- Platform: meet
- Source: Read.ai 01KTSFQBY8KZTM8HNJ7C1FNM30 (via Ben Ornstein’s account, imported 2026-08-06)
Transcript
Alan Thomas: Congratulations, dude. That’s amazing. Ben Ornstein: Thank you. Thank That was pretty recent. And I’m heavily involved in of the AI stuff that Hauk is talking about. So it’s been a ton of fun. I’m sure you know, if I saw it, I’m sure you saw it that Will Remnick just got engaged. Alan Thomas: I did. Yes, sir. Ben Ornstein: That’s our mutual friend, Hauk. That’s how we Oh, cool. UNKNOWN_SPEAKER: That’s dope. Bro. Alan Thomas: We’re going to Dan’s uh bachelor party this weekend, so you know, it’s uh, everybody’s getting married, bro. I got another couple years, I’ve been dating my girlfriend for like so that around that three and a half, four-year mark, we’ll uh we’ll talk about it. Ben Ornstein: No, that’s awesome! Michael Houck: Good for you, brother. Um, okay, well, let’s let’s dive in. I don’t think we’re gonna again need the full time, but Alan, just for some context, um, some of the like Ben has basically become an AI engineer at this point. Like he leads all of our agentic engineering and product stuff now across the whole business, which is really cool. It’s been see him sort of dive deep into that over the past, like, what been like three, four months at this point? Ben Ornstein: Yeah, that’s about right. Yep. Michael Houck: Yeah. So we have this little team of agents that can dramatically reduce the amount of time it takes for and increase the volume of content that our team can create for our clients. So very briefly, there’s Casper, who’s our ghostwriter. There’s Eddie, who’s our editor. There’s Trent, who’s our trend spotter, looking into the X API and things like that. There’s Connie, our account manager, and then there is Red, our sort of client advocate, like voice of the client in these sort of agentic conversations that they have with each other. And one of the new areas we’re looking to tackle on. Is not just writing the content, but having visuals attached to it, right? And so we’re exploring that from a design perspective for like infographics and things like that. But what we thought could be really interesting is obviously memes as well. And so, wondering, like, I don’t know what it would look like or how much Ben you would kind of need on a back and forth basis, but it would be very cool to kind of set up some channel where. Just able to share notes basically back and forth about, like, hey, here’s what we think the structure of this agent could look like if it’s like generating ideas for and producing actual memes that are funny, relevant, et cetera, for the client. And then, Alan, you sort of like, you know, poke holes in it, tell it it’s stupid, like, you know. And hopefully it becomes actually useful for you too, right? Like, our goal, our barometer for like, do we ship this would be like, do you actually think it’s useful and get use out of it? Alan Thomas: Okay, um, so just some clarification: do you want it to generate the visuals from scratch, or do you want it to tap into a database of already existing memes and then try and pull them for relevant people? Yeah, okay, because I was gonna say, banking on it to make funny images would be a little tricky. But yeah, I would that second part definitely makes sense. I use Meme Lord69. I don’t know if you guys have heard of it. Yeah, that’s like one of the best databases. And shout out Jason because that guy’s brilliant. If there’s a way to link that to what you just said, then that would be dope. Michael Houck: Yeah, I have to imagine there would be. I know Jason pretty well. So I can reach out to him and see if they have some kind of API that we can use. And then we would just need to choose the right meme for the situation and add the right sort of caption on it for that specific client. And that would basically be the game that the agent is trying to solve, essentially. Alan Thomas: That makes sense to me, man. I mean, I think the trickiest part would be explaining like context to the AI where it’s like, why is this image used in these specific examples? But I feel like if we give it like three or four examples and show how it’s being used, it be smart enough to understand, how to do that. Because again, if you just show an AI like a random picture of whatever is trending, it’s be like, why is that funny? That’s the trickiest thing with AI, is humor, man. Not there yet. I’m sure it will get there. But also, that definition of what is funny is different for everybody. So, it’s a tricky one. Michael Houck: Yeah, I think this is an area where I’m not quite as optimistic as I am with some of the other agent jobs we’ve been setting up, specifically for that reason. Like, the humor. Contextual humor, the cultural contextual humor, I’m not sure if it’s going to be able to grab that at like the level we would need it to. So that’s why this is like total experiment. Like, let’s throw some shit out there. Let’s spend a couple and cycles on it on our end and see where we get essentially. Alan Thomas: Can we train it? Could we give it like three or four Twitter accounts that are kind of consistently funny and be like, this is the tone of voice that I would try and lean into? That might be a decent idea just because, you know, like you there might be a spectrum of humor. But if there are a certain batch of accounts that are going viral consistently and using trending images already, we can maybe give those as examples. Michael Houck: Yeah, Ben, what do you think about that? Ben Ornstein: Yeah. There’s APIs we can use, or I can literally just manually pull the content from those posts and throw them in like a markdown file and train it that way. So there’d be ways to do it. The visual part, like the tone of voice in terms of like the actual text, that’s the easy part. Images, I’m not so sure about. Michael Houck: I guess like on the images side, if we are connected into like a database, like Meme Lord’s database, right? Each one would have like an ID in there. And we wouldn’t be training it to like read the image necessarily. We could just train it on like this ID with these types of caption structure tends to be funny in these types of situations. You know what I mean? 100%. UNKNOWN_SPEAKER: Okay. Michael Houck: Yeah. Alan Thomas: Because the images also have to be trending, right? Like it’s a little bit weird to send content to clients of images that are outdated. Because even like a week is a long time. So I like make the AI jobs easier by saying, if it’s past five days of relevance, ignore. So that might be making it a little easier for them. Michael Houck: That would be a lot easier because that would considerably cut down the like set that it’s considering at any time. So what we could do is we could hook into the XAPI or I mean, maybe even like, Instagram API or something. I don’t know. Alan Thomas: I don’t think the easiest thing to do would be: okay, whatever is not on the first page of Memorial69. Like, I’m serious. Like, he’s done such a good job at aggregating trendy images. Like, it’s an absolute steal. So, he’s already built it. So, definitely would use that. Michael Houck: Yeah, if there’s info like that already built that we trust, then 100%. Let’s just leverage that. That sounds great. Ben Ornstein: Yeah, we could use a separate version of Trent. To like search as well. So, like, that would be another filter. So, that could be more. That could be, that could be interesting. Michael Houck: Okay. Um, Alan, what else what are we not asking about that we should know of, like the art of the game, so to speak, in terms of how we would design this guy? Alan Thomas: dude, it’s timing also, right? Where it’s like, okay, something happens, something is trending. The earlier that you use it, the higher upside that you have. So it would almost try to tell the AI prioritize images that are trending in the last hour, three hours, six hours. Because again, like it’s funny, like I posted content, like the exact same image in the exact same copy, and it flopped one time and then it went viral the other time. And I’m like, bro, how what am I supposed to make of that? And the only thing that I can think about is timing. So that’s something that would be a little tricky to figure out what that looks like, but I would even 15 minutes to like 15 minutes, hour, three hours. Well, things things just pop off, right? At a you’re watching the NBA and like it pants crowd and somebody is instantly a meme, right? And it’s okay now, that’s starting to be used. And the earlier, if I know like 15 minutes might be tricky, um, but the earlier that you can hop on a trending image, the better that it is for sure. Michael Houck: What would make this something that would actually be useful to you? Like, what would make you actually use this as part of your workflow? Alan Thomas: I mean, it would probably be like an event has happened, right? Like, we’re watching the World Cup and, like, something happened. And I’m just like, hey, like this just happened. Show me the relevant images in hour that could potentially make a funny meme. And you’re, you describe the situation. Like, it like a one or two sentence where it’s like the referee just accidentally said, or yesterday. Let’s just take yesterday for example, because there’s a viral video. A light literally fell from the soccer stadium and crashed onto the field. And never seen that before. We posted the video yesterday. It’s at like 15 million views. So, like, that could be something that, like, hey, this just happened. Show me the last 10 images on Memorate 69 that you think could be paired, right? It’s like a wine and a steak or a wine and an entree or pairing things that you think would be best for that. Ben Ornstein: That’s super interesting. Okay. Michael Houck: That could be, Ben, that could be an entire agent’s job. It’s just the pairing. Ben Ornstein: Literally, I mean, that’s enough of a job. There’d be enough context in that to its own. Yeah. Alan Thomas: Like you don’t drink white wine with sirloin or whatever right? Like you need, it’s got to be the right pairing for sure. Ben Ornstein: Oh my God. Yeah. Okay. Michael Houck: This is very, that’s a very interesting way to think about it. And actually, yeah, Ben, this gives me the idea that like maybe this is a little team that’s working together on like very discreet jobs within this job, you know? Yeah. I don’t know. Like as you get into it, you can sort of map out both directions and we can see whether it’s like one guy with a lot of skills or different guys with like one or two skills each. But Alan Thomas: And also have it feed into the aggregator account of like, it needs to be following certain Reddit subs, because where does the news break? Right, how is it supposed to even know that something happened on the TV on YouTube TV or wherever you want to watch it? It needs to have access to wherever the news breaks. So, I don’t know if that’s like specific subs on Reddit or specific accounts on Twitter, wherever the breaking news is, it needs to have some kind of access to that. Michael Houck: So, you would recommend, like, not just, for example, with our trend spotter for like topics to talk about for our clients. We have the guy who’s integrated with the XAPI, and I think he’s also integrated with Reddit and Google too. But what he does is every, I think it happens at different cycle times on different days or whatever, but like every 15, 30 minutes or an hour or something like that, he goes in and he looks at what topics have spiked the most since the last time he checked. And then if that topic is relevant to that client. And that client has something unique to say about it and it meets some like overall threshold. He presents that to the client as like, hey, we should comment on this. Here’s a draft of a post. Like, you know, okay. Alan Thomas: Yeah. I mean, that’s the, I mean, that’s not my business, but we do definitely need to figure out how to make it go quickly, right? Because when something is trending, you can’t have the client take all day to get back to you on something because then like it’s probably not going to do it as well. Yeah, so that’s just something to as well. Like the idea might work, but if the execution takes too long on the client side, it might not. Michael Houck: Yeah, so really, maybe Ben, this is an area where we just sort of ask the client to like give us carte blanche and we just have like sort of strict, like very clear guidelines given to the agent about like how edgy is this client able to withstand on their account. Ben Ornstein: Yeah, it’s kind of crazy because we have the ability to like straight up post the image with we could just automatically just post it if they do give us the cardi blanche. It’s kind of nuts that we can just do that with the agents, but yeah, it literally could be instantaneous, like thing happens. Michael Houck: And like if the checker is checking at that moment, it could be posted within you know seconds. So maybe that would be useful actually. If they give us that trust. Alan, do you normally just have everyone you work with sort of give you full trust and like running it? Alan Thomas: Yeah, I mean, I don’t have to get permission on an individual tweet. Like they, they might for like the first couple of days be like, hey, send to me what to post. And then if like the last 70 that I’ve sent them, they’re like, fine. Like, okay, just start doing it. Like at Polymarket, I just post on Twitter. Like, I don’t ever ask for permission. I know what is allowed, what is not allowed. And that works the best, man. Because, like, even yesterday, that video of the light falling, I found it in like the first five minutes that it happened. And then everybody started using it, right? Because like stitch videos on Twitter. So, like, we wore this, we’re the source of the video, which is why it blew up, which was sick. UNKNOWN_SPEAKER: That is sick. Alan Thomas: Speed is everything, bro. I think I’ve told you guys before it’s the F1 pit crew, man. Like, they’re in there like three and a half seconds or whatever, and they’re gone. Like, a 0.5 seconds, a one second, that’s costing you views, man. That’s costing you likes. That’s the truth. Michael Houck: Okay, Ben, maybe this is a job for Red once we have, and now that we have Red Live, like having him check on, like, not just have guidelines for what the client is willing to talk about, but have like meme-specific guidelines for the client in there, too. Ben Ornstein: Yep, that makes sense. That makes sense. Michael Houck: Okay. Michael Houck: Okay. I mean, look, I think this is meaty enough, Ben. Do you have any questions we should go through? Because I think on my end, this is meaty enough for us to sort of dive in and get started with. UNKNOWN_SPEAKER: Definitely. Ben Ornstein: Yeah, I might have things that come up, but as I like research and figure out the configuration, I’m sure there will be things that come up. UNKNOWN_SPEAKER: Cool. Michael Houck: Alan, can we bug you if we have other questions? Alan Thomas: Sure, man. This is good for me to even realize what’s being built. And I’m trying to be early on these tools. I don’t want to be late. So 100%, man, anytime. Michael Houck: Once we have something that’s like at least testable, we’ll set up a little Slack channel and you can just. Fuck around with it and let us know where it’s uh, where it’s breaking. Michael Houck: Basically, that’s perfect. Alan Thomas: Well, cool, guys. Sorry again, I missed the call last time. I apologize for that. Um, yeah, if you need anything in the future, just you know, let me know what it is. Michael Houck: Appreciate the time, brother. Talk soon. Awesome. All right, guys. Peace. You want to stay on? Oh, he’s gone. Ben Ornstein: There we go. When I was about to leave, I saw you say, like, like about to say something. I’m like, that was probably a cue for me to stay. Michael Houck: Yeah, I was I was trying to get it quick enough before you would hit. I would hoping he would hit leave and then you wouldn’t hit leave right away, but it was too, it was too close. Yeah, look, I think that’s exciting. You recorded that, right? Ben Ornstein: Yeah, my radio is in here. Michael Houck: Okay. So I have some little notes on my end that I’ll. Go through, but sounds like, I mean, it sounds doable. It sounds hard, but it sounds doable. Ben Ornstein: It’s another one of those things, like the design agent. If the execution is be very difficult, but if it’s done, the value is infinite. It’s that’s, and I feel like that’s where we should be living. We should be living in that area of like, this is going to be extremely difficult, but if we do it, holy shit! Michael Houck: Yeah, I totally agree. Um, and I think that having someone like Alan who’s willing to be like an alpha tester and like consult on it and not charge us on it is like, I think it gives us a unique edge. Um, I’ll talk to Jason and see if they have some sort of API because I don’t actually know. I mean, maybe you could just check that when you’re setting things up, but um, I’ll message him anyway and see if there’s some way to like, partner on this or something. I don’t know if there’s anything there. Ben Ornstein: Sounds good. Ben Ornstein: Yeah. I had these weird bugs today with Ani where she started, like, I’m still setting her up obviously. There’s a lot to configure. But she like shut down, and an error on, like, I think it had to do with Casper’s soul file. One of his files was too big. But it wasn’t even that big. There was like a constraint of 12,000 characters or something. That’s the first time that’s ever happened. So, and we’ve been over 12,000. Michael Houck: So, I don’t know if that’s something OpenClaw just implemented or no, oh, no, no, it that shouldn’t be the case because you can have up to a million tokens. Ben Ornstein: I thought it was um within the markdown file within the soul file, it was really weird, it was a weird error. And honestly, good because I had to clean up the soul file anyway. There were things that didn’t need to be there. They were listed twice. Somewhere shouldn’t have been in that file. So I did shorten it a lot and cleaned it up. But yeah, these bots are annoying me. There’s so many little bugs. They’re so, there’s like, they continue to be. Continue to be so many little bugs? I’m getting Venus messaging me saying Trent didn’t run, do the second run for proactive engagements for like two clients. I don’t understand why. Like, I don’t know why it happens to like two people, but the rest are fine. I don’t understand. Michael Houck: Yeah, I, I mean, these things are finicky. We know that, right? I don’t know. I don’t know. I think it’s. These type of things that, like, eventually you’ll want to just delegate to Raphael and just have him figure out. Yeah. But while a million things are going on, probably impossible to do that. But eventually, like, I do think that’s a good area where you can focus on, like, you can focus on the 80, let him focus on the 20, Yeah. Ben Ornstein: Yeah. Built like the new agent creation, I can definitely do. Like, I can spin these things up quickly, but they’re not like, they’re not great. Maybe my instructions aren’t. I blame myself. Like the instructions clearly aren’t clear enough because it’s a markdown file. It’s plain English. And so clearly, I haven’t defined edge cases enough or other things. But I mean, it’s amazing when these things do work. And it’s amazing to see where we’ve come, you know, where we’ve gotten to. But God, they’re so finicky. They do stupid shit so often. But and because that, like, I still haven’t gotten to the jolt stuff. I need to get to the jolt stuff, but I can’t get there with all these other things. Michael Houck: No, you mean, don’t we don’t need to be on the jolt stuff yet? It’s fine. Jolt, Jolt, it’s even launched yet. Yeah, there’s no, there’s no huge immediate rush there. Tatum’s time is not constrained yet, so it’s no big deal. That’s good. Um, okay, let’s use the next 10 to go through this voice stuff, though, because that is something I want to get live. Soon. Obviously, sure, you saw the screenshot. Like, we didn’t tell Raphael what to do here, so we shouldn’t expect him to do a lot, but the sort of system prompt that he’s giving the 11 labs agent is like, you know, I mean, it’s let’s put it nicely and say sparse. UNKNOWN_SPEAKER: Yeah. Michael Houck: Right. So basically, what we need here is a prompt that is going to instruct. The agent how to run, I think, these calls that we do texts. I’m not sure if the texts are associated or not. But basically, you know, most of the people who do this are going to be new subscribers, but initially, probably a good chunk of them are going to be existing subscribers too. And we’ll have the call link in the newsletter so people can do it whenever, basically. But the idea is we want to learn more about them and that’ll let us send them the right email campaigns. That’ll let us sell the right services to them. That’ll let us know what kind of content they’re interested in. And it’ll help when we want to set up an expert network later this year as well. So, our goal is information extraction, but like positioned in a value-driven way where, hey, we can help make intros to relevant people for you, or hey, we can help connect you with advisors, expert network investors. You know, basically, everything that Boardy pitches itself as doing, we just want to take that. That and personalize it around me and around Founding Journey. So we need instructions to give it for the call. I don’t think we need anything more broad than that, but it’s just like, what information do we want to get basically? So I can provide like those data points that matter most. I think there’s actually a thread. Maybe Blake can find it, where the three of us went back and forth on this a while ago. So, like, the playbook is there somewhere, but we just need like similar to how you’re creating these like notion docs for the strategy for the agents before we build them. We need something like that for this agent, even though we’re not like building the agent, we’re just sort of prompting 11 Labs’ agent system. UNKNOWN_SPEAKER: Yeah. Ben Ornstein: yeah. So, when Raphael and I tested this, um, basically, I mean, like, it did work in the sense that like he got it called me, um, it was going back and forth with me, and like answering the questions. Um, but like, it was like a very, you know, it was right asking the questions, and that was it, right? Michael Houck: It’s like it, it’s, it’s like, it’s like you’re filling out a form just with a voice, right? That’s not what we want. We want this to like, this thing is going to do tens of thousands of calls over time, maybe hundreds of thousands of calls over time. And so we’re going to want it to be able to handle a lot of different situations. We’re going to be able to want it to have conversations flow organically and naturally so that the person doesn’t, you know, stop after three questions. Like, oh, this I’m just, you know, filling out a form or talking to a wall. Like, when I’ve called, have you had calls with Boardy? Have Boardy? Ben Ornstein: No, I know who he is, obviously, but I haven’t done it. Michael Houck: You should sign up for Boardy and do a call. Like, spend like 30, 60 minutes on the phone with it and just like give it just ramble, give it your answers, and then kind of see the experience. I think that will make clear a lot of sort of what we need to consider for this and how the call should feel. Ben Ornstein: Got it. Okay. Is this higher priority than making the Casper and friends production ready? Michael Houck: No, definitely not. Ben Ornstein: Okay. Yeah, because they’re just, they’re not production ready yet. It’s like bothering me. There’s so many little things. But Raphael is still like, I still, I guess, reconfiguring the infra first. Michael Houck: Well, he’s bogged down with Emil stuff. So everything right now on the engineering side is delayed because Emil decided to ask us to help out with this engineering, which we because the margins on it are great and we’re making good money on it. But it’s just delayed everything. It’s delayed the landing pages. It’s delayed this. It’s delayed that. So I think that will wrap up by like next week. I don’t know when next week, but at some point next week. So, yeah, as far as I know, he has not. And I, and I told him, like, you know, since we’re getting paid externally for this, like, prioritize that first. So don’t hold out, don’t like go too far down the rabbit hole on that yet. And if you want to even, you know, sort of cycle back and forth between a couple of different things until he’s back on it, do it, you know, make that judgment call, you know? UNKNOWN_SPEAKER: Yeah. Ben Ornstein: Um, because Connie is, I have to go back and forth because that bug earlier today like took like an hour, so I got to keep figuring that out. Um, and then I assigned the red build to um to Raphael, so that red has not been built. Michael Houck: Okay, that’s fine. Um, when red is built, um. It’s not like we can use it for the meme thing, too, which is great. Ben Ornstein: Yeah, for sure. So, okay. Yeah, I’ll just that’s what I’ve been doing. I’ve kind of been cycling between different projects. Like, I’ve gone to one and then taken a step back and been like, All right, well, I’m kind of at a blocker for X reason. Michael Houck: Sometimes that’s helpful. I’ve been there too. I do that all the time, actually. Michael Houck: Um, but um, okay. Ben Ornstein: For the viral examples thing, so there’s one. The last here, I’ll show you it. I tried to explain it. I’ll just share my screen quickly. Just because this is like the last thing that’s stopping, I think, like the rest of that, all that stuff. So basically, in this referral program, We’ll definitely need to. Michael Houck: Sorry, I know how to fix this. I saw your message. I’m going to, I’ll take care of this like either tonight or in the morning or something like that. Ben Ornstein: Oh, okay. Got it. Yeah. And then, like, the copy, obviously, like, it’s horrible because I’m not good at this, but no, it’s you’re not like the worst copywriter, but please don’t think that. I know it’s not a strong suit of mine. And so, like, I would never ship something. Like, this is technically live right now, but like, I don’t think anyone’s. Michael Houck: Yeah, yeah, yeah, for sure. No, it’s fine. I’ll get in there before the issue goes out tomorrow morning. Okay. But also, look how great that graph looks, man, for viral examples finally. Ben Ornstein: Like, it’s, and so the big driver is we’re throwing ad money. Is that why? Michael Houck: Yeah. So basically, um, we’re spending a couple, I think it’s like, I think it’s like 300 a day right now on these. And I had it set at 200 pitch check database, 100 LinkedIn playbook. But now it’s be, as of like yesterday afternoon, 200 LinkedIn playbook, 100 pitch check database. So two-thirds of the new folks we’re getting are technically for viral examples rather than FJ, although we’re adding them to FJ too. Yep, because they’re founders anyway. But yeah, so viral examples is be growing. Today we’ll probably get close to 100 new subs and I’ll probably be growing close to 100 a day for now. And then next week, we’re going to launch the X playbook and the newsletter playbook. And so between the three of them, you know, we might be growing at close to 150, somewhere around there, subs per day. Ben Ornstein: It’s legitimately a great newsletter. Like, I like putting it together. And it’s like legitimately very helpful. Like I can tell. UNKNOWN_SPEAKER: Yeah. Michael Houck: people will reply to it sometimes. I think the click to post thing is so brilliant. I was like, blown away when you, it was like such a little hack. I loved it. Ben Ornstein: I was so excited. I was doing some research and I was like, oh my God, wait, this is going to work. So I was like, I was so pleased where it’s like, oh my God, click, boom. Then they can literally just edit right there. I was so happy with that one. Michael Houck: So, I think if we, if we can get this, like, what I would like to have happen is that we get this newsletter big enough where it can just start doing sponsor revenue too. Um, we can monetize. Obviously, we, what we’re really going to monetize on is people who sign up for Casper or sign up for Megaphone, right? But if we can monetize through sponsors too, it just makes us even more valuable. But again, it’s all dependent on the acquisition, right? Which I’m still. Working through. I have another comments to address tonight or again tomorrow morning. So we’ll see. Ben Ornstein: Is it getting close or is it still unclear? Michael Houck: Well, we all the tax stuff is wrapped. Um, and so now they’re just deeper in the like phase of their diligence process or whatever. So, um, it’s like new people on the team I’m dealing with now. Ben Ornstein: Oh man, well, I guess that’s a good sign because that means it’s progressing, maybe. Michael Houck: Yeah, I mean, it is right. But again, if we can nail this ad funnel ourselves and we can grow the business substantially over the rest of this year, like we need to see more leads come through here, though. And we need to see them convert to sales calls for Megaphone. So that’ll be a big, a big factor for us is like, does that happen over the next couple of weeks as we sort of wrap this up? Ben Ornstein: Got it. That makes sense. Michael Houck: But, all right, man, I got to run to this thing with Tatum and Emil’s PR guy. But I think we’re good. I guess I feel like you’re juggling a lot of different things right now. Do you have a good system set up to like crack that and not lose track of all? Like the meme agent, for example, is not a high priority, but like it is something we should do. So I just want to make sure you have it like tracked somewhere. Ben Ornstein: it’s kind of mental right now. I have it like juggling, but like I’m I’m I know like all the balls that are in flight, it’s just like yeah, yeah, action on them. Yeah, I’m trying to that’s why I constantly keep asking, like, what’s the priority? So I can like focus on that. Yep, so Catherine Friends is still highest, and then Hawkbot, yeah, I think so. Michael Houck: I think so. Um When so I use the Slack reminders too much, and I’ve actually stopped doing that to some degree. I still do it for some stuff, but I recently set up, I’ve always hated like to-do list apps, but I recently set up this app called Superlist. It actually works for me. So check it out. If you actually think it’ll help you sort of structure the chaos, you know, you can expense it. But if not, you know, don’t worry about it. There’s also a free tier, I think, too. Ben Ornstein: I pay for a called Todoist and it’s like changed my life. So I do that one for a lot of stuff. So, like, when there’s things I know I’m going to forget, I just put it there and then forget about it. So, okay, cool. Michael Houck: Yeah. I just feel like I feel myself losing track of some stuff. And so, if I’m feeling that, I imagine everyone on the team is sort of feeling that too. So, just wanted to flag that, but sounds like you’re on top of it. Ben Ornstein: One last thing. I saw your message about Slack or Stripe. There were the two separate ones about Stripe. There was the one to take the lead on with the Grok chat. That seemed like I read it once and was my head was spinning. But I’ll read it again. Michael Houck: Is that that’s low, very, very low priority. We just basically all of our revenue runs through the Rye Valley Stripe account right now. Now that we have the other one set up. I want that revenue to show up in the right bank account automatically. Don’t have to do like end-of-month calculations. And I want it to show up in the right spot on the PL automatically. So I don’t move that stuff around. So it’s not a big deal. It can even be like next month. Like there’s no urgency there whatsoever, but we do need to do it at some point. Ben Ornstein: Got it. Okay. And do I need access to, you know what? I know you got to go. I’ll talk about it later. Michael Houck: Yeah, I don’t think you need access. You have full Stripe access, which I think is all you need. But you know, again, when you get into it, we’ll handle whatever we need to do there. Ben Ornstein: Okay, sounds good. UNKNOWN_SPEAKER: All right, man. I’ll talk to you.