But what would you say helped you get into Resend?
00:00:55Because you've done a lot of things in your past.
00:00:56So how would you explain that journey of getting into Resend?
00:00:59Yeah. I mean, it's hard not to just say that I feel super lucky.
00:01:03That's a lot of it.
00:01:04I just feel like right time, right place with a lot of stuff and connected with Zeno in particular.
00:01:09He's the CEO of Resend.
00:01:10I started my YouTube channel kind of as a chance for me to kind of talk through what I was learning.
00:01:16And one of the things I was learning at the time in 2023 was React email.
00:01:21And then Resend when it came along, I think Zeno actually reached out and gave me access to the beta.
00:01:27And I posted about both of those things right around the time that they launched.
00:01:32I didn't know it at the time.
00:01:33But when the team saw it, then Zeno reached out to me.
00:01:36We just kind of struck up a friendship.
00:01:38And then I think we met in San Francisco in 2024 that summer.
00:01:43I was in town and I just messaged him and said, hey, do you want to connect?
00:01:47And he's a super busy guy.
00:01:48And I figured he didn't have time, but he did.
00:01:51And we ended up just mostly chatting about life.
00:01:53Like it wasn't me trying to pitch myself to him or they didn't have an existing role open.
00:01:58But we kind of struck up a friendship and he reached out that fall and just said, hey, we have this DX role open.
00:02:05I'd love to have you consider it.
00:02:06And I hadn't really even thought that direction.
00:02:10And one thing just led to another and felt like it was a good fit.
00:02:13And it's the people at Resend are really what make it.
00:02:16And there's a real camaraderie with the team.
00:02:19And so all that kind of fell into place.
00:02:21Met Zeno, met the team and jumped right on board.
00:02:24And I think it was like four weeks from him talking to me to me being in Brazil with the entire team.
00:02:30So it was a pretty quick turnaround once everything happened.
00:02:33Yeah.
00:02:33And I just feel super lucky to have a chance to be able to like teach people Resend and try to push what we're doing to others in a way that's hopefully friendly to devs and not super salesy.
00:02:44That's not really in my nature anyhow.
00:02:46Yeah, that's the brief version of the journey.
00:02:49Nice.
00:02:50I think I got some misinformation from Grok because apparently you made a massive video on your channel about React email.
00:02:57In fact, I got loads of views and they reached out to you from there.
00:02:59But I guess that was incorrect.
00:03:01Well, that's not totally wrong because that's how I first met Zeno because he reached out to me having seen that first video and said, hey, yeah, basically like, you know, I'd love to connect.
00:03:11And so we kind of started chatting over there.
00:03:13Then I did the one on Resend.
00:03:14Then we met in person.
00:03:15So I kind of it started there.
00:03:17That's the first thing I think that brought me onto the radar.
00:03:20Nice.
00:03:20Okay.
00:03:21So AI isn't completely hallucinating.
00:03:24Not this time.
00:03:26I was saying before the podcast started, actually, one of my first videos on like developer YouTube is React email.
00:03:32It was the third video I ever made was on React email.
00:03:35So sort of it was a similar time in like 2023 when I made it.
00:03:38React email first came out.
00:03:39But yeah.
00:03:41Yeah.
00:03:41I didn't even know it had just come out.
00:03:43I think I did a video within a week or something of it coming out.
00:03:46I just saw it on Twitter and thought, oh, this is cool.
00:03:47And I did a video on it.
00:03:48So lucky timing.
00:03:50Did you have a full-time job at the time that you guys connected or was YouTube kind of the main thing at the time?
00:03:57Yeah, YouTube was like this side thing that I was doing.
00:04:02So it wasn't my main thing.
00:04:03I helped at my church.
00:04:05I helped.
00:04:06I did.
00:04:06I had like a side thing that I was doing myself.
00:04:09And then I also worked part-time at a company doing documentation.
00:04:11And I did that for about eight years.
00:04:13So that was kind of my main thing, that documentation job.
00:04:16So this, you know, life was kind of multiple things all at once.
00:04:19And the YouTube thing was honestly just a side thing.
00:04:22Like when I first started YouTube, I didn't tell my wife for six months.
00:04:25I just would record and talk and like put it up.
00:04:27I never talked to anyone about it.
00:04:28It wasn't supposed to be a thing.
00:04:29I didn't mean for anyone to ever see it.
00:04:31Like it was just, I'm very verbal and I need to kind of talk it out to know if I know something.
00:04:36And, you know, you all know doing videos, like sometimes you can sit down to record and you're like,
00:04:40I don't really understand how this works.
00:04:41And so I need to like, I need to start over.
00:04:44And so there's been lots of times where it was just like me trying to talk something out.
00:04:48I'd record it.
00:04:49It didn't work because I'm like, I clearly don't understand this.
00:04:52So I had to redo it.
00:04:53So it was always just this side thing.
00:04:55And then eventually I'd get a video out talking through it.
00:04:57In a way that I felt comfortable, that I understood whatever the topic was.
00:05:00So, yeah.
00:05:01So the YouTube was always a side thing.
00:05:03And kind of the other roles were the things I mostly spent my time on.
00:05:06But no development experience before Resend really, like as far as like part of a company
00:05:12or anything like that.
00:05:13So, like I said, I just feel super lucky and, yeah, super, yeah, glad that Zeno took a chance on me.
00:05:20Yeah, teaching or like, yeah, like creating content or even teaching, like just teaching myself helps a lot
00:05:27in just understanding how things work and putting the pieces together.
00:05:30So super blessed for that.
00:05:32It's like you think you know something and then you go to talk through it and you're like, I really have no idea what I'm doing.
00:05:37And then there becomes this artificial thing where you do it enough that you like you prepare these videos and you get it out there.
00:05:46But you, it's very prepared.
00:05:48And so one of the things I started doing a couple years ago is I started live streaming, just literally sitting down and starting something fresh.
00:05:54And because I'm like, I need to hit the wall a little bit more and really see do I know something.
00:05:59Because I got to the point where I could curate a video pretty quickly that made sense, but I didn't really understand it all.
00:06:05Or I had to look up so much stuff during it.
00:06:08And so I feel like live streaming is like the ultimate, like, do you actually know what you're talking about?
00:06:13And a lot of times you don't and you hit your head against the wall a ton and that's just part of the process.
00:06:17But I think if you're not afraid of failure and you're not afraid of like hitting those walls, whether it's in public or not, you know, isn't important.
00:06:25But I think you can really stretch and grow.
00:06:28But you have to kind of be comfortable with that feeling, which is honestly kind of what makes a dev.
00:06:32Like you have to be willing to hit the wall over and over and over again and say, I'm going to figure this out.
00:06:37I'm going to research it.
00:06:39And now with AI, you know, some of that's easier and some of it's easier to avoid because you can kind of skip past you having to know stuff.
00:06:46I think it's really impressive that you kind of came into this with no professional coding experience.
00:06:51Like I wouldn't know where to start when it comes to making videos.
00:06:55So how did you find your topics?
00:06:57Because you've got stuff on Astro, you've got stuff on React.
00:06:59How did you find out what to make videos on?
00:07:02Yeah, I feel like for me, it was very organic the way it all happened.
00:07:05I was literally just whatever I learned that week, that's what I did a video on.
00:07:09And that's how I've always been with my channel.
00:07:11There's no problem if somebody wants to make it a career or a big thing or try to draw attention.
00:07:16I think that can be a really helpful way to gain an audience, a community, all that kind of stuff.
00:07:21But for me, it's never been that.
00:07:23And I've had lots of chances to make it that, but I've really just tried to keep it to whatever I'm doing that week.
00:07:28That's what I do a video on, which means I don't have like, I don't have any videos that have blown up, like traditional, you know, dev blown up.
00:07:35I'll have several with 100,000 or something like that over several years, but I'd have never had anything, I don't think, approach a million or anything close to that.
00:07:43Because it's just whatever I'm learning.
00:07:44So it's not like hype topics.
00:07:47And so literally when I sat down, I was like, okay, I'm going to learn this JavaScript concept, then I'm going to do a video on it.
00:07:53I'm learning this thing.
00:07:54And I happened to do a lot of stuff with Astro when it was first kind of in beta.
00:07:57And it made sense to me, especially like as it added features, I had this progress in my own understanding where it was like, oh, it would be really nice if I could do some things on the server.
00:08:07And I didn't even understand the difference between like server-side rendering and static rendering.
00:08:11And then it was like Astro was adding stuff as I was needing to learn it.
00:08:15And so a lot of the Astro stuff is just like, that's what I did stuff on because I was learning it along the way.
00:08:21So I think for me, it was a very organic process where whenever I learn something, I teach it.
00:08:26And that just happened to progress in a way that gave enough attention where, yeah, I feel like lucky in that way that I got connected to people.
00:08:35But yeah, there was no intentionality in it, which I know is a big leg up to have arrived where I have.
00:08:42Yeah, I think that's a very healthy place to be, not trying to kind of chase hype topics and talk about what you're learning.
00:08:49I would say it was a kind of profitable place to be on YouTube a while back, but now it's completely changed.
00:08:56I think a lot of creators are saying you can't make money doing tutorials.
00:09:00You either have to do hype topics or entertaining things.
00:09:03How do you feel about that?
00:09:04And is that going to change the direction of your channel?
00:09:07I think for me, there's a lot of truth in that.
00:09:09If you're really trying to make a career out of it, I think I am, I feel like I'm lucky that I've never tried to make it that thing.
00:09:15So it's always been a fun side thing.
00:09:17It's always been a way for me to, it's kind of been for me, like hopefully not in a selfish way, but just like I need the chance to talk out what I'm learning.
00:09:24I do think that like views are down like crazy across a lot of more traditional like walkthroughs.
00:09:30I think you all have done really, really well.
00:09:32And so I think you've kind of have this nice blend of like being educational.
00:09:37It's not like a straight walkthrough.
00:09:39It's more like conceptual stuff.
00:09:40A lot of times like Richard, I love it when you get out your notepad and start drawing.
00:09:44I'm like, man, I need to do that.
00:09:45That's such a good idea.
00:09:47So I think there is a way to teach that captures a lot of that.
00:09:52But I think traditional walkthrough tutorials, you won't get the views maybe you used to three or four years ago.
00:09:58For me, I think because that's never really been my target.
00:10:01That's not a huge deal for me.
00:10:03I have kind of fallen off YouTube the last several months, but that's mostly because I'm really the only person at Reason doing all of our marketing, all of our education.
00:10:12And I have been, we hired somebody and then she's another DX engineer for us, but has moved to events pretty much entirely.
00:10:19And so I'm kind of by myself.
00:10:21I just don't really have capacity right now.
00:10:23So it's mostly been by nature of just the way life is right now.
00:10:27We're trying to hire.
00:10:28And so once I get somebody else, I think that will free up some more of my time.
00:10:31I don't know exactly what that will look like going forward for my channel.
00:10:34I think probably I will keep doing the types of stuff I'm doing, but I would love to get both more conceptual in some things and then more detailed in others.
00:10:43So my tentative plan is to like do more conceptual topics and like quick deep dives on stuff and then also to do more slow stuff where I'm like learning something from scratch, teaching it from scratch for myself.
00:10:56I don't think those will do as well, but I'm okay with that.
00:10:59Like I said, for me, my motivation has kind of been singular since I started and that helps me, I think, kind of pick my true north and just stick with that.
00:11:07And I feel lucky to be able to do that and not have to worry too much about views, but who knows?
00:11:12We'll see how it goes.
00:11:13I think quite a good example of that is Bash Bunny.
00:11:16She does quite well on sort of live streaming.
00:11:18I think she's going through a rust journey at the moment and just sort of teaching herself that.
00:11:22And I think that's quite a good example of there is still an audience for people watching people code still without AI in the way.
00:11:29Yeah, it's interesting.
00:11:30Like when I don't use AI, I get a lot of comments like, I'm so glad that finally it's somebody who's not using AI, you know, so I think there is this fatigue as well.
00:11:41But I think there needs to be a way to kind of do both, you know, because it has changed coding fundamentally and it will continue to do that.
00:11:50And I think there's a way in which like you have to figure out what this new world looks like.
00:11:54But I also think that there is something just really tactile about just touching the syntax.
00:11:59Like, and I just like doing that.
00:12:00And so I think I'll always have a soft spot in my heart for it and probably will always do some of that kind of like non-AI coding just for fun.
00:12:09Yeah, I can't imagine how you have the time to do that because you've got Resend, you've got home life, family life.
00:12:17How are you finding the time to make your own videos and to learn new skills and to write code by hand and all that stuff?
00:12:23I think like a lot of devs, like I'm very curious and always finding ways to like duplicate stuff.
00:12:28So, you know, I have three kids.
00:12:32So even like little things like we, I coached two of my girls soccer, football.
00:12:37But yeah, as we, as I coached them, I'm like, oh, I could spin up a little website for this.
00:12:42And so it's like something I do with them.
00:12:43So I sit down with them.
00:12:44We do that together.
00:12:45And I'm like, oh, there's this concept.
00:12:46I'd like to figure out how to like stream this thing from the server.
00:12:49So I've done that before.
00:12:50We had some kittens at one point that we had to sell.
00:12:54And so we, I sat down with my girls.
00:12:56They helped me code the page then, or like design the page.
00:12:59And then we coded it.
00:13:00And it looked very much like a six and eight year old at the time, I think, you know, designed it.
00:13:05So I try to like blend a lot of life together like that.
00:13:08I've got like a 3D printer.
00:13:10So I was working on some random app with my son the other day and doing the same kind of thing.
00:13:16I'm really into Liverpool Football Club.
00:13:18Hopefully that doesn't destroy my reputation with any of you.
00:13:23And so it was my son.
00:13:24So we've like spun up a lot of fun things like that.
00:13:25So I think I just try to kind of bring that creativity to all the different aspects of life.
00:13:31And I love like doing like crafts and arts.
00:13:34And like I do a lot of that kind of stuff with my kids.
00:13:35So I, I just have a lot of energy for that.
00:13:38And when I'm doing it with other people, like it doesn't feel like it pulls me away from them.
00:13:42It feels like it draws me toward them.
00:13:44So whether it's friends, yeah, or my kids.
00:13:48Like I try to spend a lot of time like using that creativity in all these different aspects of life.
00:13:54And then I learn stuff that overlays and it makes it easy to use it at work or makes it easy to use with my channel.
00:14:00Usually my channel, I'm recording early in the morning.
00:14:04And so I'll usually record like 5.36 in the morning.
00:14:07I get it pretty early.
00:14:09And so that's kind of my channel time when I have it.
00:14:12And then, yeah, work takes the rest of the day.
00:14:15And then in the evening, sometimes I'll do stuff with my kids.
00:14:18So I don't know if that really solves or answers your question.
00:14:21But yeah, just try to bring that creativity to all of life and make it connect me to people instead of pull me away from people.
00:14:27I was going to say, I think that's one of the coolest things about sort of AI coding I found is getting a lot more people who like you don't have to go back and teach the syntax of HTML, React, all of that to people necessarily.
00:14:39You can say like just prompt Claude and then we can sort of talk through an understanding of generally what's happening.
00:14:43And it's a lot easier, obviously, to get someone's like first app out there is so much easier now.
00:14:49And I think, yeah, AI is great for sort of like micro sites I found.
00:14:52Like any small thing I have now, instead of sort of Googling if there's a service out there, I was like, Claude, can you make this?
00:14:58Because, I mean, I'm its only user, so I don't necessarily care that it's the best quality it is.
00:15:04So, yeah, it's been nice.
00:15:05And, yeah, it's cool to hear that you're sort of using it with your kids in that way.
00:15:08Yeah, yeah, it's definitely sped up a lot of that kind of stuff.
00:15:11And it helps them come up with ideas and like one of them, they wanted to do a library with their friends.
00:15:16So, they built all these custom little books and we built a website that they could check books in and out of.
00:15:22And it was just like the four of these little kids, you know, a couple of my girls and a couple of their friends.
00:15:27But like that's a super fun thing to get to do.
00:15:28And they felt like super inspired and like, oh, can we do it like this?
00:15:31And, you know, normally I couldn't have time for that.
00:15:34So, I think there are ways in which like my friend who works at Resend, her name, Christina,
00:15:41she says like apps can be like a little home-cooked meal.
00:15:44Like you can just like bake somebody a little meal and it doesn't have to be super important.
00:15:48And AI makes that easier to do because you could do that very quickly.
00:15:52Yeah, no, I wish I had that as a kid because I started learning to code because of Minecraft.
00:15:56I wanted to add plugins to it.
00:15:58And that meant I had to learn Java, which I don't know how I did it, but I got there.
00:16:03Yeah, that's impressive.
00:16:04Yeah, I was going to say, you said you've got two girls and one boy and I've got two girls.
00:16:08And I think when I had my first girl, this might be too much information, but when I had my first girl, I could get away with keeping the toilet seat up.
00:16:14But now I can't do that.
00:16:16It always has to be down.
00:16:18I don't know what it's like in your house with the balance, but yeah, there's that going on.
00:16:22Yeah, you've got to change your life as a girl dad.
00:16:26Yeah, exactly.
00:16:27Anyway, let's talk about what you do at Resend.
00:16:29So it sounds like you have a lot going on.
00:16:31You do videos, documentation, you wrote the email skills.
00:16:34What exactly is your canonical job role entitled, what you're supposed to be doing?
00:16:40Yeah, I mean, it kind of connects to a lot of things in the company.
00:16:44Generally speaking, it's in the category of like DevRel.
00:16:48We call it DX Engineer because we want it to like appeal to people who are more development mindset.
00:16:56And so, yeah, it goes from everything from docs to blog posts, change logs, all of our social content, all the videos, education, kind of dev zero on every part of the product.
00:17:06I'm connecting a lot with people at events, speaking, like that kind of thing.
00:17:10So anything in that general realm.
00:17:13But like I've touched product, I've, you know, shipped features before too.
00:17:17So like it's, it ends up being fairly comprehensive.
00:17:21And as the team has grown, like there's just more and more stuff we're doing.
00:17:24So it's feels super cool to be able to talk about Resend to people.
00:17:28I kind of view my role as not just about the product.
00:17:31Like I think obviously I want people to understand what we do at Resend.
00:17:35But I think a lot of it too is like helping push people's voice internally in the company outward.
00:17:40I mentioned earlier that Resend really is about the people for me.
00:17:45And we have such amazing developers and designers and success engineers.
00:17:50Like they're doing so much.
00:17:51And so a lot of what I'm trying to do is say like, how could I amplify their voices?
00:17:55So a lot of times like I'm ghostwriting stuff for them.
00:17:57I'm taking an idea that they had that they mentioned in a meeting.
00:18:00And like, hey, could I do a quick like 20 minute interview with you about how this architecture works and why we chose this?
00:18:06And like we post this as more of an engineering post.
00:18:09I'll do all the work, but you know, we'll, I'll let you review it and we'll push it out.
00:18:12So a lot of it for me, isn't just about the product.
00:18:15It's really about the way we think at Resend.
00:18:17It's like a very, very culture first company.
00:18:20And I hope it stays that way.
00:18:22It's hard as you, as you grow larger to kind of keep that.
00:18:25But that's a lot of what I think of is not just pushing the product, but pushing the people out to everyone else as well.
00:18:31So that's kind of the full scope of my role is anything in all that stuff I mentioned.
00:18:36And a lot of people think that way too at Resend.
00:18:39So Zeno, the CEO is very much, is very good at social media and talking about what we're doing.
00:18:44And a lot of our engineers will like post their own stuff as well.
00:18:48So it's, it's not like I'm the only one who cares about getting stuff out to others.
00:18:52But I'm not like a traditional marketer in any sense of the word.
00:18:56When, when Zeno hired me, I asked him like, all right, from marketer to like educator, where is this role?
00:19:01Cause I am like past the needle over here.
00:19:03I don't care at all about this.
00:19:05And he's like, it's fine.
00:19:06We, we really just want to talk to people about what we're doing and, and let it be a more natural, organic thing.
00:19:11And, and so that's kind of how I view myself as like educating people about the product and about the people at Resend.
00:19:16When you joined, how big was the company and how big are you guys now?
00:19:21I was the 10th hire.
00:19:22Um, I think we are around 45, 46, something like that now.
00:19:28And I joined like a year and a half ago.
00:19:30Cool.
00:19:31Um, I was going to say something we spoke about in the email, but Resend is very good at getting Claude to recommend it.
00:19:38And I have no idea how you've done that.
00:19:40Um, whenever I build something that needs emails, Claude's code will always say, use Resend.
00:19:44Even though I'm sure there are many other alternatives out there, it would just get me to use Resend.
00:19:48So yeah.
00:19:49Can you speak to what, if anything you've done to help that happen?
00:19:53Yeah, I think we've tried to double down on a lot of the basics.
00:19:56Um, so a lot of the SEO stuff has always been important, but it's like, seems to be increasingly important as people turn to Claude as the kind of the decision maker with a lot of the stuff.
00:20:06I mean, devs shouldn't really have to care about email.
00:20:08You just want something that works.
00:20:09And if it works, you're good.
00:20:11Um, and so, you know, you have to really, uh, kind of cover the basics at least.
00:20:17So that's everything from like JSON-LD, like structuring the page well, basic things like making sure your headers, you know, you have like a single H1 and, you know, that kind of basics of SEO.
00:20:27Um, we've tried to do some extra stuff.
00:20:30Like, obviously we have like lms.txt for stuff.
00:20:33Uh, we have that for like our pricing page, for instance, if you, um, like do a curl request, it returns markdown, like that kind of stuff.
00:20:40Um, we've also found things like Q and A's are really important.
00:20:43So whenever we have a possibility, we have like an accordion with Q and A's, it seems like LMs prefer that kind of stuff.
00:20:49We've done some like trades with other companies where we'll do like the five best, you know, this, you know, notification API and they do one on the best email.
00:20:58And that, those also seem to be very important to AI.
00:21:01Uh, I think some of it too is honestly a recency bias.
00:21:03Uh, it seems to prefer, uh, like newer vendors with a lot of stuff.
00:21:08And I think we kind of came in just at the right time.
00:21:11Uh, the company was founded in January, 2023.
00:21:14So we really came in just as alums were starting to really get trained.
00:21:18And that recency bias seems to be a part of it as well.
00:21:21Uh, I think some of it too, is like, we're trying to constantly be quick to a lot of the AI things like MCP and, uh, skills and, and that kind of stuff.
00:21:30Um, so being there allows us to kind of get the normal SEO plays that you would get if you were quick to any part of an industry where other people who are quick there as well are talking about you or using you in their tooling.
00:21:43Um, so I think all of those things are kind of where we've started, but I don't know that we have any like perfect silver, silver bullet.
00:21:50Uh, we've tried, there's lots of companies that try to kind of crack the AI SEO or GEO or whatever people call it these days.
00:21:57Um, and so we've tried to follow those, um, you know, prompts when we've gotten the chance to, to improve in that area.
00:22:04But I think for us, it's mostly been get the SEO basics, right?
00:22:08Be first to a lot of the LLM stuff.
00:22:10And probably a lot of it, honestly, is recency bias, which means we have to keep kind of investing in this area, trying to learn what works and, uh, and hope that we stay at the top of the list.
00:22:20Cause right now I think it's about 70% of the time that Claude will suggest us.
00:22:24Yeah, I think that's amazing.
00:22:24I think you're one of the few companies that Claude suggests railways, one of them sometimes fly, um, sometimes for sale, but yeah, I don't know how people are doing it.
00:22:34They just got these crazy tricks and yeah, I think Claude has, have a few things that they recommend people do, uh, and like put stuff in robots.txt, but no one really knows what they're doing and I think it would be interesting to see what happens and I guess models will change it so quickly that they might just change the way they work and so you'd have to redo everything again.
00:22:53So yeah, it's a fun place to be and it'll be interesting to see what happens.
00:22:57Yeah.
00:22:57And there's like, you don't want it to be too gamified because then everyone could just do it.
00:23:01So like there is a little mystery to it as well, like normal SEO, I think, um, and I'm not totally sure like the model providers always know exactly why stuff gets picked up either.
00:23:10So, um, so I don't know, I think it's definitely going to be moving target and you don't want to like celebrate before you cross the line.
00:23:17Like it's very possible that, like you said, a model changes and suddenly we're not recommended as much, but right now, at least it seems to be, uh, like a combination of word of mouth with, um, and then also, you know, Claude as well.
00:23:29Uh, and other, uh, LMS, but we're hoping to stay at the top of everyone's mind, whether agent or human.
00:23:35And how do you think the AI stuff specifically has helped with recent growth?
00:23:39Because I think Xeno must have posted a tweet with a chart of it massively going up, the amount of users on, uh, recent, and do you think that's related to AI or do you think that's just other things or a bit of both?
00:23:51Yeah, I wish I was like, and these are times I wish I was like a true marketer because I'd have data and all this stuff, but like, I, I don't, I'm, we're trying to get better at that.
00:24:00Um, and, uh, I think it's a combination of a few things.
00:24:03I think AI is certainly a part of it, but I don't know that it's a direct line between somebody saying, Hey, what should I use?
00:24:09And, and recent, for instance, like we have had an integration with Supabase for a really long time.
00:24:15And as Supabase has gotten picked up by a lot of like, like lovable and a lot of these other, uh, AI coding agents, like people are wanting to add email and that's an easy recommend, uh, through that integration.
00:24:25So I think that's one, like, as we've been connected to other companies and AI is affected, not just us, but also them, like it's kind of, uh, you know, what is it?
00:24:33The, the rising tide raises all boats or something like that.
00:24:36Like we've all kind of gone up together.
00:24:38So I think that's a lot of it.
00:24:39Um, I think the best kind of marketing is always word of mouth.
00:24:42Like people just talk about you, they love you and we thankfully have a lot of people like that.
00:24:46And so I think that's also a big factor as well.
00:24:49We're still very dev focused, like that's our focus is like on developers first.
00:24:54And, uh, so with this kind of new age of development, we have a lot of more vibe coders and creators, uh, who are using us, but it's also like, you know, transparently mostly has been for developers.
00:25:07And so we're trying to figure out how to adjust to that.
00:25:09And so I think right now, a lot of it still has been word of mouth because it's been developers in particular.
00:25:14And sometimes when we get people who are more in the vibe coding, uh, creator area, they can struggle because it's working with DNS and a lot of weird stuff.
00:25:22That's not about us.
00:25:23It's just how email works.
00:25:24Um, but I think it's been level and, uh, super base.
00:25:29And some of these integrations that we have, uh, that have made it easy to kind of, you know, go up when, when they go up.
00:25:35And while we're on the topic of AI, how have you kind of personally changed the way you work because of it?
00:25:41Yeah, man, uh, the last, what, five months now has probably been the biggest change.
00:25:46Um, I think there's a lot kind of personally, and then there's a lot with work stuff.
00:25:50So maybe I'll, I'll talk mostly about work cause that's really where I've mostly invested in it.
00:25:54Um, for me, it's like really helpful gathering context from, uh, from others.
00:25:59So for instance, like we have a changelog coming up and I, I have a sense of what the feature is, but I'm like, Hey, go do research.
00:26:05Tell me, look at, you know, use the gap CLI I'm telling Claude this, um, you know, tell me everyone who's worked on this project, uh, like read the RFC, that kind of stuff.
00:26:14Give me a bullet list of like the things we should cover.
00:26:17I really think like there's some power in crafting sentences when you're writing stuff.
00:26:21So I really try to write all this stuff, but a lot of research stuff like that has been super helpful.
00:26:26Um, there's a lot of common patterns we have in our code base.
00:26:28So like if I'm doing a customer story or like a new landing page or like, there's a lot of stuff like that, where I can borrow from what we've already done.
00:26:35So I'll often have an agent working on like, Hey, get, build me a new, uh, landing page.
00:26:41I'll drop in all the content.
00:26:42I'll write all the stuff, but give me the structure there.
00:26:44So there's a lot of stuff like that.
00:26:45That's really sped me up.
00:26:47Um, I think we were fairly early to agent skills.
00:26:50It was before cursor supported them, that kind of stuff.
00:26:53Um, and so when I heard about it a lot, I'm like, we really need to jump on this.
00:26:57Um, and did a deep dive in like, what are these things?
00:27:00Why are they important?
00:27:01You know, that kind of stuff.
00:27:02And, uh, and so like, that's also been a big factor is like just figuring out the tooling, like how I would work tooling.
00:27:08So figuring out which skills I enjoy working with, how to invoke them, Claude ships, seven new features every day.
00:27:14So I'm trying to stay up on that.
00:27:15Um, and then like probably a more on the personal side.
00:27:18I, I like, I have an open call agent.
00:27:20We actually have several within recent as well.
00:27:23Um, and that we use for certain like very limited tasks.
00:27:26Um, but I use my own to basically manage all my tasks through an obsidian vault that syncs to GitHub that I also can look at locally.
00:27:33And that's a whole complicated system you don't want to hear about, but, um, but that's been really helpful as well to kind of have this dialogue more with my task manager.
00:27:40But I can also open it up in a program and, and kind of, uh, scope through it as well.
00:27:44So I think on a personal level, it's helped me kind of be able to throw stuff at the wall and just get back to it later and trust that system.
00:27:50And for work, it's mostly sped up a lot of my research quick, quickly understanding kind of what's going on in certain areas.
00:27:56Um, giving me an outline of kind of where to go or with coding stuff, mostly around like small feature requests, like, hey, cursor, um, you know, in Slack, hey, cursor, do this, um, small thing that we saw this bug.
00:28:07Uh, or it's helped me, um, like, spit up new landing pages quickly, or, you know, get the structure of, uh, some feature done quickly.
00:28:16And then I kind of get in this, into the code still myself.
00:28:19Yeah, but that sounds really good.
00:28:21Um, I, I didn't mean to send to an AI podcast, but you mentioned, you've got several open clause at recent, and I, I think a bet is like, we don't have any, I think personally, I have a fear of security.
00:28:33It's just like getting access to things it shouldn't have access to.
00:28:37And so number one, how many, or why have you got several open clause at recent and what are you using them for?
00:28:44Yeah, they're mostly, uh, very scoped to certain teams or actions.
00:28:48So like we have one for marketing, um, and we have like, we've scoped it, it lists on its own instance.
00:28:55Um, it has access to like its own one password vault, so we can give it just a very limited access to things we wanted to have and scope, uh, those API keys.
00:29:05So we'll use it for things like it has its own Twitter account.
00:29:08It has its own Gap account.
00:29:09And like every Monday it will, uh, spin up.
00:29:13It'll look at like all of our, um, analytic data for, for Twitter, for LinkedIn, for our SEO stuff, for, uh, our website.
00:29:23So it gathers all that and kind of curates like what are areas we're missing?
00:29:26It'll look at our documentation, like questions people are asking.
00:29:29So we'll kind of feed it very particular information.
00:29:32And then it kind of gives us a report each week.
00:29:35So that's one thing we use it for, for marketing, for instance.
00:29:37And then other teams have their own.
00:29:39Um, so like one that quickly can jump in and see like what, you know, what's the main cause of this incident?
00:29:44So that when the team jumps on, it has kind of all the details right there.
00:29:48Again, we've tried to scope it by both limiting where it lives, what it has access to through API kind of management, but it's all the interface has been Slack entirely.
00:29:57Um, and so I think some teams find it more useful than others and nobody's had to do it just as we started ours first.
00:30:03And then other teams were like, Oh, this is kind of cool.
00:30:05Let's, let's see what we can do with this.
00:30:07Um, and so I think for the most part, it's been helpful for those kinds of automated generated things, um, that require a little bit more analysis.
00:30:14Um, and then we also use it to like ask questions.
00:30:17Like we'll see a post on, on X or something and say like, Hey, here are these 10 companies mentioned or any of these customers.
00:30:24And it has access to be able to look up some basic data on that kind of stuff.
00:30:27But a lot of it has been like scoping it to very particular, um, abilities and then finding ways in which it would be helpful to like do some grunt work that I don't want to have to go search through, you know, and something I could do, but it would take me 10 or 15 minutes.
00:30:41And I can tell Hermes, which is our, uh, open club, uh, and, and the marketing team, uh, Hey, go, go figure this out for me.
00:30:47And we kind of go from there.
00:30:48So yeah, it's, it's been a cool journey.
00:30:50Zeno is actually the one who pushed it to start with.
00:30:52And so it helps when you have the CEO behind that kind of stuff.
00:30:55And, uh, we've tried to be really careful, like I said, to scope it to very particular things.
00:30:59Um, and so it doesn't have access to anything that's super crucial.
00:31:03And, uh, it's also pretty easy to, to shut down on stuff if we need to as well, which I think we have on a couple of occasions, we said, I don't know that we want to have access to this.
00:31:11So let's pull this or limit the API key.
00:31:14Um, but I think as people use things like open claw or even clawed more and stuff like that, API, like scoping is a lot more important.
00:31:22So recent, like for instance, wants to invest a lot more in that.
00:31:26And I think a lot of other companies you're seeing do that as well.
00:31:29So that way you can be a little bit more confident to give the keys, uh, to an agent to do some very particular tasks, but nothing more than that.
00:31:35Yeah, that's wild, I think, like you said, it's important to scope things and I think Stripe have like a way for agents to spend money and do things on your behalf.
00:31:44But, um, I want to dig a bit deeper into that because you said some stuff that I've never thought of doing, and it sounds really, really clever.
00:31:50And I think I might steal some of those ideas, but you have a Hermes agent that goes to Twitter.
00:31:55So it's got its own Twitter account, and so it logs into Twitter and, and it reads people's tweets.
00:32:00And what does it, does it suggest tweets for you guys to write, or does it look at what's popular and give you information about that?
00:32:07Yeah, we've tried to get it to write stuff.
00:32:09It's not very good at that.
00:32:10I'm sure that somebody with more intelligence could make it, intelligence could make it do that, but I can't.
00:32:14Uh, so mostly like it'll do competitor analysis.
00:32:17Like here's what everyone's tweeting about right now.
00:32:19Here are things that are like important, like you should consider doing something on this.
00:32:23So that kind of stuff.
00:32:25Um, but I can also just feed it a tweet and say like, Hey, could you give me more background on what's going on here?
00:32:30Like our own little grok, you know, kind of thing.
00:32:33And I should say ours is an open claw agent that we called Hermes before Hermes was a thing.
00:32:38So now it goes to Hermes.
00:32:40So yeah, um, yeah, so that's, uh, that's what, uh, yeah, that's, that's basically how we use it with social media in particular.
00:32:48Uh, it'll do competitor analysis once a week it posts like, uh, it used to post daily.
00:32:51Like here's what everyone's posting.
00:32:53And we gave it like, here are competitors, let us know what they're talking about.
00:32:56Um, and now it's more like as needed basis where we'll interact with it.
00:33:00Yeah.
00:33:01Well, that's really cool.
00:33:02I think, yeah, we need to do some more of that.
00:33:03We didn't do any of that.
00:33:04I mean, we have our own, I've got my own skills.
00:33:06Everyone's got their own things they do on their own, but we don't have like a central.
00:33:11Open claw or Hermes bot that goes to Skype Twitter and gives us really cool topics or areas that we're not making videos on.
00:33:17I think that would be very cool.
00:33:18So you're pretty much a one man team kind of for a while and AI obviously helps increase all of our productivity across things.
00:33:27But what else do you kind of rely on to keep everything going?
00:33:33I know you like Raycast.
00:33:35So if you want to talk about that, that's cool.
00:33:36But is there anything else that you're actively on right now?
00:33:39Yeah, I'm fairly obsessed with Raycast.
00:33:43I have used it since the early betas.
00:33:46I have been accused before of jumping on things too early, but Raycast is when I will happily stand behind.
00:33:52So yeah, I use Raycast very heavily.
00:33:53I think my stats every year are almost embarrassing when they do that Raycast wrapped thing.
00:34:01So that's a big thing as well.
00:34:02I've used Obsidian, like I mentioned, to kind of run this vault that handles a lot of my task management.
00:34:08We have our own like system internally as well.
00:34:11I think a lot of it too is like the team itself, like recent itself, cares a lot about
00:34:17engaging with customers with concerns we have or things we're pushing or that kind of stuff.
00:34:21So everyone's really, I feel like the whole team thinks very much externally, which if you've,
00:34:26you know, a lot of places when you work, they're focused mostly just on product in a way that
00:34:31can sometimes be like, well, of course, people will just use this.
00:34:33But a lot of the devs are super proud of what we're working on.
00:34:36And so they're talking about it and they're pushing it and they're saying, oh, we should do this.
00:34:39And so while in one way, like I'm a one person team currently ish, you know, Christina helps as
00:34:45she's able to, I think the whole company really thinks that way.
00:34:49And so that makes my job easier.
00:34:51They're always bringing stuff to me.
00:34:53I'm always listening and saying, oh, could we do something here?
00:34:55And then they'll help me kind of push it across the line, design jumps in on stuff very quickly.
00:34:59And a lot of the stuff that I'm pushing out has their fingerprints all over it as they're, you know,
00:35:04helping me design graphics and stuff like that.
00:35:06So it is a whole team effort.
00:35:08But yeah, for my personal stuff, it's mostly down to Raycast.
00:35:13I use Obsidian.
00:35:14I've used OmniFocus for like 15 years for more like global mental management.
00:35:19We use linear internally as well at Resend.
00:35:22Slack is basically where everything happens.
00:35:24There's too many channels.
00:35:25It's always nice to watch a new employee come in and you just see their eyes when they open Slack
00:35:30for the first time.
00:35:30And it's kind of like that.
00:35:32So I don't know that I have any secrets other than like I try to just be as realistic as I can
00:35:40about what I can get done.
00:35:41And then, yeah, you can get a lot done if you're realistic.
00:35:44And I think a lot of it is like warm mental mind tricks for me.
00:35:47When you kind of overextend yourself or pack your day too full, you always feel behind.
00:35:52And so I try to leave a lot of margin and that way I always feel ahead.
00:35:56And I know that's a dumb mental trick, but it works on me.
00:35:58And so I try to spend a lot of time just making sure that like when I say I can do something, I do it.
00:36:04And unfortunately, like it's harder and harder, the more you have to do.
00:36:08But I don't know that I have any secret magic trick around kind of how we, how I do the work
00:36:12other than listing some of those tools and kind of the way I go about it.
00:36:16Well, I read your article from, I don't know, a few days ago now, right?
00:36:19It was predictability is a superpower, right?
00:36:23And it was like, you talked about the whole idea of planning longer than it actually takes,
00:36:28but you only plan up until like 1:00 PM you have there, right?
00:36:33And, you know, how long, you know, how long have you been using that approach?
00:36:38Because it makes a lot of sense, right?
00:36:40Like you allot this amount of time up until this point, because our tasks take a long time.
00:36:46So, you know, how's that been working for you, I guess?
00:36:49And is that like a new thing?
00:36:50Yeah, I've done that for several years.
00:36:53And I just, I think I know myself, like I get distracted and curious and a bunch of other
00:36:57things.
00:36:57And then like, just stuff doesn't go the way you expect.
00:37:00You record a video that you're like, this should not take me more than 20 minutes to record.
00:37:03And then it takes you an hour and a half.
00:37:05And so like life just happens.
00:37:06And so I think there's basically two ways to go about it.
00:37:09One is you plan for the perfect day every day, and it never goes according to plan.
00:37:13And you always feel like a loser behind or two.
00:37:15You just plan for that and realize like, that's how life is.
00:37:18I think especially with people, it's really important to give a lot of like extra, a lot
00:37:22of time because people aren't machines and you need to adjust for that.
00:37:28Like I had a meeting with somebody the other day that was supposed to be 30 minutes.
00:37:31And we talked for an hour and like, I know some people are probably busy and important enough
00:37:35where they can just say this meeting's over.
00:37:37But I'm like, hey, there was more going on than just the thing we had to talk about.
00:37:40And they kind of needed some extra attention.
00:37:43So it's like, that's fine.
00:37:44Like, I'm not rushed in that way.
00:37:46Like, I want people to feel like they have my full attention.
00:37:50So I've done that for several years.
00:37:51But I think a lot of it is just after a while of like everyday finishing feeling like, man,
00:37:55I didn't get anything done.
00:37:56Like, then just the little mind trick of saying like, I'm only going to plan this much.
00:38:00I'm going to make sure I allot for like the not perfect day.
00:38:03And then when I finish it almost every day, I'm working on the next day stuff.
00:38:07And it's really that day stuff.
00:38:09But like my mind says, it's next day stuff.
00:38:10And so ideally, when I get to Friday, I've already finished everything for the whole week.
00:38:14And then I have bonus stuff that I want to get done, which is probably the stuff I would
00:38:17have done anyhow on that day.
00:38:19But it feels like, hey, I'm ahead.
00:38:20I'm optimistic.
00:38:21Like, I can be curious.
00:38:22I can take time with people.
00:38:24So I think it's mostly more like, like hacking your own brain and thinking, thinking in a way
00:38:29that like allows you to work as your best self rather than always behind and always pressured
00:38:33and always whatever.
00:38:34And there's a lot going on.
00:38:35So it's not like I don't feel like that at times.
00:38:37But those kinds of mental hacks have like made the experience of work better.
00:38:41I was only ever going to get that amount of stuff done anyhow.
00:38:44So it's more like, how can I shape the experience to be something I enjoy rather than something
00:38:48I'm always, you know, pressured on?
00:38:50Yeah, that really speaks to me.
00:38:52I think I'm in the camp where every day I'm like, oh, I didn't get enough done.
00:38:55I wish I got this done and that done.
00:38:57So yeah, I really need to check that out because it's an easy thing to fall into,
00:39:02like an easy trap to fall into, especially if you're quite, what's the word?
00:39:06If you want to achieve a lot, basically, and get a lot done or if you've got a lot to get done.
00:39:12And so, yeah, I think it's a good mind trick to not put too much pressure on yourself.
00:39:16So I'll read the article.
00:39:17Yeah, let me know if your brain works that way too, at least for me.
00:39:20And I know, like, I'm openly talking about it.
00:39:22I know what's happening, but it still works.
00:39:25Yeah, yeah.
00:39:26Oh, that's nice.
00:39:29So we're going to talk about your video setup in a bit,
00:39:31because I'm interested, but we have a few questions from our viewers.
00:39:35They're very recent specific.
00:39:37So if you can't answer them, don't feel the need to answer them.
00:39:40So the first one is from QProductions23.
00:39:43And he says, can you ask him if they add temporary mails?
00:39:47That's going to be easy for prototyping applications rapidly.
00:39:51I'm not sure I totally understand the question.
00:39:53Temporary mail in what sense?
00:39:56So I don't understand either.
00:39:58But in my head, it's more like, can you, this is how I understand it, can you send temporary emails?
00:40:05So like, I don't know if you've got like a, or like an email, temporary email inbox,
00:40:08and you want to test something to see if it works, maybe.
00:40:10Oh, okay, so maybe like a built-in email inbox within Resend?
00:40:14Yeah, something like that.
00:40:15Yeah, that's, it's a feature request that's come up before.
00:40:18I think a lot of it is just trying to prioritize like what helps our current customers,
00:40:23versus what would bring another customer in.
00:40:25So it's, it's something I can bring back to the team.
00:40:28I guess I'll put it that way.
00:40:29With no immediate plans that I know of.
00:40:31And second question is from Life of Code.
00:40:33Ask him about spam filters, flags IPS, and how to maintain quality to prevent it.
00:40:39I got a lot to talk about with domain reputation and deliverability.
00:40:43So I don't want to go too long, but there's a lot there.
00:40:46But yeah, we've done, I've done some content on it in the past.
00:40:49So I'd probably encourage them to look at that.
00:40:52But generally speaking, like, I think devs just assume like,
00:40:57if you get the throughput, then it's going to land in the inbox.
00:41:00And there's a lot that can go into landing there.
00:41:03Right now, basically Gmail, especially,
00:41:06and they're really the ones who own email in a lot of senses.
00:41:09They mostly care about your domain reputation, not your IP reputation.
00:41:13That doesn't mean that IP doesn't matter at all.
00:41:15We do our job to make sure our IPs are, you know, are really healthy,
00:41:20but it's your domain that matters.
00:41:22So where you're sending from and think about it like a post office.
00:41:27If you just show up one day and drop 100,000 letters on the desk,
00:41:30they're going to say like, who is this guy?
00:41:32You know, and not trust you.
00:41:34You really have to build reputation trust.
00:41:36And it can take a little bit to build and you can lose it very quickly.
00:41:42And so you have to really put in a bunch of safeguards yourself.
00:41:45So that's everything from like making sure that you like sent from a subdomain.
00:41:50So that way, if you have issues with your domain, you can move it to another subdomain
00:41:55without having it affect every single sending on your whole platform.
00:41:59So there's like some basic things like that.
00:42:01Making sure you include unsubscribe links and things that are not
00:42:03merely like transactional emails around like auth or things like that.
00:42:06So there's a lot of guidance there.
00:42:08But every time an email comes in, your email inbox,
00:42:12the email inbox that's receiving that Gmail, Yahoo, Outlook, whatever,
00:42:16they're looking at that using their own proprietary set of like filters.
00:42:20So they don't tell anyone what that is, because if they did,
00:42:22then all we would get is spam email.
00:42:24So they're trying to they're trying to read a lot of signals
00:42:27that has everything to do with how often you're sending.
00:42:30There's some authentication things like DMARC and SPF and DKIM,
00:42:35which that's that's a whole set of acronyms I just threw out there.
00:42:38So like I said, there's a lot of weird complexity around email.
00:42:42But generally speaking, that mindset will really help developers to think,
00:42:45okay, do I look like a trusted sender?
00:42:47Am I sending to people who have already opted in to receive this?
00:42:50It's like we don't allow cold emails on resend.
00:42:53You have to have actual consent to send to people.
00:42:57So there's a lot of basics like that, that if you follow the basics,
00:42:59you send to people who only want to receive it.
00:43:01You send in a trusted way, not just dumping thousands of emails
00:43:05immediately like that.
00:43:07And then you pay attention to basic best practices.
00:43:10Like there's not like specific words not to use, but generally speaking,
00:43:14like your link should be the same as the sending domain.
00:43:17Your images should be the same as the sending domain.
00:43:19Don't use like spammy type words.
00:43:23And a lot of times what you'll find is that Gmail in particular will tell you,
00:43:25hey, this isn't spam for this reason.
00:43:27And you can adjust the content and see how that affects it.
00:43:30When you send with resend, we also show deliverability insights for every email you send.
00:43:34And that's in part to educate you.
00:43:36So every email you send, you can look at the emails table,
00:43:39look at that particular email, and it gives you insights of like,
00:43:41hey, make sure you're doing all of these best practices.
00:43:44So generally speaking, if you think like that, hey, I'm going to send in a way that's trustworthy,
00:43:49where people want to receive this, they've opted in already.
00:43:52I'm sending in a predictable pattern.
00:43:53Those are really the big mental things you have to think through.
00:43:57And then there's a lot of particulars that we can help you figure out as well.
00:44:00So that's as brief of any answers I can give about spam filters and deliverability and
00:44:05IP and domain reputation.
00:44:06Hopefully that was intelligible with all those acronyms.
00:44:10I think that's amazing.
00:44:11It's an amazing amount of wealth knowledge you have about emails and it makes sense.
00:44:15I was going to ask you earlier to sort of explain how resend makes emails easier for developers.
00:44:20And I think just that question alone explains how many hurdles you have to go through when
00:44:24you want to set up emails.
00:44:26Yeah.
00:44:27I remember trying to do it myself and I jumped straight for React email when it first came out
00:44:31and then resend as well, because there were so many headaches I didn't realize email had.
00:44:35And it's, I also like self-hosting and like the one thing you're told never to self-host is emails.
00:44:41Because it's like, you will not have a fun time.
00:44:43Yeah.
00:44:43Yeah.
00:44:44Welcome to being an email service provider.
00:44:46It's, it's, it's got its own set of challenges.
00:44:49Yeah.
00:44:50I think like we started with React email, which I think is such a cool origin story.
00:44:53Like Zeno and Boo, the two of the co-founders were just like, they were having to,
00:44:58if you've ever written email HTML by hand, it's like, what century is this from?
00:45:02And you know, you're writing tables and floats and they were like this, there's gotta be something
00:45:07better out there.
00:45:07And there, there are some things, but they didn't really have the modern touch.
00:45:10And so I think the story of like, they just said like, let's build something that you can actually
00:45:14use TypeScript and React and Tailwind.
00:45:16And, and so that was React email.
00:45:18And then out of that came Resend because they were like, the sending actually isn't that good
00:45:21either.
00:45:22So let's work on that.
00:45:23But there's so much peculiarities around email.
00:45:26We spend so much of our time actually protecting our users by cutting out spam.
00:45:31People trying to send that through our platform.
00:45:33So we have a whole team just dedicated to trust and safety.
00:45:36But once, once you start sending email at scale, you have to either be an email
00:45:39expert or you have to have an email expert.
00:45:41And like, I think one of the things that makes Resend special, and I really think this is core
00:45:45to who we are, and I hope we always keep this is like, we view support as part of DX.
00:45:51Like, it doesn't matter how nice the API is.
00:45:54If when you have issues and you reach out, there's nobody there.
00:45:57And like, we, you have to have an email expert and we want to be that for, for people.
00:46:00So like, you get a shared Slack channel if you're a recent customer of any paying tier at all.
00:46:06And we're, we're very quick to respond to people.
00:46:09And of course, like, we want to get better at that.
00:46:11That's always something that, that is like a stretch goal.
00:46:14But we view like, when you have trouble, you should come to us and we should be able to
00:46:18jump in with you.
00:46:19Like we'll jump into Google Postmaster tools with you.
00:46:22We'll jump into deliverability stuff with you and help you debug it live.
00:46:25And so that's really important to us that DX includes support.
00:46:31And so I hope people feel that way, like taking care of when they come to Resend.
00:46:35And like, we have your back and like, you don't have to be the expert because you have one.
00:46:39And we want to keep doing that for our customers.
00:46:41You know, my, my first experience with email was I think freelance, I was doing some freelance
00:46:45work and he was like, can you add emails?
00:46:47And I naively thought like, oh, emails must be a solved thing.
00:46:50We've had these for years.
00:46:51Like it can't be that hard.
00:46:53And then, yeah, that's when I went down the rabbit hole of, oh my God, this is insane.
00:46:57Just the compatibility between the clients.
00:46:59Like what do we have a few browser engines and hundreds of email clients?
00:47:03And like, there's no agreed upon standard.
00:47:05And I don't know that we'll ever have one because it's really up to the big players to decide that.
00:47:11And that alone, just trying to get stuff to look the same across clients is difficult.
00:47:15And React email now has this whole like linter and compatibility checker and preview.
00:47:19And we want to continue to kind of iterate on it to make it easier for people to know, like,
00:47:24I can send this with confidence.
00:47:26And like Claude can just spin up HTML email, but like, you don't know if it's going to look good.
00:47:30It might look good in, you know, your web browser, but that doesn't mean anything
00:47:33because every client is so different.
00:47:35So I think that's like, there's a lot of little sneaky foot guns in email.
00:47:40And so we're trying to make those easy for people.
00:47:42But it is just kind of a legacy industry with a bunch of stuff tacked on.
00:47:46And so by nature of what it is, you know, it has its complexity.
00:47:51So you kind of need someone who can hopefully hold your hand and guide you in a helpful way.
00:47:55So most people we have on here that do video have a kind of progression arc of where they started,
00:48:02the equipment they got and what they replaced it with.
00:48:04So it'd be good to hear yours.
00:48:07Yeah, I mean, it started with me just with a webcam.
00:48:11Actually, I didn't even I didn't do video at all for a long time.
00:48:14When I worked for the company I used to work for, I recorded like voiceovers.
00:48:19And so I got this mic pretty early on, actually, which is I'm super grateful for it's an RE20.
00:48:26And so the audio was the most important thing to me.
00:48:30And then several years into doing my channel, I finally did like a face reveal.
00:48:35It wasn't intentional, like a big thing.
00:48:37I just recorded myself.
00:48:38When I got a webcam, I got like a Insta 360.
00:48:42And so I used that for several years.
00:48:44And then when I joined Resend, you know, six, seven months in, they were like,
00:48:49let's really invest in video so that it's really easy for you to just jump straight in,
00:48:53no matter what time of day it is, whatever's going on, it always looks the same.
00:48:56So we worked with the company that designed this whole set.
00:49:00It's actually in my office in my basement.
00:49:03And I switched my office.
00:49:04If anybody watched earlier videos, I used to face this way.
00:49:07Now I face this way.
00:49:09This whole wall I painted and hung the shelves and did all this kind of stuff.
00:49:12And then the camera itself is a Panasonic or yeah, Lumix S5 II, I think, something like that.
00:49:22I don't remember the lens.
00:49:24I'm not that much of a camera junkie, whatever they told me to get.
00:49:27That's what I got.
00:49:28I think 35 millimeter maybe.
00:49:30But I've got a teleprompter, two screens.
00:49:33And then the big help was like all this lighting setup.
00:49:36I actually have it on buttons here where I can just like turn off my lights completely.
00:49:40And so it just turns on once and then everything's set up.
00:49:43So it makes it to where I can sit down, literally touch two buttons under my desk,
00:49:48hit record and go.
00:49:49And so, yeah, I'm really grateful for the investment that Zeno
00:49:54put into the studio.
00:49:55And it makes it really easy to record stuff for a reason.
00:49:58And then like I have this little computer over here, which is an actual SC30 that is
00:50:02operational that I bought in California and flew back with on the flight,
00:50:06like stuffed under my feet and hit it with my jacket because it didn't really fit.
00:50:10But I got there.
00:50:11So that was really fun to get as well.
00:50:13So I've kind of been adding to the set as I've had a chance to pick up stuff.
00:50:18And I don't have any stretch goals of what I want to add next,
00:50:20but it's been fun to kind of play with the backdrop as well.
00:50:24Can you talk a bit more about your lights?
00:50:25I think all of us here know a bit about recording and stuff.
00:50:28So I'd love to know positions and like what kind of lights.
00:50:31Yeah.
00:50:32Yeah.
00:50:32I don't know what the names of the things are.
00:50:34I'm sure I knew when they told me to set them up.
00:50:36But this is like an Ameren, I don't remember what,
00:50:38but it's probably like three feet across, something like that.
00:50:40Um, so this is kind of the main big one.
00:50:43I've got one spotlight that goes up in the ceiling.
00:50:45I've got another one behind my head that points down.
00:50:48And then I've got this little like ambient light to the side.
00:50:51Um, and so, yeah, it's not overly bright the way I've got it set up.
00:50:55Um, so the camera does a little bit more works that I'm not baking.
00:50:58When we first set it up, um, we had it all like five times brighter.
00:51:02And I'm like, I'm sweating just sitting here.
00:51:04There's no way I'm going to be able to work like this.
00:51:06And so they're very, they're quite dim in here.
00:51:09Um, but it allows the camera to kind of do some, a little bit more of the heavy lifting.
00:51:13But yeah, so it's those four.
00:51:14So this main big one, the ceiling light that kind of, I think just provides ambience or something.
00:51:20Um, this one, that's supposed to fully like the hair light and then, uh, the ambient light.
00:51:25You know, I think when I first started making videos, I had a light,
00:51:28like genuinely only like a feet away from my face on full brightness.
00:51:31And it would cook me when I was filming videos.
00:51:35It was, yeah, it was not a fun time.
00:51:37Yeah.
00:51:37Yeah.
00:51:38Um, and yeah, learning about a hair light was fascinating.
00:51:41Slowly adding pieces on.
00:51:42It starts with like one light and then it goes around and you're like,
00:51:44I need a hair light apparently to light up my hair.
00:51:48Yeah.
00:51:49It's a weird rabbit hole to go into.
00:51:51Right.
00:51:52Yeah.
00:51:52I just basically, whatever the company like told me to do, I just did that.
00:51:56So I don't really understand how it works.
00:51:57So I feel like I went from like zero to this, um, but I'm not complaining either.
00:52:01Yeah.
00:52:02It sounds like a really cool company.
00:52:03I think, uh, we need to reach out to them and see what they can do for us.
00:52:06But yeah, I've got a few lights.
00:52:08I don't have any Amerinds.
00:52:09I've heard a lot about Amerinds and how good they are.
00:52:10I've got some Legato key lights, which are working well for me.
00:52:14And I also use a Lumix Panasonic, but I think it's a bit of an older model to yours.
00:52:18But yeah, it works well and I've got a nice lens.
00:52:21Yeah.
00:52:21I, I really liked the, the way that Panasonic handles videos and they were a little cheaper
00:52:26overall as well.
00:52:27So I think that was my main motivation moving that direction.
00:52:30Does it last for your live streams?
00:52:32Do you, cause I know a lot of them overheat.
00:52:34It does a pretty good job so far.
00:52:36I feel like my cam link, uh, that I've got it all connected to has started to have issues.
00:52:41Uh, hopefully won't during our call today, but, um, so I've had some issues with that in the past,
00:52:47but I've also just like killed the camera during my stream.
00:52:50Nobody's there to see me, you know, like, so it's, it's okay.
00:52:53Um, but, uh, yeah, for the most part it's been fine, but I've started to have some issues.
00:52:58Um, so I may look into alternate things cause I think it might be the cam link that's actually
00:53:02causing issues.
00:53:03Um, cause I'll replug it in and then it works.
00:53:06So, yeah, but I think the, the camera itself has been fine.
00:53:09I had to replace my cam link the other day cause I'm pretty sure it was breaking as well.
00:53:13Um, I did end up just getting another like cam link from another company.
00:53:16So I don't know what the best solution is, but yeah, it might be the cam link.
00:53:20Yeah.
00:53:20It's like, you know how you make these mental notes.
00:53:22I was listening to Syntax, um, the podcast and they mentioned like,
00:53:27they were having issues with their cam like they got something else.
00:53:30So I don't know what it is, but I know that when I have issues eventually like long enough with it,
00:53:34I will just go search on their website and buy whatever they recommended because
00:53:37it seems like it's a common issue.
00:53:39Yeah.
00:53:39And it seems like for launch week, so I've watched a few of the recent launch week videos and
00:53:44they're, they're really well done.
00:53:45Um, was it a company, the same company that made those ones or a different company or
00:53:49how did that work?
00:53:50Yeah.
00:53:51So it was, we were in, we are at an offsite, um, actually in Barcelona together
00:53:57two, three of launch weeks ago.
00:53:59And I just said, what if we just recorded this ourselves?
00:54:01Um, so not this one or the one before, but the, the, the three back, we recorded that ourselves
00:54:07and it was like a lot more difficult than we thought.
00:54:10Um, in part because we were in this, uh, like conference room kind of thing.
00:54:13And there was a lunch happening downstairs.
00:54:15So there's like all this ambient audio and you know, we were very, very like just kind of home,
00:54:20uh, spun.
00:54:21And, uh, so we're like, that, that was harder than we thought.
00:54:23Let's see who we can connect with.
00:54:25So we connected with, I think it's open light studios, if I'm remembering correctly in SF.
00:54:31Um, and so what we've done is we went and recorded there, uh, both this week and last week.
00:54:35Um, and they give us the kind of the final cut, um, which we can, you know, ask for alterations
00:54:40on.
00:54:40And then I basically take it from there and do all the, uh, editing past there.
00:54:44Um, so that's been a learning experience, a lot of stuff that I'm not used to doing.
00:54:48Um, and then this time we contracted with somebody to kind of do the music and sound effects, um,
00:54:55touches afterwards as well, um, which added like a level of class that I don't think I could do.
00:55:01And so, uh, that was really helpful as well.
00:55:03So we've kind of slowly built on that.
00:55:05Uh, so first time we recorded it last time, uh, they recorded it.
00:55:08I did all the editing and this last launch week, just a couple of weeks ago,
00:55:12I did most of the editing.
00:55:13We had kind of a final editor come and add the music and the sound effects, uh,
00:55:17and a few small details like that.
00:55:19But I, I still did most of the actual on-screen stuff.
00:55:21Nice.
00:55:22I think the videos are really good.
00:55:23I like the transitions.
00:55:24I'm sure you had multiple cameras and you've got like these nice bubbles or,
00:55:28and text to explain what people are saying.
00:55:30Yeah.
00:55:30What, what did you use for that?
00:55:32Uh, so this last time we actually went slightly lazier, um, and I just used Figma
00:55:36and we basically just exported it to where like it was the bubble effect in Figma.
00:55:41And I made sure that it kind of lined up in the video.
00:55:43The time before that, I actually built this custom animation.
00:55:46If you watch the two, not that you need to do this, but, um, the time before we had them,
00:55:50it was actually inside of DaVinci Resolve.
00:55:53And I, I went on a rabbit hole, um, for like five hours and built this custom bubble that
00:55:59interacted in like refracted light and stuff.
00:56:01And, but it was just too much effort.
00:56:03And we're like, especially this time around, there was too much stuff going on.
00:56:06I'm like, I don't think I've got the capacity to mentally do that.
00:56:09And it was very hacked.
00:56:10Like I didn't, I'm sure anybody who knows DaVinci Resolve would say like,
00:56:13what are you doing with fusion?
00:56:14Like, you know, you take this away from this person immediately.
00:56:17But, um, yeah, so that's, this time it was very basic, but last time it was pretty involved
00:56:23and pretty fun to, to work on, but unfortunately I couldn't make it this time.
00:56:26I didn't even know you could do animations in Figma, so I'll need to look into it.
00:56:29Yeah.
00:56:30I think it's just like the, the glass effect in, in Figma, uh, like the bubble itself.
00:56:34And then I exported it over to DaVinci Resolve and did all the animations inside of DaVinci Resolve.
00:56:39Sorry.
00:56:39Yeah.
00:56:40Just, uh, the final one we sometimes ask is if you have any sort of spicy tech takes,
00:56:45anything that might, might be a good clip, basically.
00:56:47I think it's funny that React server components have been around such a long time
00:56:52and they still are in this kind of ambient zone.
00:56:55It doesn't help that we just had a bunch of security that are constantly coming out about these.
00:57:01So I think for me, like, I don't know that I'm a very hot takes type of person, but
00:57:06the thing that like, at least architecturally has struggled, like there's two things with
00:57:10React server components.
00:57:11There's clearly like the way it was implemented.
00:57:13There's a lot of security issues with it.
00:57:15But secondly, like just the mental model, um, Dean Abramoff did a post last year
00:57:21explaining React server components using Astro.
00:57:24And I'm like, anytime you have to reach to an entirely different framework to explain
00:57:28how yours works, that tells me like mentally, this was not prepared in a really helpful way.
00:57:32And I'm, I know the people at React are way smarter than I, uh, you know, will ever be.
00:57:36But I think for me, like if you like for me, every site should default to Astro, that's, that's my hot take.
00:57:45Um, and I, I just really think most of the time people are not building full on a web apps and Astro can do just about everything you need for most basic websites.
00:57:55Of course, I'm not talking about apps, but even with some basic apps, you can really get most of the way there with Astro and you get all the beauty of like content collections
00:58:03where you get typesafe markdown, you've got every rendering pattern you basically want, you've got full support for fonts and images and like, there's just so
00:58:12much baked into the system.
00:58:14That's really basic when you need it to be, uh, everything is like logically, like mentally is a clear separation of layers.
00:58:21Like what's server, what's client, like that's all the complexity that came with React server components of figuring out where this lives doesn't exist in Astro.
00:58:29Cause there's a very clear delineation.
00:58:30So my hot take is basically everyone should default to Astro.
00:58:34And I think one of the, the, one of the ways that's shown is through like these meta frameworks, trying to implement, uh, RSC in a way that clearly is not working both on a
00:58:43technical like security side and a mental model side for a lot of people.
00:58:47But I also, I'm just an Astro fan boy.
00:58:49So I guess, what did you expect?
00:58:52Yeah, no, I'm not sure I can argue against your points there.
00:58:54Cause I use Astro for my personal site as well and, and love Astro and yeah, React server components for
00:59:00people that didn't know also just had another vulnerability and like yesterday or today, basically.
00:59:04Literally yesterday.
00:59:05Um, yeah.
00:59:06So, I mean, I might have a video on that coming soon, depending on how technical it is, but, uh, I think
00:59:11I've already done two or three videos on react vulnerabilities this year.
00:59:14It's, uh, server components have gone a bit wild this year.
00:59:18And I think people are starting to sort of, yeah, doubt them because of that.
00:59:21It makes it hard to want to architect a whole lot of your, your app around react server components, but they are super helpful.
00:59:27Like, you know, it's, it's a super helpful pattern, but clearly there's some issues with it.
00:59:32Yeah.
00:59:33I quite like the way Tanstack start recently added server components.
00:59:37Cause that's sort of their own thing.
00:59:38I don't think they're impacted by the vulnerabilities.
00:59:40And, uh, yeah, it's also very explicit that you're making a server component.
00:59:45Whereas next JS is obviously pretty much everything server component first.
00:59:48And then you make a client component instead.
00:59:50Tanner does so much stuff.
00:59:51He actually lives fairly close to me.
00:59:53Um, he's just such, I don't know how he does everything he does.
00:59:56Um, and I think like he approaches stuff in a really like a fundamental, like what's the best way to do this kind of approach that somehow he's unlocked.
01:00:06Um, and I know it's not just him, um, but yeah, I've just always been so impressed with him.
01:00:11And like, I feel like every day I'm like, how are you shipping the amount of things you're shipping?
01:00:15Um, and he's also just the nicest guy in the world.
01:00:17If you ever get a chance to meet him, you'd think he's just like some nobody.
01:00:20Uh, honestly, he just acts like the most normal guy in the world.
01:00:23And you realize like, oh, he's literally rewriting every single thing we've ever done.
01:00:27In the front end world, especially.
01:00:29But, you know, I've been meaning to invite him on the podcast because I'm a huge Tanstack fan.
01:00:33And yeah, they, they have a library that seems to fix every problem in the, the JavaScript ecosystem at the moment.
01:00:38So literally everything.
01:00:40Thank you for listening to this episode of the Better Stack Podcast.
01:00:43Find us wherever you listen to podcasts.
01:00:45So Apple, Spotify, just everywhere.
01:00:48And thank you, Chris, for joining us.
01:00:50It was great to chat to you and talk about AI and everything you work on.
01:00:54And yeah, maybe in the future we could do this again.
01:00:55And talk more about AI and other stuff.
01:00:58But until then, it's a goodbye from me.
01:01:00Goodbye from me.
01:01:01It's a bye from me.
01:01:02It's a goodbye from me.
Community Posts
No posts yet. Be the first to write about this video!
Write about this video