Sweet Hang #5: My Psychosis Moment w/ Jesse Hanley

[SPEAKER_00] You're listening to Mostly Technical, which is brought to you by Lair Jobs, the official level job board and Solo, the workspace for your agents and dev stack. And now, Ian and Aaron. [SPEAKER_01] All right, we are here with a special Sweet Hang episode, Mostly Technical. Welcome to Jesse Hanley. How are you? I'm good. How are you? Very good. So we've been going back and forth on the internet about various topics. And over the last couple of weeks, I feel like especially and we're like, hey, let's jump on. Let's jump on a podcast and and hash it out. Hash out some things. Aaron's not with us tonight. [SPEAKER_01] We thought he was going to make it, but he's getting ready to travel. [SPEAKER_02] He is like a mediator. He needed to be like a mediator. [SPEAKER_01] But then I was thinking it might be better. He might just be dodging bullets the whole night. You know, might be better with him not here taking the heat. So anyway, there's some chance he just pops in. So he could just appear at some point. But otherwise, it'll be you and me, which I think will be great. We have a lot to talk about. We got business stuff. We have AI stuff. We have camera stuff. Yeah, a lot to cover. And I feel like there's lots of your videos. [SPEAKER_02] I feel like there's there's yeah. Do you have an outro list of things to cover? Because I feel like there are so many things we could we could we could start getting into. And there's lots of things like I feel like we agree and disagree on. But because Twitter and X where we mostly go back and forth is so little characters, we're not able to really understand each other. [SPEAKER_03] Yeah, yeah, yeah. [SPEAKER_01] This is why podcasts are the best. This is why we go long form. We locked the people in here with us for an hour, hour and a half. And we settle all family business, as they say on The Godfather. So well, let's do it. Let's settle. All right. I do have a list. I do have a list, but we can go anywhere. But let's go. I don't know. Do we want to go? Let's let's jump in deep into the pool first. You've been lately talking a lot about bot. [SPEAKER_01] And so Grokbot, which if you're not aware is like this thing, X, A.I., or whatever they're called released, which is like a agent desktop app, I believe exclusively. Or maybe there's a web version. But mobile. Yeah, mobile. So let you define bots. I mean, it's kind of stuff we've all seen, but I get the impression that like the packaging is a little bit nicer, a little bit easier to use. So maybe and you've jumped in your both feet, you've already got processes and systems using the bot. So to you, what is the bot? [SPEAKER_01] And then maybe a little bit about how you're using it. And then let's we can we can go from there. I'm sure it'll branch out. [SPEAKER_02] Yeah. So so with but I feel like this is kind of my psychosis moment. You know, everyone kind of has their own. And I feel like but is the first time that it's really resonated. Like like something is really I've gone. Oh, OK. Like things I think are about to change quite significantly. And I feel like for a while, you know, you've got the open claw people and the Hermes people and all these people that have kind of been doing similar stuff. [SPEAKER_02] Right. But there's just kind of like an aesthetic about them that is quite repulsive. I don't know why. Like, I just I just see them posting and I'm just like, I don't think that's for me. And there's too much. [SPEAKER_01] It also seems like a maintenance headache. It's like they're like, this thing is great. It's doing everything. And also, I spent all my time keeping this thing alive and it went off and it destroyed something. You know, there's a lot of like that kind of stuff to it. [SPEAKER_02] Yeah. And I used like an analogy last week where it's kind of like, all right, well, I pick planet scale for my database. I want something like that hosted. There's a lot of critical stuff. I'm not giving something my Gmail that is self hosted on Hetzna. I'm just not like the security risk of that is just so substantial. And it's just it's just something that I I didn't want to do. [SPEAKER_02] So with with Grok, the stuff that is interesting is like you install it, you get it set up, you add all your MCPs. And the way that they've modeled it is basically you create a bot and a bot is analogous to like an employee. Right. So that bot has its own Linux VM. It has its own context. It has its own potentially local skills. So you tell it some things. It can use a web browser so you can be like, hey, go do this and it will log in. [SPEAKER_02] I'll give you like a specific example of something I wouldn't have done in another tool. I have my money with a management company in Australia. They have a very old school portal once a month. I need to get my cash statements from that. So I actually physically log in, grab that and then send that to my bookkeepers to point to zero. Grokbot's really good at that. So I just, you know, I show it how to do it. It does it. And that task is done. And I would have probably got like I was doing it myself. [SPEAKER_02] So that was like human time or I probably could have got someone on my team to do it. But yeah. And then over the past like week, there's been a lot of those little tasks where it's been something that I've done or something that Anya on my team has done. Or and I've just taught bot how to do it. And it's done all those tasks. Right. And it's super reliable and it feels really good. And so I don't know. I just really like it. But it's touching all areas. [SPEAKER_02] I think this is the big thing. It's not just dev. Like, it's not just dev. It's touching like even our support, the way that support was handled was like we had AI back and forth for the first year of support. And then it would get escalated to me usually. And then now I'm training bot to handle all those. And I'm even getting it to close things just from my mobile phone. Right. Using the admin MCP, which I got that idea from you. So there's just there's workflow changes that is is just kind of changing. [SPEAKER_02] It's even this is a tangential thing. We can go off on this later, but it's even making me like slightly rethink products, for example. And like with help spot and also like bento chat, you're building AI features. Right. And I imagine those AI features are ticket comes in. AI use a whole bunch of tool calls and then gives a great reply back. Right. So kind of the fin, fin esque thing. [SPEAKER_02] But then now it's like with bot, you can get it to do a routine where it manually checks, you know, let's say a help spot MCP every hour and does all that stuff and then does all the replies. Like you can, you can do that now. So then you're like, all right, well, are people in the future going to want to bring growth bot to these tools and just interact with them and use them basically as management systems? Or is there a lot of value in having your own agents do a lot of that? [SPEAKER_02] And it's just made me rethink a lot. And like some customers, yes, they're going to want to bring their growth bots, but then some customers, they're not going to want to do that. And so you need to offer both. But it does change how you think about building them. [SPEAKER_01] I don't know. I've just gone on a tangent with that, but yeah, no, I think that's like, there's like three great things for us to dive into there. So like we should mention you're, you're the founder of Bento. So everybody's heard about Bento on this feed from the first day. So thank you, of course, for sponsoring the show and believing in it. I love this show. [SPEAKER_02] It's my favorite show. [SPEAKER_01] This is not a paid ad. This is you're here to hang as a founder. So anyway, yeah. So let's take that kind of phase one. Let's put the business part aside for a second. So just you're a dude using your bot, right? [SPEAKER_01] One of the things I've seen from this, I think one of the big differences from the outside, I haven't tried it myself yet, is that like it seems quite good at what you just described, which is use the browser, I assume on its end somewhere on its own computer to go do stuff. And so like, even if there's not an MCP, whatever, it's just like pointing out a website, tell it to go wander around, log in, do the stuff, get your statements, do whatever, right? [SPEAKER_01] So like this has opened it up to people more than just, yes, the things you use that have MCPs. And even MCPs are kind of annoying in some ways, like for someone less technical, just, hey, go to this URL and log in as me is like much simpler, even conceptually than the MCP, right? Even though it's not that complicated. So it's that kind of, it sounds like that's what you're finding a little bit. It's like you can just, it's a little more flexible. [SPEAKER_02] Yeah. A friend messaged me, like an e-com friend actually messaged me. And something that I told him was that don't come into this. If you are one of those users that is like a chat GPT pontificator where you just talk to them all day. And that's not what, that's not what bot is, but bots are like action orientated because the back end of it, it's like the Grokbot naming is horrible. And so we're going to completely move away from that. It is a wrapper on cursor cloud agents, right? [SPEAKER_02] And what cursor cloud agents was, was just like every GitHub repo had its own VM. And then now they've just changed that to every agent has its own VM and that persists. Yeah. So that's, that's the main change. And because of that cursor is very action orientated. It just wants to do code. It wants to write code. It wants to find stuff and do stuff. And so Grokbot is just applying that. And so it's always over eager to do stuff. Um, where, you know, if you're like, look at my tickets, it'll actually want to respond to them. [SPEAKER_02] So you have to tell it, don't respond, like stop and, you know, drop some stuff. And then I'll tell you to go ahead, but it's very action orientated, which for all tasks, all human tasks is very good at. So if you're like, do this thing, it will try its best to do it. We'll try to do it. Yeah. Yeah. But you're not going to talk to it like a human, like strategy stuff. [SPEAKER_01] Right. It's not really set up for like what we, what is traditional of two years of like, here's an empty text box. Let's have a chat back and forth. That's going to be long winded and not necessarily inclined to do anything unless you are explicit about telling it to do something. This has got sort of the reverse, which is, that's like one of those things that's quite interesting, right? [SPEAKER_01] It's like just this like little tweak conceptually that's not, you know, hardly anything on like whether it's in the system prompt that does it or some, whatever the harness is adding in there. Like it's minor, but like for the usability of it, it's actually quite different and opens up some different doors of how people might approach it or use it or just that initial vibe you get from something that maybe changes your feeling. Like it seems like it has for you. Um, I'm always fascinated by those little things in software that are just like, we just, we just took this thing. [SPEAKER_01] We made it 3% different. And that's like what really opens the door, you know? [SPEAKER_02] Yeah. Yeah. And then, so like the, the outcome of doing a lot of this stuff, right, is like, as I have automated more and more Bento. So like I mentioned in the previous past, like Bento is low seven figure in Rev. It is just me and Anya. So very, very tight. Um, but Anya's workload is like freed up so much. So like now I'm having conversations with her where I'm like, like our support is honestly mostly automated now. Um, and it's, and it's good. It's good support. [SPEAKER_02] Like I had, um, Tyler from less annoying CRM do like a little, he spoke about in his podcast about the support that he got, which felt highly technical, but it was all automated. None of it was, was us, but it was a highly tailored personal experience for him, uh, enough that he spoke about it. So I, I feel like these tools are now good enough where you can deliver amazing experiences. And then now your attention is pulled elsewhere. So it's like, all right, well, I'll give more personal support to people using my time. [SPEAKER_02] And then Anya, I think we're going to probably shift her into until cursor does like a multiplayer mode where I can get her to like manage bots with me. Um, we're probably going to start doing more sales stuff. [SPEAKER_01] Um, but I, so right now it's just, it's your personal bots. It's not, there's no team structure to them or permissioning or all that stuff. [SPEAKER_02] Yeah. But I'm talking into the cursor team. They've added me to like a friends of Grokbot and they're working on a share feature. So, so, so you can do, you can do like, let's say I make a bot. You could potentially, I could invite you into that and we could collab. An example would be, let's say I hire a copywriter. So I hire a copywriter. They write a whole bunch of content. They give that to a dev bot who will then go actually implement on the marketing website and then send the preview to both of us being like, how does this look? And you say, yeah, it looks good merge. [SPEAKER_02] So that's how I think things will collaborate kind of moving forward is you'll invite people into these group chats. They've got like a group chat thing. So right now you can actually have chats in multiple bots. You'll probably have humans in there also chatting and then everyone's kind of delegating each other work. So I feel like that's how things will change. Yeah. [SPEAKER_01] I mean, it's interesting to think about like, I think one of the things where we differ a little on the bots is like how much, like how much of an actual change there's going to be in like kind of corporations, human behavior. You know, I think like companies are size to me, you're not like representative, right? Like you were already trying to stay small. You're two people. You were already doing even way more than I am of like trying to optimize for your smallness and like, you know. [SPEAKER_01] You know, just automating things even pre-AI and definitely even very early with AI, right? Like you're big into like how can I automate something? How can I make it repeatable? How can I have a video? Your videos everywhere in Bento, which is like a different kind of thing. You don't see everywhere. Like there's a lot of like that kind of optimization, which is not something you see everywhere. So like I just don't, I don't know that that's how everybody's going to approach it. So you seem to feel like bot opened up to you the idea that like this is how everybody's going to work a little. [SPEAKER_01] Well, before we were on air, you were giving the example of like, you know, our job is going to go away because they're going to automate these things with bots. So, I don't know, let's get into that whole sea of stuff. [SPEAKER_02] So do you remember in like 2015 to like 2019, like pre the Pandy? Do you remember like SOPs? Everyone was like always talking about SOP documents. Like everyone's got to write SOPs for every single job that everyone does, right? Right. And especially in like the spaces that I hang out with, which is like marketers and people that were running like little indie businesses traveling around and stuff. [SPEAKER_02] Everyone was obsessed with SOPs, whereas like you hire someone, they do their procedure documents step by step and they do that every day and everything was repeatable. So if someone left, you could, you could move them in. Those jobs, you can't argue that Grot can't do almost all of those. Mm hmm. And that represents a very large. [SPEAKER_01] I can't argue that actually. I can't argue that a little bit. I can't argue that a little bit. Because I think there are, there's of course certain jobs that are like that, right? But in knowledge work, there's not, there's not really that many jobs that are like, this is literally a list of automatable things and that's it. There's not none of them, but there's not, because there's still elements of like meetings. There's still elements of like different aspects of responsibility. There's still aspects of, I mean, I've talked about this on the podcast before. [SPEAKER_01] There's still aspects of like, I don't want to be the manager who's like, oh, sorry guys, my fleet of bots went south. I'll take the blame for my fleet of bots. No, I don't. I want a person between me and the fleet of bots that's responsible for the fleet of bots. Right. And so that means that there's more people in all these rungs because they, those people want people below them who are like, well, I don't want to be the one that the bots that come down on me when the bots screw up. Like, so there's like aspects to like the human nature of this that I don't think are so easily like the bot just did the work. [SPEAKER_01] It was steps one through five. And that's it. Again, not no, no job like that. But I think when you really dig into a lot of jobs, they're just not that clean. Like maybe there's like 30% of the job that's that clean or 50% of the job that's that clean, but the other 50% isn't that clean. And so it's not so easy to just say, oh yeah, these people are all gone. Like maybe there's less of them. Maybe they do different things, whatever. But it's not, I don't know. It's gonna be so easy to totally wipe them out. [SPEAKER_02] Well, I mean, it's kind of like the lawyer thing, right? It's like you still need a lawyer to look over your ducks because the lawyer is the one who is accountable. Lawyer is perfect. Right? [SPEAKER_01] Lawyer is the most automatable thing in the world. I think a lawyer is actually more automatable than most of what we're discussing, like in terms of like jobs, even in a software company. I feel like a lawyer is like, just look up the fricking law and spit out what the thing that happens is, right? Like it should be the most automatable. And I don't believe for one second that anybody's gonna be like, oh yeah, I got my fully AI lawyer and that's all I use. And that's it. [SPEAKER_01] Like again, of course, like internet people, people on Twitter. Yes, they're just going to the AI and they're writing their legal documents. I have no doubt about that. But that's not the 99.8% of commerce is not gonna act that way about their lawyers. So that's like a perfect example to me. The lawyer. How do you replace that lawyer? [SPEAKER_02] Yeah. Okay. Well, I'm not gonna replace the lawyer, but even that with like, I've hired lawyers to do my stuff and then I've got an AI to find their mistakes and then I've sent it back to the lawyer. So like, I'm pro lawyers. And I'm pro AI checking my lawyers work. But I think that there is some very hard maths with like risk tolerance and your willingness to like implement these things. Right. [SPEAKER_02] And so for me, it's like, I'm comfortable giving AI access to the PlanetScale MCP on read only. I'm not comfortable giving it access to, you know, right access to, you know, do whatever it wants. So there's, but some people are, it's like, some people are happy to do that. I'm what's it? What's something? What's something? [SPEAKER_01] Hold on one second. I'm gonna keep going. I'm gonna open my door here. Cause it's getting, it's getting a little hot in this little office. [SPEAKER_02] Well, I guess like a counterpoint, a counterpoint that I want to post to you. Um, and I don't want everyone to be fired. Okay. I'm not, I'm not, I just want to make that really clear. [SPEAKER_01] It's about, it's about if that things happen, whatever. People used to be ride horses to deliver the mail, you know what I mean? Like shit changes. But is that, is this one of those moments? Maybe it is, but maybe it isn't. [SPEAKER_02] Well, it is, I think for like, and you're right. I think there's a, there's classes of companies. The larger companies may be a lot slower. Um, but the SMB, the online SMB, I think, yeah, a lot of jobs are at risk because the, the smaller SMB is price conscious. They're looking at their support costs and like, all right, I'm making, let's say as an example, I'm making 50 grand a month. I have three support people. Each one costs me, I don't know, five or one or eight grand. [SPEAKER_02] Whatever, whatever. Yeah. Right. 150 grand a year. It is, if you get that down to one person who becomes more efficient, delivers better support, actually has now career progression because they go from a CS person to a swarm bot manager and they're delivering amazing customer experience that is above and beyond because they're able to elevate themselves. [SPEAKER_02] Um, that, that's where I kind of think things are more headed where, uh, things will get automated. Stuff, stuff will, will be reduced and key performers will be elevated to positions where they can actually like thrive. I think that's like where things are naturally headed. And at least amongst like my friends, that's kind of what people are talking about. People are talking about, I had a DM over the weekend with a friend who's, uh, downsizing one dev and downsizing a support person. [SPEAKER_02] He's going to save like 10 grand a month. And for his small business, that is, it's a lot of money for him. Right. [SPEAKER_01] So yeah, I, I agree. I agree with that. So I agree with small businesses for sure. Um, at least technically enabled ones. Right. Like, I don't know if I think like the local bodega is anytime soon going to AI themselves out of extra employees. Like, um, no longer term, possibly. Right. But they're what? They're doing everything themselves anyway. Right. They're doing the procurement. [SPEAKER_02] They're doing their orders. They're making sure that the sweeties are full. [SPEAKER_01] Right. And so I think when you have something though, like a, like what you're talking about with that example, I think that there's an interesting grouping there of like, they're sort of the perfect case because it's like, well, they could, they couldn't even afford the best support people to begin with. Right. So they're probably already at people who are maybe like not as skilled as somebody else. Right. So they, so the AI now has a lower bar to replace that person. So like, that's like a baseline. Like it's not such a high bar has to achieve in terms of ability. [SPEAKER_01] Um, also there per like transaction risk is lower. So like when you have like a big company, it's like if this AI screws up and costs us $50 million and if we had just had this person, they're making 70 grand, I would have avoided that. That's much harder sell than like a small business. It's like, Oh, our average order is $200. Right. Okay. [SPEAKER_02] I'm going to push back on that. Humans cause damage of that size too. For example, in our space, in both yours and my space, right? Holding customer data. Uh, I won't name them. Two very large competitors have massive data breaches in the last three to four years. They were caused by humans getting bribed in Bitcoin, dumping all the data. Right. Like, like literally a CS person going, Oh, okay. This looks nice. I'm going to get a million dollars. I'm going to go export all these companies data. Like, like that's easy enough. Right. [SPEAKER_02] Um, is my AI going to get, you know, is my AI going to get, uh, convinced? [SPEAKER_01] Maybe it's not going to get bribed, but it might get hacked. So like, yeah, I agree with that. I think that's where like, to me, the bigger companies, I think that that's the exact reason why we might see less of the job pure replacement because it's like, I want those two together. Like I want the human and everything the human brains, but I want the AI that's like watching over the human checking for easy to find mistakes that the human had a bad day and made a mistake. Obviously like security, all those types of things. [SPEAKER_01] Now you've got more elaborate security checks and things like that. You have somebody checking every log all the time to see that Bobby downloaded a bunch of data that doesn't belong to him. Right. Like all those automations to kind of check the checkers of the humans, I think are going to make loads of money. Um, and that you want the, the bet you have the resources to have the best of both worlds. And I think there will be companies that want that. Now they're definitely going to be people want to save money. There'll be, I think the, it's like the middle outcomes, the most likely, right? [SPEAKER_01] We're like, you're going to downsize. You had 10 people in a department and now you have eight. Right. Because like, and you, you brought, you brought the AI or you have seven and you brought the AI in, or maybe you still have 10, but now three of them do like you're talking about, like three of them are doing more different kinds of work. That's more like human, uh, intensive verse, like things we could automate of their jobs. So I don't know. There's obviously going to be people who are going all in on it. Right. Like I'm obviously in our world, we see tons of that. [SPEAKER_01] Like, yeah, I don't have any employees and I've AI and everything. Right. Like that's fine. [SPEAKER_02] Um, what about you? What, what, what about you? Like, you know, my stance, I'm not hiring anyone. What about you? Are you, are you, have you stopped? [SPEAKER_01] Because I knew you did a recent hire, but like, yeah, we just hired a support person. Um, yeah. Like I couldn't automate almost any of my support. Like literally we are like near 0%. Like, honestly, like, it's just not, not literally zero, but like, there's so many that are unique elements. Um, yeah, we just don't get a lot of like, Hey, how do I use feature X? Like we like, don't get that. [SPEAKER_02] Well, neither do I, neither do I, but mine's automated. So like, like I'm getting, Hey, I'm currently getting an attacked and spammed, like help. I need instant help. So I'm getting, I'm getting very niche tickets. I'm getting niche DNS tickets. I'm getting a whole bunch of whack tickets. So I'm, I'm very, I disagree. I disagree. I disagree that you can't automate your support. [SPEAKER_01] I mean, I guess there's the element of like, so our biggest support, like burdens are things like on premise customers who are like messed up their windows installation and messed up then their help spot installation because they messed up their windows installation. So like, you know, I guess to me it's like, could we just have the AI? Could we say, Hey, you got to go deal with this AI on your issue. We could, and the AI would do what it does. I mean, we've all used AI to debug server type issues, right? [SPEAKER_01] Um, now if you could put the AI on the server, that's different. That to me is more interesting because then it'll root around and find stuff, right? And fix it. But when you're just having the chat with the AI about, Oh, I tried this. Oh, I tried that. And the other person is going to know it's obviously just an AI chat that's getting just returning generic windows information. Like I think they don't. If it's good, if it's, if it's, if it's good, they don't. [SPEAKER_02] I'm sorry. I'm sorry. I'm sorry. If, if, if the AI is good and look, this is like also an advertisement for help spot AI eventually when it releases, I'm sure it'll be fantastic. The premise there is that like, if the AI has the docs, you have years and years of context of close resolutions and threads. There's like, in no doubt in my mind, can that be all read, summarized into new private docs that the doc that the AI has access to. [SPEAKER_02] You've solved windows issues probably a hundred times. So it can do that. And then you make the AI very like escalation focused. So if it does struggle, it just goes, Hey, routing you to such and such in support. And it, and it, and it does. Right. So I, I just pushed back on the notion. Like you have so much unique, valuable information and so much so that like companies are paying tens of millions of dollars for people's private data. [SPEAKER_01] Where do I sell mine? Let's go. Come on. [SPEAKER_02] Maybe we can do a package deal. My data and yours. Yeah. Yeah. We can retire. [SPEAKER_01] Yeah. Um, yes. I know. I think there is that, I mean, I think we've done various, like, even when we've handled issues as humans, like searching back through our own history, the number of times we see the same issue is surprisingly low sometimes, but I do agree. Like we do a lot of data. Um, it will be used in that way you're describing. I think that I, even though even me though, I feel like it's, we're borderline on. So we're going to have the help side AI. We're going to use ourselves. Obviously we're going to, we're going to see how much of it can take, but I'm a little bit. [SPEAKER_01] Um, I don't know. I'm a little bit like our dollar value per transaction is enough that I feel like I don't know what I'm gaining by automating it. Like I'm not gaining that much. Now I like the idea of like gaining. I like the idea of the first shot. AI attempt. That's like, right? Like maybe if we can do that instant answer, well, that's better than the human because it gave you the instant answer. Right. And so that's, that's great. So, you know, we're definitely going to have that on and we'll definitely be attempting [SPEAKER_01] to do that on some level. I do think there's a downside of like when the AI gives bad answers or wrong answers, it's very bad and very frustrating, which I've had. I mean, I've had it. I've had the bento AI give me a wrong answer. I've had the Laravel AI give me a wrong answer. Stop it. I've had lots of AI giving me wrong answers. So like that's the thing is like there is a balance there that this is to me why like everybody just doesn't use spin already because like the wrong answers are so bad to occur. [SPEAKER_01] It's so negative when the answer is wrong that like the upside you get is, is not sufficient. Like when it's 90% right and 10% wrong, the dollar value savings is not there to for that offset in my opinion. I absolutely disagree. [SPEAKER_02] I absolutely disagree. Why can't we disagree? No, I, I, I can't, I can't disagree more. And I'll try and like, do you know how long it takes for a human on average to respond to a ticket? For me or just in general, you're saying your team helps. [SPEAKER_01] Yeah. I'm sure it's like probably about, I mean, it depends a little bit, but it's somewhere between like 30 minutes to an hour, usually something like that. [SPEAKER_02] Okay. So assume an AI goes back and forth in those 30 minutes. The human can still intervene. A human can still respond up. AI was dumb. I do. I do that sometimes. I'm going to go fix that. Here's the actual answer. Right? Right. But if it can help that customer instantly with something that is morally tailored. Right. So it's like, and I feel, I honestly feel like 90 plus percent of your tickets can be answered that way. [SPEAKER_02] And then the team can be focused on better docs making their garden of support way cleaner so that the AI is more accurate. And a lot of the time, like when Bento gets it wrong, that's very slanderous of you. But when Bento does get it wrong, you know, when Bento does get it wrong, it's like, yeah, like, like actually we got, we got one wrong this morning, which was, um, it didn't know the demo booking link because I updated it. And so it kind of shocked. It was like, Oh, I can't, I can't schedule a demo for you. And I was like, Oh, that's bloody annoying. [SPEAKER_02] Um, so then I go in and actually Anya did that. She went in and then just changed the docs. And so then the next time it knows how to do that. But I could, I think Anya still replied, Hey, here's the actual link. Right. So she still jumped in and it was still fine. And they booked a demo with me. So it's, it just, it changes the scope of the role where support is becomes more of a gardener. I think then a frontline, you know, person who's, who's always on the attack. [SPEAKER_02] Even, um, I saw Amar from Zenmaid right this morning that someone's husband was basically being super abusive to his support staff. And for me, I'm like, AI is great for dealing with that. I can't deal with abuse. Like I am, I jump in and I start, you know, I can't deal with abuse. [SPEAKER_01] No, I don't want anybody abuse my staff. Yeah. No way. [SPEAKER_02] Of course. But can they abuse a bot? You know, it's, it's probably, it's probably fine. The bots have at it, you know, go back and forth a hundred times. I don't, you know, I don't, I don't, I don't. Yeah. So, and then it saves the mental health of your employees in that regard. So there's, there's other benefits as well. [SPEAKER_01] Yes. I think the kind of jobs that get a lot of that kind of abuse straight, I think are the ones that are most, uh, so I, so in my opinion, like what you're saying with Bento, like, I think I, I agree for Bento software companies are going to be highly automatable in all kinds of ways for like customer service for all sorts of things. They're like kind of one of the premier use cases of it. I think like big consumer company that doesn't get that many different issues. [SPEAKER_01] Cause they kind of do one thing like, uh, I don't know, like a Klarna or something like, Hey, there's like 20 types of issues. Like somebody tried to, you know, get the credit check to be part of this. Somebody tried to buy something they couldn't like, whatever. There's like the 20 things that are like people have issues with Klarna. Like that's yes. All automated. Even though Klarna did try to automate and I know they pulled back a little bit from whatever. Yeah. But I think eventually like, okay, like you're getting kind of a set group. Once you have like the better AI, once they have the sufficient tools, like all that stuff, [SPEAKER_01] I think it'll be able to do a lot for those. Uh, but I think that's like actually a relatively small percentage of the companies that are like so clean, like a software company or like a one, a single purpose, limited entry point sort of consumer type element. Um, I just think a lot of it, a lot of companies just don't operate that way. There's going to be human element. Now I do think, so this is how we're building health.ai. It's like the real focus of it is on, can we get rid of that? [SPEAKER_01] Like the Pareto, like 20%, like, Hey, there's 20% that are like instant answers that are like a simple back and forth that are like all that stuff. Yeah. And now maybe over time, right? Yes. Then maybe eventually it is 80, 20. Yeah. Oh, you think it's Twitch. Yeah. I don't see this one. I think now it's switch. Yeah. So, but I don't see was like in one year, there won't even be help desk applications. All support will be AI. And I think that's like, doesn't make any sense. It's just not human nature. Nobody's going to operate like that. [SPEAKER_01] Um, now in 10 years, this is a, to me, it's like a one year, 10 year thing. Like people underestimate, overestimate what they can do in one and underestimate in 10, like in 10 years from now, might it be 80, 20 and you won't and or more or every company has their own AI mega brain and that does all your customer service and all your accounting and all your, maybe in 10 years, but not in one. Um, so yeah. [SPEAKER_02] Well, well, like with all this stuff, like we are obviously like super early, right? But we're super crazy early. Let me just go into emphasis there. Crazy early. We're very, very early, but very early. If people are listening to this podcast and they're early and they're interested that it is a time to experiment. And like, for me, the reason why I think, um, I, I am trying to die of all in this because [SPEAKER_02] the last time I dove all right in was when I started doing like online business stuff in like 2015 for myself. And I met crazy people that were hanging out in Thailand and weird places in Europe and living like amazing lives. And I was like, Oh, you can earn a thousand dollars a month selling skipping ropes online. That sounds like the best ever. And, um, and when I pursued that, I ended up like where I, where I am now. And so I'm, I'm very interested in like, all right, like what's being at least whispered here. [SPEAKER_02] What can we take advantage of as a small business? Like how can help spot be super, uh, cost efficient? How can I, I care a lot less about the super corpse and like the impact on them because you know, at the end of the day, you and me putting food on the table for our families. [SPEAKER_01] And so like, that's, so for me, that's kind of interesting. Like, I mean, definitely I agree from a, like if you're a founder or just potential founder out there listening, I think that's a different set of advice for sure. Like that's like, yes, what might be pot? Like if I start now, what's the world going to look like in five years and what tools are going to be needed and how's the world going to change? And how do I take advantage of that? Not that we should be thinking about that too, but I do think it's even a little different, uh, in that regard. [SPEAKER_01] Um, for somebody who's like, I'm building a product that's for that time in the future is a little bit different. Uh, but yeah, like for me, like even so for me, in terms of us as small businesses saving money, like I'm never going to have less than one support person. Like I have one support person. Yeah. And I'm never going to have less than one support person. So like, there isn't actually any money for me to save in that regard. Now there is like, I'm already, this person we hired, I already established some of this with the idea of like her job now entails more than just support. [SPEAKER_01] Like in the past, it was like pretty much purely support. And with the idea of like, AI is going to take at least some of it. We'll see how much that is. Um, but there's gonna be more of like, sort of, uh, success training aspects to the, her role that wasn't there so much in the past. Cause it's like, we're gonna have all these new tools. Nobody's gonna know how to use them. Nobody's like us out there in the corporate world, like just on the bleeding edge figuring and thinking about how to use these things. [SPEAKER_01] So like, we're going to build these tools, but we're gonna have to really train people how to use the tools because they're not going to just inherently grab them. Like, uh, like we do. So, so that's already, yes, I'm on board with that, uh, aspect of it for sure. [SPEAKER_02] Um, something, something, something I actually want to, yeah, something I want to touch on cause it's been like a theme in like some of the like group chats, like just business owner chats is that like a lot of people also are engaging a lot less with agencies. I'm not talking about dev, like dev agencies. I still think are highly needed, highly valuable. I'm talking specifically like marketing, marketing, marketing agencies and SEO agencies and ad agencies because what, what people are finding is it's like, would I, I saw the [SPEAKER_02] guy who runs, um, hopefully who ran groove before he posted this morning. He's a, he's a good social media poster. He posted this morning. Like he got rid of his SEO agency and replaced it with a guy internal with AI. And it's, it's been great. Yada, yada, yada. Um, so I am seeing like a shift of people actually bringing stuff in house. So they're hiring and taking out of agencies effectively. So you bring someone in house, give them as much token usage as possible and just could [SPEAKER_02] be hyper efficient at your job and working a little less outside. So it's like another company behavior change where maybe in the future we're working, you know, maybe we've got more employees, um, but they're more efficient and we've got less outside contractors doing that stuff. So that's not possible. [SPEAKER_01] Again, I think like, that's the thing for like, we're talking about very small businesses. Uh, I think that's true. I don't, I think even just slightly above very small businesses, I don't really buy that whole narrative because I just don't think they want to do that. Like they're like, we're not good at this. Okay. We can hire a guy who does it and give them AI, but that guy's going to leave. And now we have to find somebody else. Like now it's on us to be good at SEO. We don't want to be good at SEO. Like we want to sell machine part or whatever it is we do. [SPEAKER_01] And so like, we're not interested in, in doing the SEO. Like this is where like, to me, like this is what like, when you get to like Twitter, like, oh, they can do everything. And yes, it can do everything, but not everybody wants to do everything. So like, that's the thing where like, ultimately somebody has to do the thing. And then do you want to be responsible for the person doing the thing? Or do you just want to pay five times more money and have somebody else do it and never think about it? And like, there's a lot of places that are like, yes, I just want to pay five times more money. Like I don't care because you got to remember, it's not their money, right? It's not their money. [SPEAKER_01] This is a thing I've run into all the time with like bootstrappers and startups. Like it's not their money. They don't think about the way we think about it. They're like, somebody gave me $500,000 to spend on marketing this year. I'm not getting any bonuses if I spend 450. Like I just spend the 500. Like, do you know what I mean? Like it's not, I saved some money. Nobody's gonna give me a pound back. Nobody cares. Like they said, if anything, they're gonna be like, why did you spend all the money? Next year, you get less money because you didn't spend all the money this year. Right? [SPEAKER_01] So it's like their motivations are not the same as our motivations. Like that is a difference. And I think sometimes you and me are talking through each other in that regard of like, on our level, I totally agree. Hey, you save me $10,000. That's like $10,000 to me. I know I can buy a fancy lens, right? Like I could do stuff with $10,000. Like, but when you're like a middle manager, like nobody gives a crap that you saved $10,000. Totally irrelevant. So I do think there's like some of that stuff, but still, I think the AI is being hyped a lot. Right? [SPEAKER_01] And so people are gonna come and be like, people's bosses are going to more and more start coming to me and be like, hey, can we save $200,000 of this through 500,000? Right? And now that's real money. And I might be aware of that as your manager. Um, but so then the question is like, is that because you're going to have an internal person do your SEO or is it that AI makes it more competitive for SEO agencies to offer SEO less expensive to people? Do you know what I'm saying? That's probably correct. [SPEAKER_01] Yeah. I still have an SEO agency and now I have a bunch of bots and people with the bots. And so I can do everything for half the price or whatever. [SPEAKER_02] And maybe it's not an agency. Maybe it is a hyper efficient freelancer or a contractor or whatever. The shape is different. Yeah. Who knows how it's going to, yeah. [SPEAKER_01] That stuff I think is going to be, could be all weird. Like what happens in those spaces, like where it's like the money is a lot of money still, but it's like not an insane amount of money. And then, so like the efficiencies could be really important there. And then how that comes to market, uh, will be intriguing. I think. [SPEAKER_02] Um, yeah, it's, it's, it's funny how, um, the end result of a lot of this is that like less people do more work. Like let's have more responsibility. [SPEAKER_01] I mean, are you doing less work? I don't know. I'm doing more work. I feel like I'm doing more work. I'm doing less work. [SPEAKER_02] Yeah. Probably more, more, but I feel like I am less anxious. Like, uh, an example is like, um, a growth part example. So we'll go all the way around. Um, it was my four year olds, uh, birthday, uh, yesterday. And she had a birthday party the day before. And, um, I'm in the, I'm in the kitchen, you know, making some birthday stuff. And then, um, you know, there's four year olds running around screaming and stuff. And, um, I get a little alert on my phone being like, oh, response times have just shut up on Dento. [SPEAKER_02] And I'm like, oh, and so then I go to Grop, what I go. So WTF is this. And then it, it hits a better stack, honey badger, dah, dah, dah. And then basically just gave a full synopsis and was like, Hey, do you want me to do X? Cause someone was basically DDoSing us. And I was like, oh yes. And then everything, all the response times dropped down. Everything was calm. And that would have been an incident where I had to run off into my, uh, office to go, you know, create some firewall rules and all that kind of stuff. [SPEAKER_02] Um, whereas now that's not a, that's like less of a thing. So there's stuff like that where it's like, yeah, it's, uh, you know, and in that is I've got like an infra bot that has access to those MCPs. I've got a dev bot and I've got a couple of other ones that are in that like group chat and they all go off and then in a couple of minutes come back and it's like, yeah. And that's crazy. [SPEAKER_01] It's so crazy. That's crazy. I don't know. That's crazy. Like, are you not worried though that he could like do the reverse and screw up your infrastructure? Or you feel okay with that? Uh, well, like, yeah. [SPEAKER_02] The way that I think about it is like anything that is reversible, I am okay. So like, I'm not going to give it access to like Redis clusters raw or Postgres clusters raw. Um, am I happy to give it access to the ability to add a couple of firewall rules here and there? Right. Yeah, sure. You can fix that easy enough. Yeah. Turn off some rate limiting or whatever. Like, so anything that's like I deem is reasonably reversible, even like, yeah, like emailing customers and replying to stuff, that's, that's reversible. Cause you'd be like, oh, sorry. [SPEAKER_02] I didn't mean to say that. Like it's not, things, things actually, things actually aren't as high risk as I've had humans like, you know, make issues with bad human support. [SPEAKER_01] Sure. Yeah. Yeah. Yeah. [SPEAKER_02] So, so yeah, I might, but like the benefit gain of that is I stay present with my family. I, I'm not grumpy. I'm, you know, it's, it's that, that is very huge for me personally. It's like, that's very impactful. [SPEAKER_01] So now here's the, so this is the interesting, this is like a perfect little knife's edge thing here because like, to me, then the next layer of that is still ultimately, I don't want to get that message from the bot at all. And I want somebody else to get that message. Oh yeah. Yeah. Because at the end of the day, it's still messaged you. You still had to be pulled out of the party mindset. You still had to be worried about the thing, even though it was 10 minutes and like it got fixed quickly and you'd not have to go to the computer. Like it's still is you who got that. So like, are you, it sounds like you're obviously not to the point yet where you want that off [SPEAKER_01] your plate, but at some point might you even want that off your plate or you think the bots will get there fast enough that you can just let the bot auto handle those kind of things two years from now or a year from now. I don't know. I don't know. [SPEAKER_02] Like, like it's, it's, it's something I think like pretty deeply about because it's like, um, but there's just, you know, by hiring a human to help me with some of those things, there's just like additional, there's, there's additional risks. There's a whole bunch of SOC 2 compliance stuff. I now have to deal with, um, SOC 2 compliance is quite nice when you're just one person. Sure. [SPEAKER_01] I'm doing everything right. Don't worry about me. Yeah. [SPEAKER_02] Yeah. Yeah. So, um, but you know, and there's a locality thing here. Like if, if someone is touching customer data, I really want to be able to be near them, you know, legally. [SPEAKER_01] Like you're very, okay. Interesting. Yeah. Yeah. [SPEAKER_02] Well, like I'm, I'm, I, um, I, I am paranoid and I very much want to be like a cockroach in this business. I just don't want to die. And so for me, it's like, if I'm going to hire someone, I actually do want them to be local to me. I want them to be under the Japanese legal system where they'll go to jail for a very long time. If they ever sell my data for Bitcoin, um, and Japanese prisons aren't nice. So I, uh, yeah. So, so that, that's, I think that's how I kind of think about it is like, I'm comfortable [SPEAKER_02] with the workload. I'm comfortable with the work burden. Um, for now, cause I'm, I am, I am all over it and I, I understand that, but I'm using AI to, to give, give myself more time and more space and, um, be happier with that burden effectively. [SPEAKER_01] Right. And it's like, you know, it's actually the limits of like what one person can do. But like this automation is super awesome for you. Cause that was already your mindset. And it's like, well now it's really relieving you of a lot of burden. Um, you know that, yeah, you would, you would have to hire somebody generally to relieve you of that. And then also like, there is a lot of small businesses have this and I'm sure you have it. It's like, there's like a weird amount. It's like you could hire somebody, but it's like not that much work. So, but you, but to pay them like appropriately, now you're paying them for like the not full [SPEAKER_01] time amount of work really. And then you're like finding stuff for them to do at the other time. But really you're just hiring them for this like message you get once a month. But really like, like that's why you're really hiring them. Um, but it's like, but now you have to fill the space of the other 39 hours a week of like for the stuff I'm new or that you're hiring them part time, which is not the greatest solution either. Cause then they're like ephemeral sort of, and sometimes people have great part time. People last a long time, but often like, you know, they go get a job or whatever happens. [SPEAKER_02] It's harder to get part time and contractors risk. And I think like that's, you know, yeah, it's hard to have a hard time to be like, right. [SPEAKER_01] You got to answer this message anytime of day, but you only work 10 hours a week. [SPEAKER_02] Like, like I'm getting that and this, this is the thing is, is like, you know, um, like the, the problem that you're kind of touching on is like, I'm 24 seven on call. Right. Right. And so like that has, you know, people talk about like the, the burden of being on call, even at a corporate job or whatever. Right. And, you know, people, people will make like a YouTube videos where they're like, you know, my week on call at Amazon HQ. [SPEAKER_01] One out of six weeks I'm on call or whatever. [SPEAKER_02] Yeah, exactly. And I'm like, Oh dude, I'm like on call 24 seven, like 365. So yeah, but, but yeah, so it, it changes the scope of the challenge where you're just trying to be okay with that and be comfortable with the burden. The rewards for that burden are quite high. So you're obviously okay with that. But yeah, it's like even a full-time person, it's like if an incident pops up during the day, they're probably asleep. So then I would be on call anyway. So, right. I know it's, it's, it's on a Sunday, that person may be away. [SPEAKER_02] Do I want them to carry their laptop? I'm always got my laptop. All that stuff. Yeah. It's true. [SPEAKER_01] It's weird when it's like not that much work in a sense, but it's like the most important at the same time. Yeah. And it's random. All right. All right. So let's, we have one more take that I think we'll probably disagree on than a couple other things. We're already at 50 minutes. Um, what do you think normie people will do with a bot? Cause I don't think normie people have any use for a bot. Like almost no use for it at all. Speaking of bots, my, uh, cleaner just went off. Give me one sec. [SPEAKER_02] Let me kill it. One sec. Look at this guy. Everything automated. It, it, it hurt itself. The vacuum. Hey, hey vacuum. Say no. Do you hear that? Oh, that's a little Japanese vacuum. A little, little Japanese vacuum. It, it, it, it, it, it heard us talk about the utility of bots. I was like, I need to come and put my input in. Um, all right. Uh, I think, I think I agree with you. [SPEAKER_02] So I, I, I think, I think, I think I agree with you. I don't really think normies are going to do bots and I'm going to kind of, uh, preface it with, with bot. It's very useful if you go into it wanting to do stuff and implement. Real life stuff. Right. You know, go into my investment portfolio, grab my thing, dump it here. Like, uh, check, check our support every hour, you know, very action orientated. You'll have a lot of success with a bot if it's very action orientated. [SPEAKER_02] What I noticed, especially like on Twitter. Yeah. And I think why, why the Hermes and open call people were, were a little bit icky is that like, they're kind of don't really have necessary like full jobs. And so like, they don't have anything to do with the bot. Right. Nah. So they're like kind of making up stuff. And so like, you know, that's exactly it. [SPEAKER_01] That's exactly it. [SPEAKER_02] Yeah. Yeah. They're making up stuff. Yeah. And like, if you go into bot and you're like, I'm going to build a full, you know, go to market sales team in here and you've never actually built a go to market sales team. And like, you don't know how to find leads and richly, like you don't understand that whole system, then you can't, you can't replicate it. If you, if you haven't written SOP documents for a specific role, then you can't really automate it. So I, and I feel like normies, they open the app. [SPEAKER_02] It's a blank canvas and they don't understand the potential here. And also they're not allowed. Like I was speaking to my sister. She works at, um, uh, big company in Australia. I won't, uh, add her on this, but then she's not allowed to use AI in, I asked her, I was like, are you using like AI for your job? And then she's like, I'm explicitly not ever allowed to touch it. And, um, and half her job honestly could probably be like, she could probably just be chilling. [SPEAKER_02] Um, you know, it's, it's building reports. It's, it's doing a whole bunch of stuff that, that could be more efficient. But her manager says, you're not allowed to touch it. So there's, there's that, there's that, that thing. So I, I agree. Like, I think these tools right now, we're like so early on the, that, whatever that curve is that the people who are benefiting and like the ones that are willing to take a lot of risks and implement it. And the normies aren't going to catch on until it's like, I don't know, like a Slack default [SPEAKER_02] or a Microsoft Teams agent that people can actually do. Um, yeah. Yeah. [SPEAKER_01] And even then that's still all like within their work. That's like work is going to set it up for them. Like you're saying, right? Like, okay, work has ultimately been sold something by Microsoft or whoever. And now in teams has access to all documents, whatever can make a portfolio. But like you have a normal job. You work at the mall, whatever you do, whatever you do, right? You're a security guard. So like you do what you do your job, which is not going to be AI. And then you're gonna come home and you don't, what in your life is like automatable? Like we have things to automate. [SPEAKER_01] Cause we're like, Hey, we run this business and there's shit to automate in it. Like if you're just a normal person, like I'm going to go to the movies. I'm going to watch Netflix. I'm going to go out to dinner. Like I'm going to like, there's not actually that many things you're going to automate. Like I'd be at every normie I talked to ever use the AI strictly as Google search. It's just Google search slightly better. And that's like the only way they ever use it. Um, and now maybe they could, the more advanced ones might like automate their finances at some point. [SPEAKER_01] Like, yeah, go ahead and connect to like my 401k and my this and that make me a monthly report. Like I could see some of that stuff, but like, I don't know. It's like the hype is so high that like everybody in the world is going to use this. And I'm like, I don't really see how they're going to use it. Like, I don't, I literally don't know what they're going to do with it besides like ask it like a Google search things or, you know, again, like once in the, in the future, when I have a home robot, like you already have, cause you live in Japan, but when the rest [SPEAKER_01] of us get the home robots and the robot uses the AI, of course, that'll be awesome. And y'all tons of uses for that. But that's again, like where it's something has abstracted the AI. Uh, so you're not using the AI and programming it. You are just somebody packaged it into a usable thing for you and you are using it. So I wish there'd be tons of money to be made there, but I don't know that the normal people are going to look at it as like, I'm going to use AI, which I think is a huge part of the problem with like why people don't want data centers and why people hate AI. It's like 80% of the people hate AI. Right. [SPEAKER_01] Cause they're like, I don't, I don't get it. I don't. Okay. It's like, it saves me three clicks that I used to make on Google search. Now it just shows it to me at the top, but like, I don't get it. Like, I don't, I don't understand what we're doing here. So that's kind of, I don't know. Well, it sounds like we're semi in agreement. [SPEAKER_02] Yeah. No, I think we're in agreement. I think like it will be very interesting. Like I'm very interested in like what the next generation of products kind of looks like, like using the, you know, when you're talking about like how, how will normies interface with like AI in the future? I do think like, you know, uh, businesses less likely to log into like a WordPress instance and more likely to just share a Google doc with like some bot in Microsoft teams to get something published. Like probably. Yeah. [SPEAKER_02] Um, and so there's different products and stuff you can build to fit that workflow, like in the future. But I still think we're all trying to work out what these workflows are like. It's very much like remote working in like 2015. We're trying to work out, you know, what is zoom? How do we do these remote calls? What's like, how does Slack work? It's still, it's like that. [SPEAKER_01] Yeah. And that's, that's a great example. I also think that people just, there's just this, like the Silicon Valley, Twitter take on things is always so extreme. Right. And it's like, I had the tweet about this the other day. Like there's still like 20 years ago, 20 years we've had self-service and we've had email support and we have chat bots. And we're like, you could have had to take 20 years ago. There's not going to be phone support anymore, but there's still phone support at like almost every company. You could say, Oh, we have zoom now. You're never going to have in-person meetings ever again. And yet all these people fly. [SPEAKER_01] I mean, they fly in all the way everywhere to go to in-person meetings all the time. There's still consultants flying it. Like these things have not been eliminated. So I think there's like just a lack of like the human element of what people want. Like you could replace all these things with technology today. But if the humans don't want to pay for that, then it's not going to get replaced. Like if there's people who don't want it to be replaced, it won't be replaced. Like even if you can do every meeting, every conference in the world could be zoom right [SPEAKER_01] now, but there's more to it than just the information that comes over the zoom. Right. So like there's all these subtleties that I think are fascinating. [SPEAKER_02] I think you're right. Yeah. Like I think, I think you're right, but I still feel like a lot is changing in our income brackets of businesses. Right. So I feel like businesses, right? [SPEAKER_01] Like there's less phones. Like 20 years ago, everybody had a huge phone call center. You had a huge call center. Everybody had a huge call center. And now it's definitely less call center than there was before. And so like, there are changes, right? It's just that like, it's probably not all or nothing. There's probably like levels. There's people who are going to use the technology better as you're trying to do and gain a competitive advantage by doing, by using it more extreme than someone else might be willing to. Like, can you extract more value? [SPEAKER_01] Can you be faster, better, all those things? Like, I think you would say you're taking more risk on some level. Right. But like, but you're, you know, so far so good. I think you would say, right? It's going well and you're reaping the benefits of taking that risk. And so, and that's a very entrepreneur type thing to do, right? Like that's a founder type thing to do. It's like, let's take this risk. I'm willing to fail, which is not what most people are willing to do. [SPEAKER_02] And I think like the Hermes open claw, you know, on that, the adoption graph was like right at the start, right? Where it was like, like right at the start. And then as that little ramp comes, the GrokBot's here, like GrokBot's kind of, and I'm seeing a lot of friends just test it and try it. And like, because Cursor is managing the security, not yourself. That's where the risk has come. Yeah. The risk comes down, I think quite substantially. [SPEAKER_02] And like, I, you know, I agree. Grok is a horrible, horrific brand. They should have nuked it from orbit. I think like, but. [SPEAKER_01] Why did they keep the Cursor brand? It makes no sense. Cursor has a great brand. It's so crazy. [SPEAKER_02] I agree. I agree. But like the Cursor team, I think is, is excellent. And they're really running it. Yeah. Yeah. And arguably more, I think more consumer ahead than Anthropic and open AI. And my own behavior, like I use Grok4.5 almost by default. I'm toying with 4.6, but I don't like 4.6 is like, it's kind of like soul and fable. It's a little bit verbose. [SPEAKER_02] And I like 4.5 because it's, it's punchy. Yeah. I like the composer was at 2.5. I use that for sub agents. Cause that's, you know, very fast acting. And so it's like, yeah. And so I was kind of, I'm using Cursor. I'm going to try their new origins product today. Um, I'm kind of in that ecosystem. Mm-hmm. And that's why I took the leap. Like if opening, I release this, I probably wouldn't be trying it. It was just, I was already in this ecosystem. Um, so I was a little bit more confident. Yeah. [SPEAKER_01] Yeah. I do wonder if like, um, there is an element, I think Claude has like won the corporations, which is why they're so like, they kind of have corporations on lock. I feel like open AI has won the like total normies. Like I'm just using chat GPT for the free mode of chat GPT for Google search. Like I just have chat. Yeah. Um, which is why they're like giving away tokens and like, they're like trying to get into the corporate world. Like there is this like spot or like maybe grok. [SPEAKER_01] I think it's like, I think bot is really smart from a product perspective of like, there's this group that's between those two things. Like us, we're talking about here of like, we're ready to go farther. We're ready to automate. We don't, we want the safety net of a bigger company. That's like securing it. We want the like rough edges. We don't, we're not like trying to like write the whole thing ourselves from scratch and have it rewrite itself every day. Like we just wanted to like do the task we have to do, right? It's like, so we're above the consumer. We're like a prosumer in a weird sense. [SPEAKER_01] It's like more pro than that. But like still like where we have tasks, but we are not a corporation with like, uh, you know, levels of permissions and all that, you know, layers of management and stuff. So we can just start using it if we want to. So I think that's sort of an interesting market play on their part to like try to come in here and obviously like leverage Twitter, I think, and do all that kind of stuff of like spread the word, get some of that benefit for themselves. [SPEAKER_01] Um, so yeah, it's, it is intriguing, but it's definitely the first Brock thing I've been intrigued by at all. Um, and really the first, I tried cursor, you know, pre, um, XAI and everything and like didn't really click for me. I don't know. There was like some of the clunky and this didn't really work for me. Um, so I don't know. So this is the thing I've been most intrigued by getting back into that end of the world. I don't know if I don't think I actually will. Cause I think like Claude and open it out, I will just have bought like within a month. Yeah. [SPEAKER_01] Uh, you know, cause like I, I can't imagine them letting this sit out here very long, but uh, I mean, Claude already has like the managed agents, which is basically bought, but it's like been in beta for like ever. And you can't just like sign up. I don't really get what they're doing with that. But, um, anyway, all right, enough bot. We've already been on over an hour. Do you still have time? Yeah. I can talk for, I can talk for another two hours. [SPEAKER_02] I'm good. Yeah. As long as you don't know, cause it's late your time, right? It's like only. No, I'm good. Yeah. [SPEAKER_01] I'm good for at least a, it's 940. Yeah. We're good. We're good for a little more time. Um, I have a couple of like non tech things, but before we talk about those, like any AI tech stuff, cause we've been going back and forth on various things, anything. We've kind of been mostly in this bot, the future zone. I don't know anything you have that you want us to discuss on here and the tech related side of things. [SPEAKER_02] Uh, system. We got it. We got it. That's a lot of AI. [SPEAKER_01] That's a lot. [SPEAKER_02] Yeah. I, I, and that's enough promo for Grok as well. I think that's, um, that's too much, too much, too much unpaid. [SPEAKER_01] Yeah. Go try Bob if you're interested in all this stuff, but that's it. Um, okay. Yeah. One thing that I'm super, I've been loving is your like video founder in the park video series. So let's talk about founder in the park, founder on a bike, founder in a Japanese house with a cushy chair. Like I love this vibe. It seems like, I don't know how much was planned, stumbled upon. So give me, give me what you got. Cause I'm going to, I want to steal some of this, but I want to hear what, how you did it so I can steal it properly. [SPEAKER_01] Give me what you got here. I think you have a huge advantage being in Japan. It's already giving you like a different vibe. You're not just in San Francisco somewhere. So that's awesome. But like, how'd you start doing it? Like what inspired you? Give me the backstory for those who don't know, Jesse on Twitter has, I don't know, the 10 minutes, little 10 minute videos of his like kind of one shot, one shot, take type things, right? Yeah. One shot. He sets up a camera in the park or wherever he is, you know, nice backgrounds under a tree somewhere. And it's like, here's his take on thing X. [SPEAKER_01] So I don't know. Talk to me about that. Those are really great. [SPEAKER_02] Yeah. So I think, I think this all ties into our like REI discussion as well. Um, so something that has been interesting with Bento is that like, oh, I don't know. I've been like, so in my background was as a marketer and I've been so in dev for however many years, just basically doing a little bit of everything, doing dev, doing some customer development, doing support sales, all that kind of stuff. And, um, after I did the planet scale migration, which I covered on the podcast with Aaron, [SPEAKER_02] um, things became like a lot calmer for me, like on the product side and on the infraside and stuff. And I was like, wow, you know, I, I really should. Um, and there's a lot more work, not just the planet scale stuff, but, uh, things just became a lot calmer and I just had the capacity where I was like, I need to like think about the marketing side. Um, the murmur between a lot of my friends, I don't know if this is also true with you, but like the murmur with all my friends is, Hey, our website SEO traffic is all gone down [SPEAKER_02] or such and such channels have really been impacted. I've got, I've got a friend who said that web traffic went from 50 to 70 K a month pending to like hundreds of visitors a month. And, and that's across like a lot of web pages and a lot of investment in search. And that's a lot of top of funnel. So then the next thing was like, all right, well, um, one, I should have been doing SEO like probably more aggressively give my background for a while, but I just didn't. And then now I'm like, all right, well now I'm not even going to do it. [SPEAKER_02] I'm not just not going to do content writing. What's the point? Um, you know, I'm pretending I'm smart. I'm like, I knew this was coming or whatever, but I was like, I need to, I need to figure out a way to stand out. And in the past, like I'd experimented with this. Like I just filmed a video, uh, at our local park, like a couple of years back and it did really well, but then I never like kept it up because it's super embarrassing to be on camera. [SPEAKER_02] Um, but then something I think I realized, uh, actually thanks to you guys. And also, um, the Laravel live in Japan was like how impactful video and like this form was regardless of the audience size. So I have no idea what the listeners are mostly technical. You've never told me, so I have no idea like how many eyeballs. [SPEAKER_01] It's around like 5,000, I think across everything, you know, it's like hard to say. Yeah. But phenomenally impactful for me. Right. [SPEAKER_02] So like, regardless how many people are listening, it's like. Advertise on the show, people. [SPEAKER_01] Advertise. [SPEAKER_02] Anyway. So they're, they're, they're, they're smart, amazing, you know, they're, they're, they're high quality listeners. [SPEAKER_01] Beautiful, very attractive people. [SPEAKER_02] Very attractive. Um, um, so I think my realization was like, and then when I went to Laravel live Japan, like so many people knew me because they had just listened to the podcast I did with Aaron like the day before or whatever. [SPEAKER_01] Oh, that worked out perfect. [SPEAKER_02] Yeah. Yeah. Right. So, uh, it was very, yeah, it was, it was very crazy for me to like walk into a venue. I was like, no one knows me. I'm a Ruby dev, you know, like no one knows my face, but then a whole bunch of people did. And I was like, okay, so, but, and I've got a small audience on, um, on X and I was like, well, what if I just start like actually do this regularly for a change? And what if I just, I just commit to being embarrassed and just give it a go. [SPEAKER_02] And I did it and I did, um, I did three the first week. I'm just like, I'm going to hammer, hammer it out. And I was actually, I got really sick the week after. So actually that week, I think I was in like a coming into a fever. So I did the first three videos Monday, Wednesday, Friday, and I think I was actually getting like deliriously sick. So that's probably why I just did it all. Shipped it, got sick the whole next week, but something really interesting happened. I got some very big leads. I got one of Tatami's largest leads. [SPEAKER_02] Um, it was like, uh, $5,000 a month. Um, one for Tatami's that's for the side business. I got, uh, one, a very large Australian lead who I think they're going to pay me in two days. Um, for Bento. Um, I got, I got loads of messages and like really nice things. And the view count was not large, right? It was just X. And like, we're talking a few thousand, a couple of hundred likes. [SPEAKER_02] So like, that's cool, but it's not like, yeah, it's got out there, but it's not, yeah, [SPEAKER_01] it's not the whole internet, but it's like, yeah, it's a nice. [SPEAKER_02] Yeah. But you just kind of forget that like every number is a person. Right. And, and, and, and, and maybe, and maybe with X, maybe like, you don't know if they're actually really people. Yeah. It's half a person, but yeah. It's, it's half a person or like people's grok bots viewing it. Um, but you know, there, there, there, there are people watching and it's impactful and, um, it led to a very large, um, yeah, like bumping MRR. Like, I think, I think I'm, I'm, I'm, I'm quite a lot this month. [SPEAKER_02] This is like our largest growth month and it's off the back of those videos. Oh yeah. Um, and so, yeah. And then I had a whole bunch of other people copy me. So they watch my videos and then they did it. And it's also been impactful for them. That's my plan. [SPEAKER_01] Yeah. I'm just copying it. [SPEAKER_02] And I think, and I think the reason why, like, I'm like even currently in this chair with like the background and stuff, like, um, the background is awesome. Yeah. I think people are super over, uh, the led light background in their, in their bedroom. People, it's like, Oh, people don't like over optimize stuff. Um, and I think there's this like deep yearning for like, just tell me what's up. Like, what are you working on? [SPEAKER_02] Um, and I'm trying to keep the videos very focused on that, which is like, I only want to talk about stuff I'm doing. I don't want to, I don't want to ever talk about like theoretical stuff unless it is like impacting me directly. Um, or impacting me like building these businesses. And so the combination of like, I only ever talk about what I'm doing and film it in something that is unique to me. And like, even if you live in America, um, there are parts of America that are interesting. [SPEAKER_02] Like, could you go to a Denny's and record in a Denny's and talk about your business? That would be cool. You know, like, like, yeah, you know, um, where, where, where you're in, you're in New Yorkish. [SPEAKER_01] New York. Yeah. North of New York city. But like, yeah, there's like a, it's like a beautiful area. Like the, it's not beautiful to me necessarily cause I've lived here my whole life, but like to somebody outside, they'd be like, Oh, that's really cool. You're like on the world's longest pedestrian walkway or like whatever. There's all kinds of stuff around. Exactly. [SPEAKER_02] And like Ian talking in a New York scene, maybe having like a sandwich from a bodega or [SPEAKER_01] something, you know, like to the city and do something. Yeah. [SPEAKER_02] And it doesn't need to be complex. Right. But I think like as people are scrolling, everything's like hyper optimized. Everything is a clip. And, um, yeah, so it allows you to stand out. People are willing to like, listen and all this kind of stuff, but you do have to have something I think to say. So anyway, the summary, it's been like a very successful experiment. I'm just going to continue it regardless of, you know, the views or anything. I just want to get stuff out. It's not the downside, you know? [SPEAKER_01] It's like, yeah, it's all upside. Um, yeah, yeah, yeah. It is hard though. It's like real hard. It's hard getting on camera. There's that part of it. That part's definitely. Yeah. I mean, it kind of ties in with our other conversation. I feel like in that, like, um, well, two things. First of all, like the AI only makes being authentic, like so much more valuable. Right. It's like there's so much AI content, whether it's like text content, video content, like video content that's real, but looks AI video content. [SPEAKER_01] That's real, but looks overproduced. Right. It's like, doesn't feel authentic. And I do feel like this is like desire. Everybody has, it's like the same thing about like going to conferences or whatever. It's like, we want the connections. We want the humanity. We want the like authenticity. And those are the things that are really big right now. And it makes sense with where everything else is. And yeah, it's kind of like, again, like people want to do what people want to do. [SPEAKER_01] And so like you could produce a bunch of highly, very fancy videos and spend a lot of money. And you're like, Hey, this means that people will want to watch these videos. But at the end of the day, it doesn't necessarily mean that. Right. Like the people are going to do what the people want to do. And a lot of times people want the human connection. I think especially coming off COVID then on top of COVID, you have the AI. Right. So it's like now we have all this like disconnection from humanity. And yeah, like these videos where it's just a dude in the park, like that's, that's clearly real. [SPEAKER_01] Like you come across as like, Hey, I'm, I'm just shooting the shit here with you. Uh, we're not like, I don't have a team here videoing me. That's seven angles. Like, it's just like, here I am constantly clipping, like, you know, um, Have you put them into short form or anything? Or you're not, you haven't, I don't think I've seen any like that. [SPEAKER_02] Um, and I've had people, I've had people like reply and they're like, Hey, like I recolor graded it for added captions. And I'm just like, stop. Like just, like, just stop. I actually, and this is thanks for asking. The point actually of this experiment was I want to get on my bike, have a coffee, shoot, come home, post within a few hours. Basically like everything that I've done has been done in a two hour block, almost exactly. [SPEAKER_02] And it's like, yeah, drink coffee kicks in in about 30 minutes, 45 minutes, ride to the destination, try. I, and like this, I've got, um, I've got this great, this is a, this is a great tip. I got this little remote. Okay. And, uh, it just lets me, got the camera set up, press record. Bing. Yeah. Oh, didn't get it right. Stop. Try again. And then, and then that's my setup. So, and I got this, uh, same wireless mic. So wireless mic, remote control, try it. [SPEAKER_02] And when I've kind of got the flow of it, I'll just commit to the bit. Like the Grok video, um, was really interesting. Like the Grok video, I think got the most views cause it was like topical. It was also interesting cause I was just talking about what I was doing with it, which I think people found was good. Um, but I don't think I, I feel like critical of myself. I didn't really like speak too clearly. Um, and, and I actually, it was very hard. It was a hard video to do, but I was like, it doesn't matter. I'm just gonna get it out. So yeah. Get it out. [SPEAKER_01] And it's like, it's not like talking about bento where you, you know it deeply. It's like, this is this thing I just started using two days ago or whatever, you know, like, uh, yeah. So it's a different kind of thing. [SPEAKER_02] And so, um, that one, like I was embarrassed to put out and it was like my most successful video. Um, and so, yeah, I don't know. Like, so if anyone's like on the fence, I just recommend just like get your camera, try, but then like, just maybe even don't even edit it too much. Like maybe crop it. So it's like positioned nice and then just like push it out and then just move to the next one. Cause if you just keep doing the next one and the next one, um, it's probably the same with you guys who like the podcast. It's like, you start off a little bit like, what's this going to be like? And then you're great at it. And Aaron's great at it. [SPEAKER_02] And you just, you get the flow, but you've got to get the reps in. Right? [SPEAKER_01] Yeah. Yeah. Once you got the reps in, you can just go, it's so much easier, but yeah, it's hard. That first step, right? Nope. Nobody takes that first step effectively. So like if you're the person who takes the first step, you're already in the like less than 1% of people who even tried to take that first step. So totally agree. Totally worth doing it. Do you put them up on YouTube? I would at least put them on YouTube. [SPEAKER_02] I'm studying. I'm studying. I've actually got my Grokbot where I can link it a tweet and we'll push it up. Yeah. You should do it. [SPEAKER_01] Cause YouTube is awesome. Like it's really the biggest like kind of growth area in some ways to like reach like wholly different groups of people, you know? [SPEAKER_02] Totally. And like, I think people have found that and they started like subscribing and stuff. But X, X has been good because of the virality aspect. I think it does reward video quite well at the moment. And so like, I'm pretty sure that people like both the leads that came to Tatami and Bento, the big ones, um, that was pushed to them. And they weren't already following me. And so that, that, that was kind of interesting. That's great. [SPEAKER_02] And yeah, I want to loop this into like more of like a broader conversation about kind of like SaaS go to market at the moment, because I think that's like just, it's just changing so much where it's like, Oh my gosh, the previous, the previous go to market, you had a couple of businesses. You had one where it was like, I make a Shopify marketplace plugin. I sit, I sit in the rankings and I get whatever comes to me. Right. Or you have, you're on a platform. Yeah. Yep. Or you have a business where you're like, I wrote some content, a couple of them ranked number one. And I sit there and read the benefits. [SPEAKER_02] Like there was a lot of this stuff going on. [SPEAKER_01] That's me. That's mine. [SPEAKER_02] Amazing. [SPEAKER_01] And you've got an amazing business. [SPEAKER_02] And then an amazing 20 year old business. Right. But so much of that is coming into question now. And as we spoke about, it's having like such a large impact on some businesses where they're like, okay, based on our churn and our current net new growth. Like I don't know if you've read the Jason Cohen, Max MRR article. Have you read that? No, I don't think I did. I haven't seen that one. It's like, it's amazing. And he's like, latest book is really good as well. But the Jason Cohen. [SPEAKER_02] That's on my list to read. I already have that. Yeah. I could also talk an hour on that. It's like, it's this really great stuff in there. [SPEAKER_01] Maybe when I read that we should do, we'll do a redo where we talked about the book. [SPEAKER_02] Yeah. Okay. Yeah. A book review. So, yeah. The, yeah. So the Jason Max MRR one is basically like based on your cancellation rate and the total net new revenue dollars in like absolute dollars. So $1,000 extra a month, $2,000, whatever. You basically mathematically can calculate your maximum MRR. And that's actually always been true with Bento. When I've done all the back numbers, his math basically is flawless, at least for my case. Yeah. [SPEAKER_02] And so our churn and cancellation rate is very low, probably similar to help spot. And so really it's like, well, the only lever we can do to get MRR up is to push growth. SEO is out of our reach. I've got, you know, paid sponsorships, you and a couple of other things. So I'm actually pretty limited on like what, what I want to pursue. And then, and the video plus actually like sales led stuff is kind of the approach. So I think we're going to be moving forward. [SPEAKER_02] And that's also coming to the Grok one. I'm going to try and move Anya from support into more of like a sales role. And so it'll be my videos, try and gather some inbound, nurture, bring them in. And it also changes the type of business that we will focus on serving, which we're obviously going to serve the small businesses anyone wants to come in. But the ones that will probably get the tailored one are the larger businesses, which require a lot more movement. So I don't know. Yeah. [SPEAKER_01] Do you want to? Yeah, no, like, I think that's the whole area is super fascinating. I mean, I think there's another area that you shouldn't sleep on, which is like growing the revenue from your existing customers. So like, this is one of the most exciting things about the AI product to me business wise is that like, I think there'll be companies that pay us five times more for the AI than they pay us for seat licenses. Because we have companies, some of our companies are like, hey, they pay for 10 licenses, but they do 500,000 tickets a year. [SPEAKER_01] So it's like, you know, when they automate some of that, like they might end up paying us 100 seats worth of licenses in AI costs. And for them, it's still going to be way cheaper. But like, so you can grow this. I mean, this is how HubSpot is in, but it's like definitely our like new customers have gone down from like what it was 15 years ago or whatever. But it's like our churn is super low. And, you know, we increase prices just straight up increase prices sometimes, which most SaaS companies don't do. [SPEAKER_01] And then, but then selling new, which we haven't done much of this, but like with the AI stuff, we will be selling a new product into the existing customers. And so that'll be a big new revenue driver. So, and I think Bento could do the same thing. Like right now you kind of get everything, I think for like the one Bento price, but is it two tiers? [SPEAKER_02] No, no, no, no, no. Yeah. So use the advertiser should be on me. And so... [SPEAKER_01] I don't use the chat product. I thought the chat product was included, but maybe... [SPEAKER_02] No, no. Okay. So, yeah. So, the way that I shifted this was, you were right. It was all in one. And then I basically wanted to split out the transactional so people could come in. Cause we had, we had a lot of people and like a lot of people that listen to the show where they're like, I just want to send password resets. I don't want to pay for marketing. And so now they can, you can come in transactionally and send password resets and all that. Yeah. Now it's separate. So that's completely separate. [SPEAKER_02] Bento chat, which is like kind of like help scout ish, help spot ish, not help spot, help scout, I guess, adjacent. That I'm, I am not sunsetting, but like, that was always something that like I built for myself. And then, um, so even if you look at pricing, it's like marketing transactional and then chats at the bottom kind of like, if you want it, you can pay us 30 bucks and you can try it. Yeah. It's not really like a core tool. [SPEAKER_01] And I think... That's interesting. Cause I, I, yeah. There you go. I just about that. I always thought that was like, I think it's interesting that you, I do feel like you need to pick a side with that, which is interesting which way you go. I think you've kind of gone the smarter way with it, but like, cause the previous positioning to me was more like we're an all in one solution for your small business. Like there's like a little help desk kind of thing and your email. And it's like, if you stay with that, then I feel like you could lean into that even more. Like, well now there's a CRM and now there's like, whatever. You could expand it into like all the little tools. [SPEAKER_01] Right? So it's like the one box shop for all your small business needs thing. Yeah. But I imagine that makes it harder to sell like transactional email, which to me seems more valuable, which I could be wrong. Yeah. It sounds like you're going that way. No, no, no, no. [SPEAKER_02] Like... No, no, you're, you're, you're right. But there's also like, um, there is like personal capacity to like each skew is its own everything, right? Like, you know, the marketing product, a lot of our revenue is tied to people that send newsletters every single week. Like, you know, they've got large lists. We've got daily senders. We've got some senders that send like 30 million emails a month, just like every single day. Right? Um, and that represents a lot of revenue for us. [SPEAKER_02] We have people that, uh, use us for transactional and go, Hey, I can actually also move my MailChimp account over. And then they do that. So I think I just wanted like a clearer path there where people can like dip their toes in, come to marketing or, you know, they join us with that. The chat thing, it'd be like, it'd be, I feel like it'd be better as a dedicated standalone product. Like it would be cleaner. It would be cleaner. Yeah. Yeah. Yeah. Yeah. [SPEAKER_02] I'll give it a new name and it's a new journey and all that. And, and I just felt like the story of, you know, it makes total sense for someone to go transactional, go to marketing, go to marketing and Hey, our deliverability is great. Let's also do this. Yeah. But then chat was like, Oh, what you offer chat. [SPEAKER_01] And then, um, that's what I was always confused. I was like, weird. That's why I was always like, you've got to add all the other tools. And it's like, this is just your small business playbook or like just do the email makes more sense to me. Cause exactly. It's like, well, we get them on transactional and then we, then we, then our upsell is like, we get you on marketing too. Right? Like, let me get like now if you're gonna have a person who has more dedicated sales time, it's like, Hey, go to all our people who only send transactional and let's get them on their marketing teams. [SPEAKER_01] Right on this and vice versa and double your sales or 10 X, your sales and one account. Right. We're like, they were only using transactional, which is, you know, always less than marketing. And then, Oh, and then we, we get their marketing and now we have 10 X their volume. Right. So yeah. [SPEAKER_02] And I had so much more awesome. And I had so much more personal confidence in like the marketing of the transactional product anyway. I'm like, and, and, and again, like the support we don't fruit it. I love it. I think our AI is good enough. Um, but like, could I, could I sell it against? Is it help spot or could I sell it against intercom? Like I just, I just have no appetite for that. [SPEAKER_02] So it's, um, yeah, so I'm going to sunset it, but in terms of like product direction, if we're going to talk about that, it's like, all right, where, where, where, where could, where could Bento potentially go? Um, there are other things and other channels. I think I will probably open up to, um, I've got like a PR open for push notifications actually. So like having the ability to do. Yeah. Yeah. So if like a company, cause we've got iOS apps and Android apps. [SPEAKER_02] And so it's more of that like multi-channel thing where someone sends a broadcast. [SPEAKER_01] Deliverability. It's delivering things. [SPEAKER_02] Delivering things. [SPEAKER_01] Right. So like that. Yeah. [SPEAKER_02] Yeah. So someone sends, cause you know, the Australian customer that came in, they send a lot of like deal type stuff and then they've got a lot of mobile devices. They need to do push notifications along with that. And they don't necessarily want a separate tool. So I think Bento will kind of start moving more into that direction where it's like multi-channel marketing and sending high volume messaging of different types. Um, but still even that, like I need to work out what my appetite is for that. [SPEAKER_02] And a part of me also wants to hold back a little bit just to push email as far as I can. Cause email can actually go really far dedicated. So for sure, it's kind of, we're in like a weird world where it's like in a world where you can build anything. You really gotta have some self constraint. Right. Where you go. [SPEAKER_01] I mean, this is every founder's problem, right? You're always like, Oh, I can build this other thing. And like when you really, you should just be doing. Yeah. You should just be focused on email. Right. Because like, yeah. Right. That's already doing well. And like there's gonna be all these bots. They're gonna be sending email. And the thing is that like, you don't want to build your own email sending system. Like nobody wants. Yeah. So people are going to continue to need solutions for delivering email. Right. [SPEAKER_01] Like that's to me is like a, a no brainer, you know, that's like a market. [SPEAKER_02] Yeah. And like, um, with, uh, the Jason Cohen book, which we'll have, uh, mostly technical after dark book, book chat about, um, he's got like a section that called like find your Carol, which is like basically find your person that you're actually talking to. And he's got a great actual, right? Yeah. He's got a skill for this, which, um, I recommend if people can find it to do it, but basically it's a whole bunch of questions. And like, if you give it, if you give your agent access to stripe, it'll even pull in all your actual real customers. [SPEAKER_02] You don't BS yourself, which is great. So, so like, yeah. So I was kind of like, you know, it's like, it was like, oh, um, when people cancel, like, what do they say? And like, this is an example of a question. So I made it pull a Slack cancellation messages actually from discord. I pulled that. So it has that. It was like, which customers represent the most of your revenue? And I was like, the top of my head, go hit stripe. So pull that in. [SPEAKER_02] Um, so that, and after that exercise, you then start asking him like, all right, well, this Carol entity that is my ideal customer, like, does she actually want bento chat? Not at all. Like, no, no way. It's like not even a question. Yeah. Um, does Carol want a calm email experience where she can deliver a whole bunch of things and her jobs on the line? And you know, she doesn't want to send a bad email. There's a whole bunch of features that we have that make that perfect for Carol. Right. Right. [SPEAKER_02] Um, now does Carol need push notifications before I thought she did, but now I'm actually like that PR may actually have to close, you know, maybe Carol doesn't want that. And so that, that, that, the exercise in his book was very helpful of giving me a, uh, something to look through when I'm looking at these features that I'm just generating and going, does Carol actually want this? No, she doesn't. And really I should stay focused on my mission, which is delivering the best email ever, making videos on email, all that. [SPEAKER_01] So yeah, this is what, like, I find it all over Twitter last week has been these people who are like, Oh, I just have it all automated. So like a customer emails in and then the thing gets built. And I'm like, I don't understand how to build software this way. Like you're just building lots of features and shit. And like, that's not what the people who are paying you want. Like, so yeah, I love, this has actually been one of my biggest issues all time with help spot is that like, what we mostly have relied on is like the concept of help desk software is a known thing. [SPEAKER_01] And so like our customers are crazy all over the place, like colleges and banks and car dealerships and little random tiny software companies, like not too much software. Like most of these real businesses, like not real, not that software isn't real, but like physical world, things like that. And so like, and there's not like a lot of overlap between them because it's just like, they need help desk software cause they need to answer email. Like the, the, the unifying thing is they get email and they need to answer it. [SPEAKER_01] Um, so when you try to market to that though, it's a little bit weird because it's not like, uh, sort of an industry or even a, uh, even the same use case. It's like an HR department or it's like an IT help desk or it's like a customer facing support for an e-commerce company. And these are all like really wildly different use. It's like the things coming in over email, but like, that's the only commonality. Like, it's not like they're, they're asking a wholly different types of questions and whatever. [SPEAKER_01] So like, so I do want to do this Carol thing. Cause I'm kind of intrigued on like in this new world order, we're like having SEO for help desk software is probably not going to like be, it's not very useful. Um, and with the addition of the AI features, which gives us kind of a new thing to push on in different ways, I do want to try to come up with more of a like ideal customer, even though I still think we'll have mostly a pretty broad base, but like even just that for the [SPEAKER_01] marketing to have a focus, even if it reaches other groups, like the focus is this person, you know, I think that's a, you'll, you'll love this. [SPEAKER_02] You'll, you'll love it. Yeah. You'll love, you'll, I think you'll, you'll absolutely love the book. And like, you'll love the find your Carol thing because I felt so much like you prior to doing the workshop. Now I still feel like you're a little bit like I, um, I feel a lot, I feel a lot clearer in some angles. Yeah. And actually like after we talk, that's like kind of what I'm, I'm thinking about today. And, um, you know, like I think a trap that I run into is kind of position myself as like [SPEAKER_02] the only one email platform and all, cause I'm doing what you're doing where it's like the source of my revenue, such a large portion is Carol. And, uh, for me, Carol is actually usually like a founder usually buys Bento, but Carol [SPEAKER_02] is usually the one that's sending the regular newsletters, doing the marketing automations and the one who her job is on the line, right? Like if deliverability cracks, Carol loses her job. Right. [SPEAKER_01] So, so this is interesting though, because like we have a similar situation a little bit, but it's like the fact I feel like you reach the founders and you, this is why X is working well for you. X does not work well for me. X works well for you. I'm getting a drink. One sec. Yeah. Go for it. Okay. Okay. Okay. So you, the founders are the people you're reaching and you relate to them. They relate to you. [SPEAKER_01] Uh, but to actually reach Carol without going through the founders is quite difficult. So I'll be curious to see if you've, uh, cracked that nut of how you do that, or maybe you don't need to. Right. And it's just like, you're, but X you're dealing with a lot of like more software type companies, a lot of founders and X as a primary path with maybe some YouTube sprinkled in at some point, uh, along with like kind of founder oriented things like mostly technical and things like that. [SPEAKER_01] Um, that the founder angle is still just the best, uh, for what you're doing. So that's kind of intriguing. [SPEAKER_02] Yeah. But I think the thing, I think founders still resonate with the stuff that, and like the founder can be Carol, right? And it can kind of like be a little bit of both. Um, but the things that deeply matter to Carol should also deeply matter to the person Carol's reporting to, who's probably the one that's procuring the software. Right. So like a lot of people find Bento prior to doing transactional marketing. A lot of people found Bento when they were in the shit with another tool. [SPEAKER_02] So they were using some company, their deliverability went from 30% to five. They didn't have any clarity and it could may have not even been their fault. Right. But they didn't get the support or the feedback. Yeah. It doesn't matter. And then they came to us. We spoke to them, we onboarded them and then they felt kind of good. And this is what happened with that large customer that came from the video. Their previous was an enterprise does the whole multi-channel thing. [SPEAKER_02] And they came to us because they had a lot of problems with Outlook and they weren't getting into Outlook inboxes and their previous provider wouldn't really tell them why. Yeah. And we basically helped them warm up properly, correctly, gave them way too much information and they felt safe and well prepared. And then now they're moving to us, even though we potentially have less channels for them to tap into, but they felt better. And they very much fit the Carol that I had built in my little workshop and stuff. So, yeah. [SPEAKER_01] Yeah. Dude, I feel like that is like the home run angle for you. Like the deliverability. I don't know anybody who has a newsletter who isn't like, well, my deliverability used to be 50% and now it's 30%. Like that's like all of them. That's everybody. Like I don't know if it's just a natural thing that happens with lists over time or whatever, but I feel like there is a, that is a pain point angle that you need to like stomp on. Cause like we fix your deliverability issues. I feel like that is like a big one. I think, I think, I think, I think, I think you're right. [SPEAKER_02] And I think I need to change. And, and, and, and, and I think we can do like, we can wow people and like they'll come in from that messaging and the thing that we're built has resonated. Um, and then a lot of the content that I can make, I, yeah. So I guess like this Carol exercise, it basically, once you know who you're talking to a lot of your go-to market makes sense. Your position makes sense. The content you make and the videos you make and how you talk and all that makes sense, how you nurture customers into the product. [SPEAKER_02] So yeah, anyone's listening, Jason Cohen, find your Carol workshop is the Google search or the chat GPT search. And then also get his new book that he just released. And it's on Spotify for free. So if you do Jason Cohen, you'll find his latest book. Yeah. [SPEAKER_01] You can listen to it for free. Yeah. I already bought the audible, but I did not start it up just yet, but that's on my list here. So I mean, I've been waiting for him to do this book forever. Like, I mean, his blog is like indispensable. Of course, like he's had like the best articles forever, but I feel like, yeah, I feel like he's like between this book and he started a podcast. I'm like, yes, give me Jason Cohen in the modern era. That's what I'm talking about. So I'm very excited for that. [SPEAKER_02] Did you see him rip? No, I guess he kind of did rip Jesse from dropping blog yesterday. [SPEAKER_01] Aaron said that I didn't listen to it. Yeah. But Aaron said he was like, it was like, it was a little uncomfortable. I'm listening to it. [SPEAKER_02] I love, I love Jesse. Like, um, he's, he's a, he's a good friend. And he was, um, he goes, Oh, I just chatted to Jason. I need my head. I'm like, Oh no. Um, and, and Jason just basically like rips into him, which I think is, um, it's good. It's like great content, but I listened to it. Like someone tagged someone's like, Oh, get Jesse on. I was like, my ego is too fragile. [SPEAKER_01] Oh my gosh. No, I'm, uh, I'm definitely going to listen to it. And that is what I heard. It was, it was intense. So, um, yeah, I have a lot of Jason Cohen queued up, I guess here. That's good. I love the Jason Cohen. I mean, he's a guy who's done it. He's legit done it. There are all these people on Twitter who pretend to have done it. Jason Cohen. [SPEAKER_02] I feel like as well, there's been a, there's been a famine of like good, deep bootstrapped content for a while. Um, even since like, you know, since like COVID and stuff, like, I feel like, like 2015 to 2019 was like a really good era. And then I don't know, things started to get like a little bit. [SPEAKER_01] Weren't you tweeting about this? You tweeted something about like the lack of bootstrappers in general, right? Like to somebody recently. [SPEAKER_02] And I, yeah, I think there's a, what there's you and Aaron and, um, um, and, um, my podcast feeds like fairly bare. Like, um, there's just not too, too, too many, um, bootstrapped podcasts to listen to now. [SPEAKER_01] Um, I think the money got so loose. There's just straight less bootstrappers, like people who are like at the go getter level to survive as a bootstrapper. Like they just went and got money instead. And so like, then they have money and then they're doing stuff with the money. So like, I think that just like, I think that's like a big part of it. It's just like, whatever in the before times there wasn't just money everywhere for everybody. And now it's just so much easier to get somebody to give you a couple million bucks to like try your idea that just changes the dynamic. [SPEAKER_01] Or then you have like tiny seeds, which are sort of like a lighter variation of that. Like there's just all these different ways that you can raise money that I think it's changed it a little bit. I don't know if it's going to last forever. Even, even if it's like, doesn't go away totally. I feel like it's so focused on AI now. I do feel like it's dried up a little bit. If you have a non AI idea, um, you're probably might need to bootstrap that now as opposed to in the, you know, two years ago, you probably would've got 2 million bucks pretty easy. Well now, no, go do it yourself. No, yeah, no, you gotta do it yourself. Um, all right. [SPEAKER_01] So we've been out a long time. One, one, at least one more. I'm happy to talk about other topics you have, but I wanted to talk about camera stuff because you're a camera guy. I'm a camera guy. I think you shoot Sony, right? Is that correct? Oh, is the bot getting you? Are you okay? [SPEAKER_02] No, no, no, no, no, no. It's just, just wait, just wait. Let me make sure my, my, uh, [SPEAKER_01] We have a live emergency here. We can see no pressure, but if you solve something in real time here, that would be really cool. You know? [SPEAKER_02] No, my wife's asking if our daughter needs an axolotl, which is the last thing I thought. I was, what if she's locked out or something? [SPEAKER_01] Um, an axolotl? What's an, okay. Do I not want to know what this is? What is, what is it? An axolotl? You don't know what an axolotl is? [SPEAKER_02] No. [SPEAKER_01] That's, that's what it's called. Now you're making me question. An axolotl? An axolotl? An axolotl? Is it a waddler thing? No, axolotl. Axolotl. I got three kids. I don't know what it is, man. Maybe some Japanese thing. No, you don't know what an axolotl. All right. Show the TV. What do we got here? Oh, yeah. You know those little guys? Yeah, yeah. Am I pronouncing it wrong? I have no idea. I think, I think they, I think they've become popular. My daughter's 13 now. I think we, we might've missed, uh, we might've missed that one, but we never had one of those. [SPEAKER_01] I don't know. Does she need one? [SPEAKER_02] No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no. You just got some fish and that was a questionable decision. So, um, what, what did you ask before the axolotl incident? Uh, camera stuff, camera stuff. You're a Sony shooter. [SPEAKER_01] Am I correct in that? [SPEAKER_02] Yeah. Yeah. Yeah. I got, um, what is it? Uh, uh, uh, FX 30. So the cinematographer. Yeah. Yeah. The video one. Um, and I feel like prior to doing like all the business stuff, like when I, you know, I used to love shooting and I used to like love my camera. And then when I was traveling and I did like a little bit added a 6,000, a Sony a 6,000. Uh, mirrorless and like love that camera. Up sensor. And then, yeah, I feel like the hobby died over COVID-ish. [SPEAKER_02] And like, I took me a while. Yeah. And, um, even with like the videos and stuff, like I got this cause I wanted to record my, my daughter, um, and just like small moments. So when she, she ran out of this to go see all of her presence over there and I've got like that great moment, you know, on camera. [SPEAKER_01] Yeah. And like in really high quality video. Yeah. [SPEAKER_02] Yeah. Yeah. Yeah. And, um, I really think like when you have kids like having a good camera and not using your phone, just, I don't know, you can capture really beautiful memories with that. Yeah. Um, and so, yeah, I've got that. I've got, um, what's the, what's the main lines? I've got like a Tamron 17 to 70 millimeter 2.8. That's what I'm doing on now. Um, I got this zoom one. Oh, is that the 70 to 200? The G-MOS the ones. Yeah, exactly. Yeah. You got it. [SPEAKER_01] What, what, what, what did you do? What did you do by last night? Uh, well the other day I bought this guy. This is the, the, uh, 302.8. [SPEAKER_02] It's a big mortgage out for that. [SPEAKER_01] That's yeah. Look at that. It comes with this weird kit. It's not quite as big as it looks, but, um, Um, that's cool. I don't know why they don't, it doesn't come with a normal lens hood, but, uh, that's, if you take all this stuff off, it's actually, I mean, it's big, but it's not that much bigger than the 70 to 200. It's a little bit bigger. It's got a big old glass on it. That looks so cool. It's very cool. So that's, see when you get the horse, then you gotta get the lens to go with the horse [SPEAKER_01] so you can take the good horse picture. So when your daughter gets a horse, you get the good, you get the good lens so that you can take the good pictures of the horse and the daughter jumping over things. So this is, this is a horse related purchase. [SPEAKER_02] Um, but the photo you took of the, the ocean the other day, did that have a mist filter on it? It looked like it had a mist filter on it. Okay. It had a long second one. [SPEAKER_01] Yeah. Has a mist filter. So it gives it that like little bit of like, um, hate hazy kind of to the highlights get a little blown out kind of thing going on. Yeah. Yeah. [SPEAKER_02] So this one and, uh, the, all the videos I shoot all have a gold mist and that, that's what gives them all. I thought that's what that was. Yes. Yeah. So, so, so, so all the videos that I shoot, all the videos I shoot have like a gold mist. [SPEAKER_01] I don't have a gold filter. Yeah. [SPEAKER_02] Oh, it's so cool. Yeah. So the gold mist works really well with kids as well because it makes them look kind of, um, dreamy and like softens their face features. Yeah. Yeah. Yeah. So all, all the videos I shoot have, um, like the gold mist filter on and it makes, it makes it look softer. It makes like complexions a lot like nicer. Um, so yeah, no, it's fun. [SPEAKER_01] This guy's got the eye. Yes. Yeah. I, I, I, um, so I didn't used to be big on filters, but what changed that for me was magnetic filters. So now with magnetic filters. So there's a couple of different companies that make them, but, uh, I have a case K A S C magnetic filters. And so you put like a mount on the front of your lens, like a little thin ribbon of magnetic, just, you know, it screws on like a normal filter, but it's magnetic. [SPEAKER_01] And then you could just magnetically take on and off like mist filters or ND filters or circular polarizers. And like you just instantly take them on and off. And to me, I like just so much, I always just hated like spin the stupid thing on, spin it off. It's such a pain in the ass. Uh, but the magnetic ones are great. And then the cool thing is you can get, uh, caps for them. So here's, here's the stack of magnetic filters, like the ND filters, mist filter, whatever. I'm looking at it now. Yeah. [SPEAKER_01] And you can get a cap for it. Yeah, it is. Well, photography is not cheap. Um, raise your prices there. Raise your prices, Jesse, but you can get a cap on both sides. So look at, so this whole thing just goes in your bag. So like in the past, you used to have the store, each one separate. You had this big stupid case to carry around. Now it's just this hockey puck of filters. You just throw it in the bag and you're done. So that I love too. Um, so that's opened that up more to me, but, uh, uh, this is cool. It's cool. [SPEAKER_02] I'm literally gonna, I'm gonna buy today. [SPEAKER_01] You're 70 to 17 to 78. You got there, whatever. 1780 you got there will be perfect for it. Probably like a 62 or 67. Um, yeah. [SPEAKER_02] Well, even, um, yeah. And I think like, I don't know, just like experimenting with my camera, getting like all these filters, buying the lenses and all that. It's just like, it's like a very fun thing to do, like with the business. And it feels like an actual creative outlet for me. And I know it's like a creative outlet for you. It's just nice to do something that's not spreadsheets and code. Exactly. [SPEAKER_01] You know? So yeah, I gotta, you gotta get the, I think most, a lot of founders have like the creative energy, but it's like, I think photography is a nice release for that or video or those type of things. And also like the other side of the coin, this is where the poker comes in is I have the competitive energy and it's like, uh, so business competition. Competition is interesting, but it's also like a long timeline. You know, it's like, well, you do this stuff and takes years and whatever, but it's like poker. It's like, no, we can get in there. I can get this competitive energy out like tonight. [SPEAKER_01] Like we can just go to battle and compete on the table. So we gotta get you into poker. We're gonna get you magnetic filters and into poker, Jesse. [SPEAKER_02] I think it's banned in Japan. So I may have to, if I'm going to practice, need to go into like the, the, the back. They got like weird systems here as well to kind of like get around all the, the bands. Oh yeah. [SPEAKER_01] Uh, I know in Taiwan they have it. If you can hop, hop down to Taiwan, you could play some poker or, uh, I think Hong Kong has it. And then there's like, uh, Macau and Philippines is big. So if you want to take a trip into the broad region, I just saw something like last year that Japan's going to open their first casino somewhere. I don't know if you've seen that. Like, yeah, there's no casinos. They're going to open their first one. I don't didn't say the article didn't say we're going to have poker though. So I don't know. Have you, have you been to Japan? No, this is high on my list. [SPEAKER_01] I want to go to Tokyo Disney. I want to go just Tokyo is like the most amazing place to take photos. I'm like the whole countryside. My son, um, my oldest kid next year is trying to go for study abroad. So if that happens, I might, uh, I might hop over. Yeah. [SPEAKER_02] Well, hit me up is, uh, does he want to do, um, I'm in a Fukuoka. So I'm down South in a Kyushu. Okay. So I'm on like the bottom part of the islands. Um, but this city is like, it's very, it's very, very cool. I really like this city. It's like a lot calmer, a lot more chill. It's still got like all the big city stuff. There's no tourists here as well. So like I'm kind of the only foreigner in my neighborhood, you know? [SPEAKER_01] Um, which is, which is fun. He is, um, it's not in Tokyo. His study abroad. It's going to be, if he does it, it's somewhere in the south of the main island there. Is that where, is that where that city is? Yeah. So like I'm lower, lower. Okay. Lower. So maybe we're not there, but, uh, do you have a baseball team? Where is going? Uh, yeah. The Seahawks. The Hawks. Okay. Hawks. Hawks. [SPEAKER_02] All right. They're like the, the winning team. They're like the, the strong one. I've got like the, the, the pay pay dome. I went to, I went to one game there and it was like, it was very, very cool. [SPEAKER_01] So yeah. So how, just quickly, how did you get to Japan? Is your wife Japanese or you, or no, or you guys just decided to move there? [SPEAKER_02] Um, yeah. So my wife's not, I met her on Twitter. I met her on X. [SPEAKER_01] Um, so I was doing like the, I was, no, no wonder I'm a grok fan. No wonder you're a grok Twitter super fan. Yeah. [SPEAKER_02] Yeah. No, I, um, what is it? Cause I was doing like the nomad things. I was working out of my, I had a little, uh, what was it? Tortuga, Tortuga backpack and had my little laptop in there and my roof stand. You were just cruising around. Um, yeah, I was doing, I was definitely like, uh, Peter levels pilled at the time. Right. So, um, yeah. Uh, yeah. So it was like doing the traveling thing. I met John O'Norland like on the, on the road as well. So, um, John O'Norland. Okay. Yeah. Yeah. [SPEAKER_02] It was kind of like that, that, you know, everyone's trying to work out what remote working is and like, how do we, uh, do these weird things online? And then, um, my, uh, Michaela had a, like she had a YouTube channel. So she, she, she was a popular YouTuber over here. And, um, yeah. And then on Twitter, I like saw that she had like an apartment available or something like that. And we started chatting and then, um, I eventually came over cause I would always like travel to meet people. I would just, you'd meet someone online and then just go. [SPEAKER_01] Um, that's crazy. That's so cool. [SPEAKER_02] Yeah. She even like booked a call once to talk to me about like a client and marketing stuff prior to me even coming over here. Right. Cause I had like my demo thing, uh, available for anyone to book a time with. So she did. Um, and then, yeah. And then I came to Fukuoka and I was like, at the time I'd been traveling, I think for a while. And I was like, all right, business is getting busier. Um, I'm tired. I don't want to like keep getting booted out of a country every 30 to 90 days. Right. [SPEAKER_02] Um, it's just very, I don't know, it gets tiring after a while. Um, and so, yeah, she had an apartment available. And so I moved in and then we fell in love and then I just stayed. And then we, we, we, uh, had Evie and, and, and now we're raising a daughter here in Japan. I dropped her off at, um, Japanese, uh, kindergarten around the corner. It's. [SPEAKER_01] Wow. Do you speak Japanese or no? Have you picked up some? A little bit. Yeah. [SPEAKER_02] No, I speak a little bit. I speak like a child. Uh, Michaela is like fully fluent. She's been on like national TV and all that kind of stuff. So she's, she's, she's good. And then my daughter is getting very fast. Like, so she'll, she'll switch. Yeah. So like she, it's a Japanese kindergarten, so they don't speak any English there. Right. Yeah. So once, once I drop her off, she's like in the, the D band. Um, this is the best. [SPEAKER_01] That's the only way to learn, you know, like this is why the people, kids grow up with it. They just, they're, they're so good at both, you know? [SPEAKER_02] Yeah. And it's like seeing, so English at home, Japanese kind of whenever she sees Japanese people. And, um, she seems to be doing like really, really, really well. So yeah. And like to round this out, Japan, I feel like is, it's, it's like a very safe country to raise a family and also run a business. Um, they could improve on some tax stuff here, but, um, net net, the amount of money that you save and like cost of living and safety and just the whole system, it does make it like [SPEAKER_02] a wonderful country to raise a family. [SPEAKER_01] So yeah, no, I like it. I gotta get to Japan, man. This is like probably, I think it's my number one country. I haven't been to that. I would, uh, that or Australia. I haven't been to Australia. So between Australia and Japan, those are the places which I would like to hit them. Maybe even on the same trip, just like we're just going over the other hemisphere. We're gonna do it all. See it all. Uh, they do a poker in Australia so that you guys know what you're doing down there. But, um, yeah. Yeah. Yeah. [SPEAKER_02] There's a lot of gambling in, in Austin. One, one, one thing on, on the Japanic. Japan's got a lot of great super family hotels. Um, I'm organizing one actually at the moment with a couple of like bootstrapping friends with their kids. It's just me and Evie, my daughter and them and their kids. And we're going to go to like one of these Japanese super family hotels where they've got like everything like pools, game centers, unlimited buffets, like the whole thing. Yeah. Yeah. Yeah. [SPEAKER_02] And so, um, yeah, I don't know if you come with your family, maybe we could organize like a, like a little place this way. [SPEAKER_01] Yeah. But he hang out. Yeah. That'd be cool. We got, we bring Aaron and he's got kids, your kids age. So that would be cool. You get the like play buddies. Um, interesting. I did not know that. So yeah, I know. I think I could spend the month in Japan and then like, I was hearing about it forever ago, like Patrick McKenzie, I knew way back in the day and like, so he was famously from Japan or from Japan, but lived in Japan. And, uh, so I don't know. I've always been fascinated by it. [SPEAKER_01] It does seem like, like even if you don't, like it's not the kind of country that everybody speaks English is my understanding. So that's not, it's not like, uh, in that way easy, but it's as easy cause it's like super safe and people are happy to help you and things like that. So it's not, uh, you're not totally, uh, on your own. Yeah. [SPEAKER_02] You're not. And it's, uh, you know, your phone's great. You can just use your phone to get everywhere. So it's not, you have the translation app and everything. Yeah. But if you know people in Japan, they'll get you away from the tourists and they'll get you to like the nice spots, which is, that's the hack. If you know, if you know, like where to go and where to eat and all that, um, it's, it's super amazing. So yeah, I recommend, please come. [SPEAKER_01] All right. We're gonna have to make it happen. It's definitely, it's definitely possible it's happening next year. So if that happens, I will definitely be swinging by to see you. Cause that would be, it's always so fun. Like it sounds you've done, you've done a lot of this. I have not done a lot of it. I'm like going to someone's home country and, uh, who you know from the internet and hanging out. So I love it. I love it. I haven't been failed on doing that, even though it's weird. [SPEAKER_02] Everyone's nice. [SPEAKER_01] I feel like there's no doubt. So it's going to be all wins. Yeah. You're getting the inside scoop. It's just cool to be together. So that's great. So, all right. It's been two hours. I think that's good. We covered the stuff. Thanks so much for coming on for the sweet hang. Aaron's going to be very shocked. This is a two hour podcast. The people are getting their money. We did two episodes last week. We're doing a two hour one this week and two episodes. The people that can't complain about the value they're getting. Thanks to Bento for the value. Thank you. [SPEAKER_01] And, um, yeah, no, this is great. This was awesome. And we covered a lot of cool stuff. Thanks so much for coming on. We will do the book review maybe next month at some point. And, uh, yeah, it'll be fun. So everybody out there, if you're listening, if you made it this far, buy Jason Cohen's book, listen to it, join us for the book club, book review, mostly technical book club. Uh, we'll, we'll do his book, which will be super fun. Um, and yeah, thanks so much for coming on. Really appreciate it. Bye. Thank you everyone. See ya.