Welcome to the panel where three founders try to figure out how to build a better business and a better life in the age of AI. I'm Justin Jackson, the cofounder of transistor. Fm.
Jordan:I am And I'm Jordan Gal. Yeah. And I just jumped on top of Brian's introduction. I was the cofounder at Rosie.
Brian Casel:And I'm Brian Casel. I'm the founder at Builder Methods.
Justin:Alright. And Jordan's host and
Jordan:take it away, man. We wanna experiment with different formats and learn along the way and how to properly have a three person podcast. So this is my attempt. Okay? So what I wanna do is start off with updates on each of our businesses.
Jordan:What are you up to? What's the big thing of the week? Right? Not like this huge giant deep dive, but what are you thinking about? What are you working on this week?
Jordan:What's the what are the big things? And then I wanna ask opinions on some topics in the tech industry on Twitter, what's going on, and just get people's takes and then see how that feels at the end of the show. A little bit of balance of updates from us and what we think about things that are happening and how those relate back to ourselves.
Justin:Sounds great.
Jordan:Sounds good. Alright. So let's start off with business updates. Brian, let's kick off with you. What's the big theme for the week?
Jordan:What are you thinking about? What's on your mind? And what do you want to
Brian Casel:share? Yeah. So I know that we're going to touch on this maybe later in the show on the question of is OpenClaw necessary? That's related to what I've been focused on this week. So this is my first full week back after last week I was on vacation in Utah.
Brian Casel:This week was a little bit of a change of pace for me. Like, I didn't do any YouTube recording. You know how like when you have a SaaS and like it used to be like every couple of years you would you would do some sort of like big big refactor. Like a redesign or refactor for the goal of like, well, if we change our underlying infrastructure or our underlying foundation, then that's gonna make it easier and faster to build a lot of stuff going forward. That justification that we give ourselves, if we if we just spend a month on this refactor, then we're gonna be able to go much faster.
Jordan:Yes. Let's slow down. Gonna go faster. Yes.
Brian Casel:Yes. Was the theme for this week for me. And and in my case, that's, like, refactoring my whole operating system. The way I organize my files, where my skills are living, and how those are synced across Claude and Claude Code and Claude Cowork and OpenClaw. I've got my agents Mac mini over here.
Brian Casel:So how do I sync files between that? And my and my Macs and my iPhone, rethinking my my recurring processes. This has a lot to do with, like, content development. I I did build I spun up a new app a couple days ago to help me with my content pipeline. So that in a nutshell, that app is like it's for my agents to use and for me to use for us to develop and draft and and move content through my pipeline, through all the the platforms.
Brian Casel:So just rethinking the way that I do everything so that hopefully after this week, I can just sort of like run my processes, but constantly tinker around the edges. But this week was about like rethink the whole foundation and so that I can start to like scale with with agents a little bit easier. That's the goal.
Jordan:Very cool. I I have a question on that. What I've seen a little bit is have you seen the G Stack? You know, Gary from Y Combinator released like what he's calling like the G Stack, which is basically a bunch of skills, MD files, and like how to organize. So how do you think about that in general for someone building with Claude or or others?
Jordan:Is that, like, part of what you teach in the content as opposed to just here's how to use it? Is it all directly connected?
Brian Casel:I'm I'm actually not familiar with Gstack. I haven't seen that yet. But one thing that I really like is to see all these different platforms like Claude and OpenCloud, but also like OpenAI and Cursor. I I like to see when they converge on standardized formats. Skills is a big one.
Brian Casel:Right? The fact that I can create skills for various processes in my business and for maybe for a couple of weeks, those skills were being done by OpenClaw agents. It's literally copy and paste. I can bring them into Clawd and have Clawd run the same skills. You can even do the same thing with other agents too.
Brian Casel:So I like that portability of things. And that's part of what I've been doing this week is, like, my OpenClaw Mac mini, without getting too in the weeds here, I've got, like, two users on that, like, a a main user and then, like, the OpenClaw user. The main user is for running my Claude, like Claude Cowork and Claude Code, so that I can use my Claude Max subscription to do anything where I need to use, like, Opus for, like, creative writing tasks and stuff. And then I've got, like, more admin tasks handled by OpenCLUD. And all of this stuff is shifting really fast.
Brian Casel:I think there is a there is sort of a migration from like OpenCLUD back to Claude because of some of the updates that the Claude team is rolling out. We could talk about that later. But, yeah.
Jordan:Okay. Cool. Look at us. We just opened a loop. Mhmm.
Jordan:And and I do wanna come back to that because I I require some additional explanation on the differences between those two, and I'm assuming some people listening also. Mhmm. For now, Justin, what's going on with you? What do you got going this week?
Justin:Dude, I'm having just the best time right now. It's right now, the world is sunny. There is money to be made. We are just cranking on video podcasting over at transistor.fm. Oh.
Justin:John and I are aligned. We are meeting every week. We're hanging out. And with that alignment, we could just say to the whole team, this is where we're going. We are focusing on this one main thing.
Justin:This is our clarified purpose. This is our clarified direction. And, yeah, John and Jason are just cranking on it using a mix of their own engineering prowess and skill and a lot more Claude as well in the mix. So I'm on cloud number nine right now, just having a great time.
Jordan:Okay. So so we all know that feeling, you know, and and sometimes it's related to revenue, and sometimes it's related to progress and not necessarily the same thing sometimes. Mhmm. So this sounds like a progress and a finality or some type of, like, resolution around what to do? Is is is that where it's coming from?
Justin:Yeah. I think just getting to alignment, I think communication, I think, yeah, that passive aggressive loop that we were in was not productive. And so now it's it's also just created a culture now where I think we all feel more empowered to bring things up and confront each other right away as opposed to letting it wait. I think the big thing for me is that we were focusing on other stuff that wasn't in my my circle of expertise and passion and excitement. And now we're squarely in something that I think has big market potential, and is perfect for me.
Justin:And, I'm just so excited about it. We're just cranking on it. We're yeah. We were part of Apple's HLS announcement at South by Southwest, which was really exciting.
Jordan:What's HLS?
Justin:HLS Video. So they're bringing video podcasts to Apple Podcasts.
Jordan:Oh.
Justin:And we're in the they had three four initial partners, and now we're in the next batch of official partners. And, yeah, it's kind of an exciting time.
Brian Casel:I'm just excited for you, Justin, because number one, I I I do think that's a really exciting, smart product direction for Transistor to get into this video, like, video podcasting as a feature. I don't know if you're thinking of it as, a feature or, a mini product or or what, but that's that's exciting. But I'm just excited as a as a fan and as a listener of our podcast because this feels like the start of a new storyline for you for the rest
Justin:of
Brian Casel:the Like, where where is this going? You know?
Justin:Yeah. Yeah. No. And, I mean, if I've been talking about this on the podcast, I think almost from the beginning, just wanting to do video, but we just couldn't quite get there. So, yeah, I'm I'm delighted right now.
Justin:Of course, there's gonna be there'll be things, you know, next week, check-in. But can I just have this moment? Yes. I think you should have
Jordan:this Hell yeah. And sometimes this leads into like revenue and growth and and all that. But just being really happy in your own company about what you're doing in your own company is kind of valuable. Yeah. Full stop right there.
Jordan:So I hope it leads into revenue and growth and all this other stuff that goes with it. But it's not a failure if it doesn't. Just taking control is you know, it's your company with a very small group of people. You guys run it together and just doing what you think is right is awesome.
Justin:Yeah. So empowering. And and I think the other reason it's exciting is because this is has the potential to really increase our expansion revenue, like those $49 plus plans. I am like, just can't wait to see if we get way more sign ups, if we get way more upgrades. My sense is we will because we've had lots of interest in the past.
Justin:So I'm I'm, like, eager.
Jordan:Okay. One more double click on on that part of it. This type of feature, I'm curious how you think about it. Brian, I'm curious how you think about the same thing. When when you introduce a new feature, how do you decide who gets it for free?
Jordan:Where is it? Where is the split? Is it something that should go in the highest tier? Is it something that should incentivize coming off of the base tier? Or is it something that everyone should have at the base tier because you wanna add more value there?
Jordan:So, Justin, how are you thinking about this in particular? And then, Brian, same kind of thing. How when you come up with a new feature, a new something, a new piece of value, something, how how do you look and decide where it fits?
Justin:Well, it with us, it's easy because there's already prior art. So there the guy that's the diary of a CEO, Steven Bartlett, he launched a video first podcasting platform called Flightcast, and they already have pricing out. It's 50, 100, and $2.50. So they became the anchor. And one of the reasons I was excited about moving into this category is I'm like, I like that pricing.
Justin:You know? That's like our base our starter plan is 19, but their starter plan is 50. And so the the baseline's kinda already been set. And there's at least one price anchor out there. But there's other examples too.
Justin:Castos does some of this already, and they do it on the 50 and $100 plans. So there's already examples out there. If customers are comparing us to other folks, they're already comparing us to that. And so now, our goal is how can we come up with the killer positioning that gets people to look at us and go, oh, transistor's just the better value. Like, we're gonna go for them guaranteed.
Justin:Okay. So that's part of
Jordan:what's behind the excitement there.
Brian Casel:Yep. I could talk a bit about Clarity Flow, but before about Transistor, if I understood that last comment correctly, it was like Castos and other podcasting tools have the video feature and Transistor doesn't yet. Mhmm. So that can also make the argument like, well, it almost doesn't even matter which tier you initially put it in. It's also about like because there's probably people who are just like, I love everything about what transistor is doing.
Brian Casel:Oh, but but they don't have that video piece. Mhmm. So then it's like, even if you put it in a tier that that's higher cost than what competitors do to get video, that could still get them over the hump. It's like, well, if I want everything and I want the video, okay, I'll pay a few bucks more, but at least I get everything I want with the company I want.
Justin:You know? We have so much flexibility. We could start it if we wanted to, we could start on the $250 a month plan and see what happens.
Jordan:Mhmm. Yeah.
Justin:I think what's more likely is we'll start it on the $100 a month plan and see what happens. But we have a lot of options. And the other thing we have is we have market positioning that the other people don't. So transistor gets recommended more often in AI overviews and in LLMs right now than those other two companies I just mentioned. So we are in a we are in a unique moment that we could capture where we have the reputation of being an incredible podcast hosting platform.
Justin:So now we just get to deliver this extra thing and hopefully yeah. So we'll see.
Jordan:We'll see
Justin:what happens. It could fail, but it I'm excited.
Jordan:I you have big customer base too.
Justin:Yeah. We got, like I can't tell you exactly, but it's, like, seven to 8,000 paying accounts, but that translates into, like, 36,000 users or something like that.
Brian Casel:Oh, wow.
Jordan:Okay. So a lot to market to very, very directly that you have their ear. We we just launched our second product a few weeks ago and have a 100 users on it already because we did something funny. We advertised inside of our own product. So we sent an email.
Jordan:We sent an email. We sent an email campaign, really. We did in app messaging, and then we just put ad units inside of our admin. So when you log into the admin, on the left side, there's just an ad unit for our own product.
Justin:Oh, I love that.
Jordan:Yes. And you can exit out and close it. Fine. But we just were like, how do we visually just make as many people aware of this as possible?
Brian Casel:I like that. I do similar things in both my products actually. So like like Builder Methods, there's really only one product, Builder Methods Pro. And so if you are a free user, like, you can go on Builder Methods and actually browse our the the whole library. You can click into any video.
Brian Casel:You can you can see but you can't press play on it. So, like, the the video will have a paywall. And and also there's, like, calls to action on the left side that say like if so so if if the system knows that you're not a member or you're not logged in, it'll promote that. Real quick, like in Clarity Flow, really my only role in that business these days is to like green light new features that that get requested, and my my teammate is sort of running the show over there. So she just surfaces these, like, customer requests, and I just say, like, yay or nay.
Brian Casel:Like, yeah. Let's build it. And then and then usually as part of that and I'll give you a bit of direction, like, some high level thoughts on how I think it should work, but she'll she'll do all the scoping and building and and all that. And usually as part of that, I I also give input on like, okay, I think this feature should be in our professional plan and up and not the starter plan. And I would say that more often than not these days, like, because we're at a point now in the product where like we've already built all the big rocks.
Brian Casel:They're already they've been in there for well over a year. So now all the requests are mostly power feature requests, you know, which naturally fit in the upper tiers. Like there's we've already established what the lowest tier is, and that's basically just asynchronous messaging. If you want anything related to automations or, like, groups or, like, workflows, like, all this stuff, like, those are the kind of requests we get. Those go into the upper tier.
Brian Casel:Just real quick to close the the loop on that thing you were saying, Jordan, about, like, advertising to current customers. The way that I've always done it and the and I think it pretty much works is in Clarity Flow as as the SaaS product, every feature and every button to get to every feature is available every user.
Jordan:Okay. So you don't hide things in the admin?
Brian Casel:We don't hide things. We don't put little badges on my like, this thing is a pro feature. Like, we don't say anything like that. If if you if you're on the starter plan and you try to click the button to schedule a message to go out in the future. Okay.
Brian Casel:That's a schedule button. I'm gonna click it. Oh. It's gonna kick me to the upgrade screen.
Jordan:Okay. Okay.
Brian Casel:You know? Oh, that's that's an advanced like like, let me click the automation workflow feature because I wanna make a workflow. Click. Oh. I'm on the upgrade screen.
Brian Casel:In order to use that, you gotta upgrade.
Jordan:So Uh-huh.
Brian Casel:So that's that's always been the way. And and I see expansions happen all the time, and it and it usually happens within the first few days of the account. Like, they they start on the starter plan.
Jordan:Start looking around.
Brian Casel:And within day two, day three, they're they're upgrading because they wanna do what they wanna do. You know?
Jordan:The the last thing that we did that worked well for our existing customers, Justin, it might be applicable in your situation. Don't know the details, though, is we applied an auto credit. So we're effectively enabling a free trial of this upgrade for our existing customers. So you go in there as an existing customer and you say, right, the modal comes up that says you need to upgrade. This is a paid add on.
Jordan:And then it's like click here to auto apply coupon to get your first month free. So you're you're just opting in to be charged for it next month. You can obviously cancel it between now and then, but it just trying to encourage when someone comes into that friction like, hey. You gotta make a decision. We then immediately reverse the risk of you have to make a decision, but it doesn't cost you anything right now.
Jordan:So you can kind of delay the decision while also hitting the value.
Justin:Yeah. Yeah. Yeah. Yeah. Yeah.
Justin:That's good. Nice to have some upgrade paths there that lower the risk. Do I
Jordan:have sharing permission? Because I put it up on the screen. You should. So this is like our inbox. This is like our new account.
Jordan:So it's only one call. But see in the top right here?
Justin:Oh, yeah.
Jordan:Right. This account already has the website texting add on. And so when I exit out, I see the next advertisement, which is for the mobile app that we just launched.
Justin:That's cool.
Jordan:Yeah. So we just use let let's see what happens if I close this. Okay. The now the website text will will will come up. That's cool.
Jordan:We have, like, a like a series of advertisers. Like
Brian Casel:a rotating thing.
Jordan:I've seen that. Yeah. Yes. Cool. And then eventually, like, if I think if I close this out now to, like, how how'd you hear about us?
Justin:Oh, just oh, I love that.
Jordan:Mhmm. Yeah. We just kinda use this vertical space of I might steal that.
Justin:For for the listener, basically, he has what's like a carbon ad. If you've ever seen those carbon ads in the on the sidebar, you can think of a menu bar in a regular SaaS, but there's he's got ads in the side. And then when you click out of them, it goes to this little survey about how you heard from us, which is that's what every marketer wants right now.
Jordan:That's awesome. We use this thing on the left hand side also to, like, become, you know, an influencer, basically share it and get credit when sharing with your customers. So we're just we're just kinda used to using this left hand space to convey evergreen messages that we want.
Justin:I'm gonna steal that for sure.
Brian Casel:Those two things, the how did you hear about us question and also the affiliate promotion, I put those into the onboarding flow in Clarity Flow. So you sign up and I think, like, maybe, like, the second screen or something is like a one question. Like alright. I think I think it might be two questions. Like, one is like, what's your main goal here?
Brian Casel:What are you trying to do with Clarity Flow? And second is like, how'd you hear about And then I think I forgot now. It's been a while, but I think you you get to like a a welcome screen. And on that, it's like, by the way, like, you can you can actually make money from referring friends and and join the affiliate program.
Jordan:We we thought about that, but I think we are I don't know if it's too scared or too careful, but we we just don't wanna get in the way. I guess the trade off that we're making is let's let them get through. We will show this visually after they get through the onboarding. And then I think this is what you're talking about here. Like, we have these little diamonds that show anything that's above your plan level.
Jordan:And so this this is on the medium level and training files are for the highest tier. And we basically this is like an advertisement for the feature. Right? This is available for growth and custom, gives you the information that you can click on give it a try, and that's when you upgrade. So very similar, That like, set set of
Justin:is slick. Yeah. You guys have done a good job there.
Jordan:Yeah. It's it's fun having a self serve product because, you know, we've never really had many levers to pull. But this is finally the first one that's, like, you know, high higher volume self-service, and you can kinda, like, do more of this UI Yeah. Framing. Alright.
Jordan:Well, we're we're we're starting to get into it. I'll give my update very, very quickly. We had two big things happen this week. The first one is that we hit 2,000,000 ARR,
Justin:which is
Jordan:very exciting.
Justin:Boom. Which is reaction? I'll put up some confetti here.
Jordan:Some confetti. There you go.
Justin:There you go.
Jordan:2,000,000 audience. Eighteen months.
Brian Casel:Audience clapping audio factors.
Jordan:You know? So definitely definitely the fastest growing thing that I've I've done in the past. So it feels awesome, and we are yeah. The that's I think it's good.
Brian Casel:So this sort of surpasses the speed of growth that you had with with I guess that was Cardhook?
Jordan:With Cardhook, not at its highest. Cardhook had a few it had, like, three or four months that we were adding, like, 30 or 40 in MRR every month. And that was just, like, a ridiculous it was COVID. It was ecommerce. It was shop.
Jordan:It was just out of control. And we I mean, obviously, was great, but it wasn't a sustainable thing, and it didn't last for very long, but it was obviously this huge growth thing. So we are it's not that fast, but this is much more consistent.
Brian Casel:Yep. Very nice.
Justin:That's amazing, dude. Congrats.
Jordan:It's awesome. Thank you very much.
Brian Casel:Oh,
Jordan:yeah. The the the second thing big thing that happened in the company was we migrated our entire infrastructure over to a new platform. And the last time we did this, it took us like four months and it was difficult. And this time, that was a year ago and we had an agreement for an annual contract and it was coming expired and we had a decision to make, and we decided to explore other options to understand what it looked like. And the difference in all of the AI tooling between a year ago and now.
Jordan:A year ago, it was three, four months of very stressful, very careful, very deliberate, very, you know, it was very stressful. And this time, it's less than three weeks. Wow. And just reached feature parity and then started pushing new sign ups to this new infrastructure yesterday for the first time.
Brian Casel:Can you can you go into that a bit? What what interest what what parts are we talking about?
Jordan:So I don't wanna name names because I don't want to put anyone in a in a bad light in terms of, like, leave it leaving one platform or another. The big pieces are the three inference pieces in voice AI are the LLM, text to text to speech, and speech to text. So those three parts, we were on a platform that enabled all three of them together. They took open source models, hosted them on their own hardware, and you were able to access this one black box that had all of it. Some great advantages there.
Jordan:We concluded that we were better off in our situation to pick and choose the best of the closed source models. So now this new infrastructure platform, I'm sure we'll announce it at some point, allows us to choose the specific OpenAI model, this specific transcriber on Deepgram, this specific voice on Cartesia. It's just and then if anything gets better, we can swap to it quickly.
Brian Casel:Just swap it out.
Jordan:Nice. We were hoping a year ago that all the open source models were going to get there sooner, that they would be almost as good as the closed source. And then this one black box that had all three in the stack together would be better. And that's just not the case. The closed source models are just flying.
Jordan:And I don't know if open source catches up anytime soon. And our model of this very horizontal agent that needs to just be smart enough regardless of who signs up and we scrape their info, the model just needs to go, cool. I get it. This is a dog groomer. Now I have all the context.
Jordan:It just needs to be the best. And that's what so that's what we were after.
Brian Casel:Yeah. I mean, I don't wanna get on too much on a tangent here, but, like, I I just I don't I know people keep, like, experimenting with all these, like, open source models and, like, self, like, locally running models and all these different things. And I just I just a, just haven't really had the time to try them, but b, like, the the popular, most well known models for me. I'm I'm pretty heavy on Claude models, but I also use OpenAI's, you know, the latest IGBT stuff a little bit. I just stick to those, and they're just so damn good.
Brian Casel:And they keep getting so they keep getting so much better that it's like, I don't jump around.
Jordan:Yeah. And I I don't know.
Brian Casel:I just get more and more comfortable with the best of those. You know?
Jordan:Here's a quick question for you. On on the coding model specifically, I saw some news today that at first was positioned as like a gotcha, and it was that cursor under the hood was using Kimi. And then and then someone on the cursor team was like, here's how it works. Like, we take this open source model, then we add a bunch of training on top of it. And, it is part of our offering, and we do pay them properly, not like people are saying that we don't pay them when we stole it.
Jordan:And so it what is that fusion between the open source models and the, you know, companies and teams and models that work alongside them on top of them both?
Brian Casel:I did see that said about the Cursor Composer model. I haven't dug into the actual story. I don't know what's true or whatever. That's one model that I have used. I was using it a lot more a few months ago and it's super impressive.
Brian Casel:And I know that they just dropped version two, but I was using it back from version one. And even that was The most impressive thing about that was it was lightning fast to turn around every request. Compared to every other model, it's like a 100 times faster. Like do this thing, boom, it's done. Do this other thing, boom, done.
Jordan:Okay.
Brian Casel:It's like, woah. But the interesting thing about what Cursor is doing there is this is a product, and yeah, maybe they leveraged a Kimi or whatever other models under the hood, but they definitely tuned it. And they have all these developer users that they can train it on that data. And so clearly, so that's a really interesting example of a product company. They are an IDE, like coding development product, making their own in house model that is purpose built only for that.
Brian Casel:They don't need to be a writing model. They don't need to be an image generation model. They just need to be great at coding because they're a coding product. So that it'll be interesting to see if, I don't know, like other tools like a like a Figma or one of these that just creates a great image model because they're Figma. Like, something that's, like, built into their product that that starts to decouple the need to like, how how can a product make their own in house LLM actually competitive with a Claude Opus or a GPT-four?
Brian Casel:Well, it's just like a Bootstrap SaaS, like niching down. You're just better for that industry, you know?
Jordan:Yeah. And I think we do see that with video, with other things, and and that'll continue. Okay. We're gonna move forward from the update part. I thought that was great.
Jordan:And and, Justin, I'm gonna go to you first on this. Mhmm. There is a product and a bit of a story developing around this company called Pulsia. Yeah.
Brian Casel:And Jordan, you and I talked about this like three or four weeks ago, right? Like when they were just starting to come out.
Jordan:That's right. And you guys know I'm on the ZHC kick, the zero human company angle. This company, Poldja, is a wild product. If you go to the site Yeah, there you go. You're sharing.
Jordan:Super creative. I signed up because I wanted to understand what was going on. The design is nuts. The execution is crazy. The the social, like, element.
Jordan:Okay. So what this thing does?
Brian Casel:I wanna know what you built. I wanna know, like, what if what kind of company you spun up.
Jordan:I I it was such a such a jarring experience. I signed up, and next thing I knew, it had sent out tweets about the new company that I was starting. And I was like, hold on. Hold on. Hold on.
Jordan:I I just signed up to look at your product. Don't go telling the whole world that I just launched this new thing. I I I wasn't ready for it. And I emailed the founder and was like, so creative, so interesting. Delete my account immediately.
Jordan:I think I might be too boomer for this. Best So, of okay. So here's the thing. The product is interesting. What it does is it allows you to launch a company autonomously.
Jordan:This thing goes out and builds a company for you. But next to it, it's almost like the real story is that this is a one person company that went from, I think, a million ARR to 5,000,000 ARR in, like, eighteen days or something like that. So what the hell is going on? What what is this thing? Is this a giant scam?
Jordan:Is it fake? Is it real but doesn't add any value? Is this the the new way to look. What what is this thing? I I need your opinion.
Jordan:Justin, when you see this thing, what what do you
Justin:well, Laura Roeder, I saw that she she was the first one questioning his revenue numbers. And that's the big thing is that there's a criticism of he's selling the story of 1,000,000 to $5,000,000 ARR in eighteen days. He's now using this word run rate. But if you look up in Grok, what are the criticisms of Ben Serra's revenue reports? The big one is his own public dashboard reportedly shows 3,218 paying users at $49 a month, which only equals 1,900,000.0 ARR.
Justin:The jump to 5,000,000 or prior claims like 1,500,000.0 in weeks is called out as including a 20% rev share on businesses that may or not be sustainable or real. So he's he's inflating. He's like, look at this. This is crazy. It's like, well, you're massively inflating what's actually happening here.
Justin:And the which all so, I mean, we none of us like this. Right? Like, none of us are none of us like people inflating their revenue numbers.
Jordan:Yes. Something doesn't feel right.
Justin:Yeah. So
Brian Casel:But and but also with that inflating thing, it's like, why inflate it? Because even just, like, 1,000,000 in in sixty days is impressive. Like, it's like, you know, why why inflate it to 5,000,000? You know? Yes.
Jordan:It it's almost like that's the PR. The marketing angle is eye popping, attention getting, one person absurd revenue growth. I don't know what to tell you. I I think everyone's keeping an eye on it, and everyone I I think the general consensus is that there's no way this is what he wants us to think it is, which is, you know, a software product that builds businesses that people are signing up for and paying for. And if those autonomous businesses that are created on behalf of users are successful, people will continue to pay for sure.
Jordan:And this thing will grow enormously. But it's almost like a believe it when I see it coming.
Brian Casel:Yeah. I'm gonna I don't know that I I personally would go so hard on this guy about, like, inflating or or or misrepresenting his company or or whatever. Like, I heard his interview on Mixergy with Andrew, which he did maybe about six four, six weeks ago, which was a really good interview. And because like you, Jordan, I'm like, what is this? This is wild.
Brian Casel:I want to hear the story. And Andrew is such a good interviewer that I knew he would get it out of him. He did go into it. Like what was interesting about that interview was they spent about thirty minutes talking about what it is Pulsia is and what he's building and the rapid growth. But finally, around thirty minutes in, Andrew was like, All right, but what about the businesses that people are building?
Brian Casel:Like how much revenue are they making? Yep. And then it sort of dropped. And he was just like, well, the most MRR that one of these businesses is making so far is $50 MRR. It's like, oh, wait, what?
Brian Casel:So there's nothing here. At the same time, I just wanna say, I haven't really heard, maybe inflated that 1 to 5,000,000 number a little bit. But like, I think he's also being pretty upfront. Even in that tweet that you were just showing, like the whole second half of that tweet, was like, look, most businesses are built on this thing are not real yet. It it's still early days.
Brian Casel:And that's true. So I I I would say that, like, I'm not actually surprised that a a new startup like this is, you know, growing its own MRR very quickly. There's just such a wild demand for, like, all things AI right now, like, whether it's OpenClaw or whether it's something like this. Like, there's there's clearly just worldwide demand. Of course, it's gonna get crushing sign ups.
Brian Casel:I yeah. I do question, like, I don't know if this one is gonna be the one that, like, sticks long term and real businesses come out of this or, like, real agent owned business, like funds come out of this or portfolios come out. Like, in concept is really interesting, but I don't know if Pulsia is gonna be the thing. Like, it it but I wouldn't be surprised if a year from now, there are completely autonomous agent run mini businesses, if you will, and people or organizations owning a little portfolio, maybe a side portfolio of just agent moneymakers. And I could see that being a thing, and maybe it becomes a more legitimate thing in the next year or two.
Brian Casel:And this gets back to the same thing that I think about. Like, whatever seems like rough and sketchy and doesn't quite work today, but there might be a kernel of interesting Mhmm. Stuff here today. Twelve months from now, more established companies and products are gonna form around this that make it real. Plus the LLMs are gonna be so much farther advanced, and 2027 could be wild.
Brian Casel:That's my take.
Justin:Can can I show you guys my inbox?
Jordan:Yeah, please. Are you getting email emails from palsia companies, please?
Justin:Bro I
Brian Casel:have I've received some
Justin:of Yes. Am. Yeah.
Jordan:Okay. Okay. Okay. So first of all, Brian, I think that was a a great take because what we're seeing is demand. Yep.
Jordan:Very straightforward. People want the people want it. Whether or not it can deliver does not necessarily mean if it's, like, being dishonest or doing anything wrong at all. The demand is there. He's capturing the lightning in a bottle element of the marketing, the attention, and the PR.
Jordan:Can't blame him on that. I hope it works out. It's really easy to hate on it because everyone's ego looks at that and is like, what? Can't be possible? Yeah.
Jordan:But, you know, a year from today, maybe someone comes out with a business building model that actually does a very, very good job of it. And online fully online businesses are, you know, heading in this direction.
Brian Casel:And all elements of that are gonna be better twelve months from now. The LLMs and all all the rest of it.
Jordan:And Yep. I
Brian Casel:do think that there's a parallel with with the OpenClaw thing. I think it's like these scrappy one person startups come out of nowhere, and they introduce a breakthrough idea that that the world goes crazy about. But that startup, you know, fizzles out. But then
Jordan:That's there.
Brian Casel:Then the market takes over, and and it becomes real like a year late. And I I think that's probably what's happening with OpenClaw into into Claw. We we can get into that later. Yes. Yes.
Brian Casel:That's the parallel.
Jordan:Justin, what are you what are you
Justin:showing here? This is my inbox for transistor. I don't know how I ended up on this. And this is just me searching Pulsia. So this is a this is a very, like, common.
Justin:This is what it looks like. Calm like, Calm Company, Calm Inbox. This is from Nightcrew dash AI at Pulsi app. You built Transistor around Calm Company principles. So it knows something about it.
Justin:But 30,000 podcasts on the platform means customer questions, partnership pitches, and listener feedback don't respect that. Night Crew is an AI email triage. Reads everything, surfaces what actually needs you, drafts the rest, calm restored. Free thirty minute trial to see it work on your real inbox. Worth a look.
Justin:And then here's the deal breaker for me. I mean, already, this is a deal breaker because I'm like, I don't I don't care.
Jordan:Not a bad cold email.
Justin:Not I mean Here's what makes This it worse. Company runs autonomously on Polsia. Any email I get that is telling me this company runs autonomously, I am giving it the big old thumbs down. I don't want a company that's running autonomously. I want a company manned by real humans.
Justin:I want human connection. What what in my world, I'm not seeing anybody that's like, man, I really want to talk to more AI bots. No. I want to talk to real humans. I want real humans that actually I'm if you guys get those pod the the emails for this podcast, it's like so many of them.
Justin:And they just feel like they're predatory to me. It's like you're grabbing stuff from our transcripts. This isn't like a human that took time to know about me and my company. You're grabbing stuff from our transcripts. Like, oh, I love that last episode.
Justin:No, you didn't. That's a fucking lie.
Jordan:Yeah. Me too. Love that last
Justin:episode when you talked about this and this and this. And then and then you get five paragraphs in, and it's like, I'd love for you to have been, you know, with such and such on your program. I'm like, fuck off forever. You're done.
Brian Casel:Insta delete. And By the way, I I just searched Pulsia in my Gmail inbox. I have 14 cold emails from Pulsia in the in the last, like, six weeks. Yeah. I don't even have a Pulsia account.
Brian Casel:Like
Jordan:I think this is a bad trade off that he's making. The virality for you know, it it kills the upside of the actual business. Anyway, so so a very interesting experiment that we can keep an eye on. Let's see let's see how it goes. Yeah.
Jordan:I saw this post from Rob today, and he's also complaining about the email spam.
Brian Casel:But, I mean, also and I totally agree with that one too. But the cold email is not new. It's been annoying for decades, and we've we've all had mountains of it. So, yeah, there's there's there's more of it from this Pulsia, but, like, people and companies have been using automated bots to send cold email for years already. You know?
Brian Casel:That's nothing new.
Jordan:Okay. Well, let's let's move on to this next topic. I'm gonna give our give us a break. Instead of diving directly into another potential scam story, let's look at a positive story here. Adam Waffen just I don't think he launched, but I think he announced a new product.
Jordan:Interesting looking product. And I think what's most interesting first of all, Adam, congrats on joining the, yes, I have no hair and I'm proud of it club. And also very happy to hear that your little fluvial baby is doing okay because I think you
Brian Casel:have a little bit of
Jordan:a scare. I think what's interesting here is we have in Adam and the team there, it's like very high end Internet people that offer products to other Internet people to make the Internet better. At least, like, the top of the Internet pyramid here. Yep. And then just confronted with this very, very new reality around Tailwind being adopted and then Tailwind being used and being open source and effectively no value coming back to the main company.
Jordan:So the business model was no longer relevant of like, we're gonna get attention because people are gonna adopt it, and then we're gonna offer these premium elements, these premium products. So this is that team moving in a direction to provide a new form of value to the market now that everyone codes in a different way.
Justin:Yeah. Do either of
Jordan:you Brian, do you have a take on this? This is starting to move into your territory, but on the front end. So what where's this going?
Brian Casel:Such a big fan of Atom and Tailwind and what they're doing with this. I haven't tried UI dot SH yet, but I can't wait to get my hands on it and try it because I'm sure it's gonna be if if not just like if not just interesting to see how they design a totally new product, I I wouldn't be surprised if it becomes like an everyday tool that I use because Tailwind certainly has. So one thing I wanna point out also from a product marketing how do I is it, like, on the website? Where did I see it? I love this.
Brian Casel:Alright. So their h one this is just like a coming soon website right now.
Jordan:Right? Yeah. It's ui.sh.
Brian Casel:They've got a main headline, like, turn your terminal into a design engineer. Okay. Interesting. And then a sub headline, like, toolkit for coding agents, Claude code, AMP cursor. Okay.
Brian Casel:Interesting. Build UIs that don't suck. Okay. You're getting on the pain point. Yeah.
Jordan:Continue to use the products you prefer. Right?
Brian Casel:The most powerful text on this page is actually the smallest text in the bottom right by the by the people who made Tailwind CSS. Everybody on the Internet or everybody who makes stuff on the Internet knows Tailwind CSS. Oh, the people who made that thing are making this product? Oh, this is like instant credibility. Gives them an opening.
Brian Casel:Yeah. Yeah. Like, that that's just it's like no matter what they what they do, it's going to be interesting and noteworthy because it's from the Tailwind CSS team. Yeah. And I also love the idea that I I don't know if he's talked about this.
Brian Casel:I was I was gonna ask him about it. If the what their pricing plan is, I hope that they're gonna be charging for this in some way and not just giving it away for free. I I assume this is like a new paid product. I I just like the idea in general of like, okay, we've tried everything on the Tailwind CSS. Yeah, we have some like some some supporter donation business going on there.
Brian Casel:They've got the pro plan there. But now it's time to like clean slate. What if we started something totally from scratch? I love the choice of product, like category. This is makes it just makes so much sense for them to make a UI design tool.
Brian Casel:I I just love the idea of, like, we sell to designers and developers. What's the next thing that that designers and developers need? It's a UI tool. And then from an AI standpoint, this this to me remains the one of the big unsolved gaps in the builder's workflow. It's something that I think a lot about.
Brian Casel:I I hacked on an idea called Design OS a couple of months ago. Some people used that, but that's not quite the the right solution, I don't think, yet. And and I'm still rethinking, like, where where does the UI design phase of of the workflow take place in this new agentic era. So there's this, and then Google just announced this new tool called Stitch. I haven't tried that yet, it's really I think from a product standpoint, I'll shut up after this point.
Brian Casel:From a product standpoint in the UI design space, think there have been many interesting products over the past year or two, like new products. The one thing that I don't think any of them have really quite figured out yet is the handoff or the integration from, okay, I have my vision in some sort of UI or a set of UI screens, like this is what the dashboard is gonna look like, this is what the account page is gonna look like, this is what the onboarding flow is gonna look like, but it's not functional yet. We have that UI figured out with using these new tools. Now, how does my Claude code agent take that and run with it and build build it out into the rest of the app? I'm still thinking through that sequence of events in the in the in in today's new workflow.
Brian Casel:It'll be interesting how how this kinda fits in.
Jordan:Very cool. Justin, do you guys use Tailwind? Is this something that your company would adopt? We we don't. So we won't, but we're kind of, like, cheering them on.
Jordan:I I think this I hope that this story in this arc is like a creative destruction. Like, in order to get here, they had to go through that and that this thing goes a 100 x whatever they were doing before.
Justin:I'm actually bullish on it. And here's my theory, and we'll see if you guys agree with me. I think we're entering a new era where to stand out, your product is going to have to be like a religion in the same way that Basecamp the the Basecamp way is the reason a lot of people sign up for Basecamp. You sign up for Basecamp, you're signing up for the Basecamp religion. And a lot of the attraction is I'm going to Basecamp because they seem to have some sort of magical, mystical process that makes them productive in this way.
Justin:So I'm not just signing up for software. And I'm not just signing up for an opinionated UI and UX, although those are both true. I'm also signing up for a framework and a philosophy and a way of working. And now that we have all of these people launching generic products, the way to stand out is to have product plus what I'm saying religion, plus an overarching philosophy, a way of working that is aspirational. So when people sign up for your tool, they're not just signing up for this.
Justin:They're also signing up for your opinions and your way of working. Like, man, I I can't tell you how many times I've been like, man, if we just fully adopted the Basecamp religion and fully adopted Shape Up and fully used Basecamp, maybe then all of our problems would go away, you know? And that thought is so powerful. And in an undifferentiated product landscape, when everybody's launching these generic products, what's going to stand out? It's going to be this kind of personality.
Justin:But also, like, we also have a philosophy, a way of working that when you sign up for this, you get this. Adam and Steve could have this in spades. It's like, do you want to design the way we do? Well, we've built the tool, but we've built it so opinionated that you're also signing up for our philosophy of work.
Brian Casel:I I love that. You know, one thing that I would add, and Jason Fried tweeted this, I think a few days ago, I think the thing applies here is that I I don't know really how how the UI. Sh is being thought through and designed from a user experience standpoint. But I do think that most SaaS, not all, but like a lot of SaaS that involve a lot of usage. Like people logging in to use the app, to click a lot of buttons and do things day to day.
Brian Casel:I think the way forward now is to make your products accessible and usable by agents. And I actually, I would say, I would I would tweak that. I would say, usable by humans, just like they always have, and welcome in potential users as agents. Which is much better than building in, like, an under the hood Jordan, your product is a little bit different. But, like a bit like, what what Jason Fried tweeted the other day was, like, I think they have I think it's basecamp.com/agents.
Justin:Let me see.
Jordan:Yeah. I've gone more toward your direction on the agents. I just think we're we'll be a little later. That's all.
Brian Casel:So this page is great. I just saw this yesterday. Basecamp.com/agents. They're basically just advertising the fact that, hey, Basecamp has a CLI now. Basecamp has an agent skill now.
Brian Casel:It's easy enough for any agent to log in to your Basecamp account and do anything that you and your team would do, but an agent can do it. So that, I think, is the way forward for any tool that is very fiddly, very user like, oh, I have to click a lot of
Jordan:stuff to
Brian Casel:spend a I have to spend a lot of time to do this stuff. I think going forward, I know this seems a little bit beyond the edge of reality today, but I think this is where things are going. More and more people are going to wanna just quote unquote hire or delegate to their agent and say, can you just manage my Kanban board for me? Can you just set up a project, set up an automation in that in that automation tool that we pay for? I don't wanna like click all the buttons.
Brian Casel:Just go do it. The agent should be able to do that using a CLI. So how does this relate to UI. Sh? I don't know how their product is gonna be designed and used.
Brian Casel:But I know that design tools historically have been extremely fiddly. To get the value out of it, I have to sit here for hours and click buttons and drag rectangles and scroll through font choices and do all the things and pick colors. Like, as a designer, that can be fun, but it could also be extremely time consuming and frustrating that I'm spending so much time on this stuff. I just wanna get to the So it would be great if I could use something like ui.sh with my Claude agents who know my preferences, who have their memory of me, who have and I can use my very best models that I pay for Claude to go use my ui.sh account to to now prompt it and and get a and get a solid UI. And then maybe I'll I'll give some input on that, you know?
Jordan:Yeah. I'm I'm excited for them. My big worry is around pricing because we have all become accustomed to getting it in a preposterous amount of value from what we're paying platforms like Clod because they're super subsidized. And so if a $200 max subscription gets you a absolute mount Everest amount of value, it's tough to it's tough to go anywhere near that price point around something that is, like, adjacent and added to and, you know, will not be seen at the same level, I assume. Don't know yet.
Jordan:But I think you're right on the interaction. When Linear came out with their Slack integration. And Claude, my engineering team were ecstatic. They were like, I can't I can't believe I just go at Linear, add this to the ticket, assign it to me, priority level medium. Done.
Jordan:Yeah. No more clicking and fiddling just like you like like you mentioned. Right? It's like this this weird version of, like, manual labor, and it's just unnecessary because the machine does it for you.
Brian Casel:Yeah. And I also I also think that there's a huge difference in quality level and power that comes from because we've seen over the past year all these SaaS tools just add in a built in magic AI button. But you know that under the hood, those SaaS companies have to manage their own costs for that. Right? So they're gonna use a less than frontier But even if they use a decent enough model, it's not trained on you.
Brian Casel:It's just trained on their own system prompts. It is much more powerful for the user to be able to bring their own agent. Or whether that means like letting their agent use it via CLI or plug in their own API keys that so that their own, you know, mean, and there's whatever, there's there's like a subscription issues with that. Like Cloud wants you to use their own tools, but that that I think is the the really the future is like people and companies using their own agents that have their history and memory and then deploy those agents on. So for tool makers, for SaaS makers, just make a great tool and value.
Brian Casel:You don't have to be AI delivery. Jordan, yours is in a different category though, because voice for businesses, I would say that's like an exception to what I'm saying.
Jordan:No, no. I think you're on the right track. We're just a little later because our users adopt a bit later and that means their tooling adopts a bit later. It just gets pushed out into the future because you still come into our product and click around. And you say Yeah.
Jordan:Perfect example. Very small little feature that's very important, FAQs. So to be able to add an FAQ so that Rosie knows this FAQ that comes into your business frequently and that you wanted to get the answer just right, Why not be able to do that from your phone? Just send a text message, add us to my FAQs or Slack or or from your ticketing. Right.
Jordan:You just something you did something over chat. You put into the CRM and you say, add us an FAQ And add add it to Rosie's context around whenever this topic comes up, this is the answer that we want them to provide with this nuance. Justin, bring us home on this, and then we'll move on to the next topic.
Justin:Yeah. I just have one more thought that came up while you were talking, which was that in one sense, I agree with you. You got to bring your own model. The other thing I'm thinking about is if Adam and Steve developed an incredibly opinionated model that was accessible via Slack chat, And I am hiring Adam and Steve's model or their harness or whatever shape it takes. I think this combination of at mentioning something in Slack, plus having some of the magic, the opinionated magic of, in this in this case, Adam and Steve.
Justin:It's like, Adam and Steve bot, I want you to go out and build out this next page for us in Slack and then report back. And then it goes out, And it's got all of their secret sauce infused into this machine, this proprietary model. Maybe it's proprietary plus your plot or whatever. But they've got some magic in there. Dudes, who wouldn't pay for that?
Jordan:That's what we've always all wanted.
Justin:Who wouldn't pay for that? Hire them.
Brian Casel:Yes. From the bits and pieces I've seen them share, little video teasers, I think it's really cool what they're building because like, they're building a UI builder, a UI design tool, literally, a UI design tool. But what makes it interesting is that I think we're getting ahead of it because they haven't fully released it. I think the idea is like you can design a full interface piece by piece and say like, I like three quarters of this, but the sidebar needs tweaking. So let's swap that out or let's swap this menu out, but keep everything else in place.
Brian Casel:And you actually need like a physical UI for that. And that's the kind of thing that like just designing in Claude or Claude code, it'll end up just like redesigning the whole thing and then you're going in circles. So like that, I think that's a perfect example of like actually build a product interface that's usable. That's, like, the core piece. And then and then we bring AI agents to interact with that new interface that they put together.
Brian Casel:Mhmm.
Jordan:Awesome. Yep. Adam and Steve, we are we are rooting for you guys. Very cool.
Justin:Absolutely.
Jordan:Yeah. I'm glad we talked about this. The the the idea of just being able to hire them to build UI is like Yeah. Yes. Like, in many ways, that's what you were doing when you were buying their elements, their talent elements, and this just moves that value over into the way people work now.
Jordan:Very cool. Alright. I wanna ask very quickly. I don't wanna spend too much time on this because this is another potential scam. YC backed 30 under 30 raised $32,000,000, $300,000,000 valuation, growing like wildfire company, Delve.
Justin:Yes.
Jordan:What Delve does is compliance. We have talked to them. We I mean, we are in a Slack channel with them because we are considering doing Hippo with them and SOC two and all this other stuff. This Substack article comes out from Deep Delver. Gotta love this stuff.
Jordan:Effectively exposing or what appears to be an expose that they're just making half this stuff up. And the the the way they're so amazing that you can get your SOC two compliance in, like, two weeks and your HIPAA compliance in, five days might just be a giant scam. So what is going on here? Look. There's this interesting story about what might be a potential I mean, this is this is kinda serious because compliance and security, like, you might end up in
Justin:jail for this sort of thing.
Jordan:So it's not like a small thing like, oh, they fudge their ARR numbers. But what I really wanna talk about is how unbelievably big of a scam compliance is to begin with. Yeah. What is going on with this company? Justin, do you have any of this stuff?
Jordan:SOC two, HIPAA No. ISO two zero zero one EU proctology exam regulation?
Justin:This is the great thing about running your own company is that we got to decide early on. And I've got some hilarious stories about talking to the big enterprise companies, like a major like, one of the big top I'm guessing, one of the oldest IT companies in the world. I'll say that.
Brian Casel:Okay.
Justin:They called us. They already had podcasts on transistor. They wanted to do, like, 50 accounts, but also make it so that all of their global offices had to use us. So this would have been eventually hundreds of accounts. And he was just like at the end of the call at the time, it was just John and I.
Justin:And at the end of the call, he's like, okay. After this call, we're gonna get our legal team contacted your legal team. We're gonna get the SOC two people. We're gonna and I was like, woah, woah, woah. Hold up.
Justin:Hold up. I'm like, this company that you're talking to is just two people. It's just John and I.
Jordan:In your field of vision right now. Yeah. I
Justin:said, you're literally talking to 50% of us. I said, if you need any of that, we're not the right provider for you.
Jordan:Okay. Good for you. So you didn't see the dollar signs and look into it. You were just like a no.
Justin:It's just so much pain. And it's like, sure. Okay. You know, in retrospect, at the time, it would have been nice to have the money. But let's say that ended up being, I don't know, a $200,000 a year deal.
Justin:That's like a lot of money.
Jordan:Yep.
Justin:But it was way better for us to just focus double down and focus on that hobbyist prosumer customer and just get thousands of them. They sign up with a credit card. They cancel with a credit card. They don't need all this extra shit. That extra shit drives me crazy.
Brian Casel:100%. Think about the we talk about opportunity cost. Like, think about the app if if at that time when you were a two person company and you're still in startup mode, and and like, yeah, you could drop everything to to get through the gauntlet to land a $200,000 a year deal. Yeah. Maybe, right?
Brian Casel:Then even if that comes through, then you're servicing that contract. Then you become an agency for this one. I know it's not that extreme, but like
Justin:It could have been. We would have had this higher more people.
Brian Casel:The other features and all the other support that you would have had to push off to to make space for
Jordan:all that. Some industries don't allow for the flexibility. Rally, which was a checkout that ended up working with larger merchants,
Justin:they all needed it.
Jordan:We didn't have a choice. So, we worked with Vanta and we paid all this crazy money and went through these absurd processes. And my real issue with it is that it is window dressing that just makes people and companies and departments and lawyers feel better. But if you're a dishonest company, are you not gonna be dishonest because you're SOC two compliant? Like, no.
Jordan:And the reverse is the 98%. The the the 98% are just honest companies that are not doing anything wrong, but don't have these accreditations and these certifications because they can't afford to go through the process whether financially or organizationally. Mhmm. And a lot it's it's almost like what Dell pretended to do should be possible. Mhmm.
Jordan:Let's just take a quick look into your company, make sure you're doing things correctly and give you a, you know, a sticker with, yes, you're compliant. Because that's Yeah. That's how it should be run. Yeah. And instead, it's this whole gigantic rigmarole, and I know I'm minimizing underestimating the importance and all this other crap.
Jordan:But just a a hateful corner of the Internet. Just awful.
Justin:Yeah. The the sad part is that I think there's actually a lot of opportunity in this space. So what Stripe did for billing and credit card processing, which was this big, hairy, complicated, like just Manual. Yep. Manual labor and the worst providers on the planet, like dealing with credit card companies before was just the absolute worst.
Justin:And they took all of that complexity. And they said, we're just gonna hide this. And we're going to make it simple for you. And they achieved it through a lot of hard work. And if you still look at the back end of Stripe, there's still I talked to somebody there and it was something crazy.
Justin:Faxes that they have to manually, like, scan and figure out and like
Jordan:So it didn't disappear. They just pushed it over to their side of the column. They said, we'll handle that.
Justin:Yep. That's right.
Brian Casel:It's about lie like, it's you're selling, like, liability protection. Right? Like, that's all really what these things are doing. But really, I think in order I don't know the details on this, but the idea of selling that as if that's the product that you're selling, we're selling you the peace of mind that your company is gonna be in the clear for any liability when it comes to HIPAA or whatever these other compliance things that your customers may end up needing or may end up in a lawsuit that involves your tool, right? So that those liabilities are codified in law.
Brian Casel:So so, like, whether you're this Delve or if you're like Gusto, like, a big part of what Gusto sells is the ability to, like, file your taxes on time and make sure you're gonna be in the clear on that.
Jordan:Unemployment stuff. Yep.
Brian Casel:You know, unemployment crap and all that.
Jordan:So
Brian Casel:so, you know, like, it at the end of the day, it's like, how do you stream? And what what Stripe did, there was a little bit of technical stuff, but it was also like they're just handling paperwork. I've done a Stripe Atlas company once, it was like, they're just kinda streamlining the paperwork. And you can sorta tell behind Stripe's clean UI, there's an office somewhere that's just pointing to like Delaware Corp websites, you know? Yep.
Brian Casel:One question I have on this is like, what does this say about YC? YC's name is
Jordan:on this. Yeah. 30 under 30, we already know. You know? That's that's like an anti signal.
Justin:Yeah.
Jordan:I don't know if it says anything about YC. Is it their responsibility? Some of these things are like a, you know, blatant frauds, they should have caught them and blah blah blah. This looks like innovation. It's similar to what Stripe did.
Jordan:Right? They said they're using AI to make things faster and better and all these other things. And I don't know if
Justin:do you
Brian Casel:think Is it innovation? I mean, because it's like again, I don't know all the details, but if it's like, it it is innovation to me if you have actually figured out a way to streamline the annoying paperwork and the actual legal compliance. If you can literally get get your customers to actually be in the clear of those liabilities, then that's not a scam. But if but if you're saying that, like, sure, you're you're you're not gonna be legally liable if your customers sue you because you're you claimed to be HIPAA compliant and you're not really, like, that is scam territory.
Jordan:So what responsibility falls on YC? Right? They are an investor in a private company that relies on claims being made by the representatives of the company. Right? And and whatever we're saying is cool, but what's in the legal docs is what's what's real.
Jordan:And in those legal docs, the people at the company are attesting to the fact that what they've represented is true.
Justin:Yeah. So
Jordan:it's like, isn't YC, like, the initial victim in in something like this?
Brian Casel:That's sort of what I mean. It's like you you would think that something like this would
Jordan:Ring alarm bells, like, go off the radar or something?
Brian Casel:Yeah. Like whatever deal they had. I guess my first question is like, what does YC do after the discovery of this? If it's in fact confirmed and all that, does that invalidate their association? But then also it's like the maybe the question is like, are there just so many YC companies that that it's like they don't, like, they they don't really have that, like, hands on or close watch on who who they're working with these days?
Brian Casel:Like, I don't know.
Jordan:Yeah. Like, is the pool of YC companies any higher, you know, more represented with some form of nefarious business than any other pool of businesses?
Brian Casel:Don't Yeah. Like like, we we remember we we probably remember the early days of YC where it's like, oh, these 20 companies in a year are the YC companies and they're highly respected. And now there's like hundreds of them every year or thousands. I don't know. Like
Justin:I think YC does have a responsibility. I mean, you've got Gary Tan back in 09/13/2025 saying, Delve is now a top YC startup. So him just saying that he's promoting them. And there is this magic stamp of approval from YC. It carries weight.
Justin:That's why startups try to get into YC. That's why they advertise it on their homepage. So if it is a badge of approval, and YC, know, prides itself on being the smartest, They're all thought leaders. They're all the smartest people in the room. Okay.
Justin:If you're the smartest people in the room, do your fucking homework.
Jordan:Yeah. Before people invest money and buy their certificates and depend on we we were gonna go with Dell for our HIPAA compliance because it makes sense for us to work with medical field. And part of why we were gonna go with Delve was because of this reputational glow around it.
Brian Casel:Yeah. Also, there must be some, like if it's too good to be true, at least just ask a few extra questions. Right? Like whether you're YC or any of the because we've all faced the prospect. We get the requests from people who need HIPAA compliance or need whatever compliance.
Brian Casel:And then we look into like, well, what does it cost and how much time to actually get that compliance? Oh, tens of thousands of dollars, hundreds of thousands of dollars, months or years to get that? Okay. No. Or maybe.
Brian Casel:Right? And then this company comes along and says, we can get you approved in two weeks. It's like, oh, that's interesting. But how? How does that work?
Brian Casel:Explain it. You know?
Jordan:Well, with AI, you know?
Justin:Except the claims are it it seems like these claims are pretty easy to debug. This is why I I think YC has responsibility here because, you know, you've got this fellow here, and he's saying it's not possible for any US auditor to issue SOC two certifications with less than three months of monitoring window. Six is typical. Dell issued a certificate in two weeks. So if that's if that's true
Jordan:Mhmm.
Justin:Then, yeah, that's a huge alarm bell. This is also why, you know, for a while, there's a bunch of these other merchant of record companies popping up around Stripe. And there was one in the MicroConf community for a while. And I was just skeptical as hell. I'm like, I do not want a little startup managing merchant of record for me.
Justin:And then sure enough, eight weeks later, they're they're closed down.
Jordan:Yeah, turns out.
Justin:You and this is actually important for anyone listening, I think, is there is opportunity here. There is all sorts of shitty enterprise compliance, all the gusto and lot of Carta, lots of places, companies playing in this field. You can make money here. I think it's worth doing, actually. Yeah.
Justin:But you got to actually do the work. Like, you got to treat it seriously. You got to be an adult. And the what's concerning about this is YC is supposed to be the company that takes these young, you know, 20 founders and turns them into the adults in the room. And it's like, well, what did you do there?
Justin:You totally fucked up. That's not you didn't even that's your responsibility. Right?
Brian Casel:Yeah. That's why I have so much respect for Gusto. I I was a customer of theirs for a long time, and, like, you could just tell that they put in the work to figure out the systems and the people and the processes to go into each individual US state and their government compliance stuff and their tax codes. And then maybe they do international stuff. Then all these different things, dealing with contractors, dealing with W-2s.
Brian Casel:They figure out all the paperwork and you could tell, yeah, they have
Jordan:a
Brian Casel:slick interface, but they have people, they have processes. They tracked it all down. Like you could just see, and it's like, I would never want to run that company. It would be in design. Like that just seems That's
Jordan:the opportunity. Nobody would steal.
Brian Casel:A thousand headaches, but like they, yeah, exactly. That's the opportunity that they figured out how to crack, You know? Yeah.
Jordan:It's it's interesting. I think it's another example of the temptation just being too great to cut corners. Right? These founders were on their way to being billionaires on paper. It was a straight line from here to there.
Jordan:Yeah. Insanely fast growth, amazing valuation, amazing investors, and they just went for that incentive and weren't careful.
Justin:Yeah. The the it's a Ponzi scheme. If if if you're if you're taking money for something that's not real, that's a Ponzi scheme. This is this is what we
Jordan:need to I disagree. But go ahead.
Brian Casel:Need to
Justin:we need to root this out of the startup ecosystem. This this move fast and break stuff in a in a way is good advice. But in another way, if you get a 22 year old founder whose prefrontal cortex isn't fully developed yet, and you're telling them to make all these decisions, and they're taking all these shortcuts. Like, the the it is a Ponzi scheme if you're selling something that's not real. So if you're speedrunning someone through sock duke appliance in two weeks, when everybody in the industry knows you need at least a three month window.
Justin:And, you know, there's people on Twitter saying, we got run through this, and we were like, wait, this can't be real. Like, this is, you know, this is not this is they're selling selling us fake compliance. It's a Ponzi scheme.
Jordan:My okay. I'm gonna finish with this. I wanna go to open cloud. Close close that loop we opened an hour ago. I don't blame Delve entirely because everyone around them is in on the same pyramid scheme mindset.
Jordan:Because all I want is to be able to say we're HIPAA compliant so we can take on medical. I don't I'm I'm not gonna steal anyone's data. We're gonna protect it anyway. We're already doing we're already being careful and responsible in our software company. I just need to buy the damn certificate so people can sign up because they were told by law they need to work with some of the it's a complaint.
Jordan:So both sides are just just give me the certificate. Just give me the shortest path to saying I'm HIPAA compliant. And they were like, cool. We'll outsource it overseas, and we won't ask them how they do it. They'll give us the approval.
Jordan:We'll give you the approval and everyone's on the same page. Yeah. So that's my only issue with this is that it was all set up to be bullshit because the entire thing is bullshit to begin with.
Justin:Yeah. Okay.
Jordan:Alright. Let's let's close this thing out with where OpenClaw is today. Brian, I wanna bring this to you because you've kinda seen this evolution in general. As a nontechnical person, I don't 100% understand the value of OpenClaw other than running multiple things at the same time. Help me connect that.
Brian Casel:Okay.
Jordan:And where it's going, again, my perspective is I'm good just kinda waiting. I don't think I need to learn all this crazy stuff. I don't think I need a Mac Mini. I'm just looking at and saying amazing innovation breakthrough and now will be absorbed in all the tools, and I'll just hang out for another month or two, and all that power is gonna come my way without having to think about anything. Help help me.
Jordan:What's the response?
Brian Casel:I think that there is a really interesting story here that's been playing out over the past, like, two months, and it's and it's playing out now in in real time still. Openclaw breaks open to the world about two months ago. One of the like, the most fastest growing thing in the world. Like, I did a couple of YouTube videos on it that that went absolutely nuts.
Jordan:Yeah. What's the thing? What what was the breakthrough?
Brian Casel:The the breakthrough that OpenClaw introduced is, like, in in a word, it's like autonomy of an of an like a true actual agent that can truly work autonomously. And and that combined with the interface of chatting with the agent using Telegram or WhatsApp or Discord or whatever chat tool you wanna use. A lot of people use Telegram. How
Jordan:is that different from just going into the chat box of Claude and saying, do this for me, hit enter? Where's the difference?
Brian Casel:The the big difference is in the autonomous nature of it. So it's absolutely true. And this and this is I'll I'll get more to how the story is changing right now with Claude. But originally, the what the the big breakthrough is that with OpenClaw, you can just chat with it and you can give it jobs. You can give it tasks, like scheduled recurring tasks.
Brian Casel:To to me, that like, as a business owner, that's how I've always thought about it. And that's one of the one of the angles that I've really been zeroing in is that, like, I'm I'm actually not I've never been interested in having OpenClaw giving it full access to my personal inbox or my calendar or my everything. No. Like, I it's it's not in my personal computers. It's on its own dedicated Mac mini.
Brian Casel:You can do a VPS too. But the the to me, the the value of it is to think of it like employees. Like, I hire employees. I give them tasks and responsibilities that they can autonomously Okay. Do things day to
Jordan:what they need to do. They walk on Monday to day. They what they need to do. They walk in on Tuesday.
Brian Casel:Know me. I I've always been a process guy back when I was it's it's such a clear parallel to me with Audience Ops Yeah. And this. Right? Audience Ops was literally built on a 100 Google Docs that I wrote out step by step processes, and I gave those Google Docs to my team of 25 humans who carried them out day to day on a repeated recurring basis.
Brian Casel:They were happy to have that sort of gig. I was happy to have a business that runs on autopilot, thanks to processes and and systems, runs predictably. Great. Today's version of that, in my view like, the the big exciting value of of OpenClaw is I can I can have agents following processes? Processes come in the form of skills now, but they are that it's a text document with step by step instructions.
Brian Casel:Like for me, I do a lot of content. So I have a content agent that every morning is pitching new ideas to me. And then I have a process that we go through with drafting. I've got another agent that watches Twitter and has some radar for stuff I want to track. Stuff like that.
Brian Casel:You can't do that, or you couldn't do that, maybe until now. But two months ago, you really couldn't do that type of thing with Claude or OpenAI. Because as powerful as Claude or Claude Code is or Claude Cowork has been, most of the power of those tools comes with me at the computer prompting it to say, do this thing, build this thing. And those builds have gotten powerful, more and more powerful, bigger and bigger. Like, I can have it almost one shot an entire app using Claude code.
Brian Casel:But that's still a task that I have to sit here and say, here's the thing. Go. Do it. Okay.
Jordan:So this cracks me open from having a conversation. Right? The man, now the myth and legend, Peter Steinberger. Right? That's his name.
Jordan:So introduces this concept. Before we go further, I just wanna ask you, Justin. When you saw OpenCLOS starting to blossom
Brian Casel:Mhmm.
Jordan:Did you see application for it in your business or personal life? Were you, like, clamoring for it? Or is it really mostly developers that saw it and kind of grokked the innovation?
Justin:I think it's the me, the innovation was I want an agent that could go do real work for me that I can talk to via Telegram. That seemed like and still does honestly feel like an innovation that I want.
Jordan:So the the channel innovation was was part of it. Of like, I don't need to be on my computer. I don't even need the chatbot that provided by the tool. This is just me having a conversation.
Justin:An interface that can actual do actually do real work. So in the same way that you could set off an agent to go build you something in Claude Code, could I have an agent go and find me a bunch of restaurants this week and come back with a report? Could I have an agent go and look through all of my Dropbox folder and then come back with, like, all of the documents I need for my accountant and then just email them to my accountant? There's like things like that. Multi step automation.
Justin:But the UI the UI the interface is an important piece of that. And then but the actual power of saying, yeah, this agent, I'll go off and look at everything and and see what I can do for you.
Brian Casel:My big thing on that is that, like, there's all sorts of different tasks that we want to delegate. Right? Mhmm. And I do use AI in almost every task that I work on today, from day to day throughout the day. But I'm more than happy to just keep using Claude and Claude Cowork and Claude dot ai and Claude Code when I'm here at the computer and I if I have a big hairy task, if I gotta organize a a big Dropbox folder, if I have to write a a complex thing, I'm gonna I'm gonna open up Claude and I'm gonna have it help me do that.
Brian Casel:Mhmm. Because I'm actively working. It's a huge help, but I don't even I don't really consider that delegation. To me, the big value is recurring scheduled tasks that are happening without me saying go. I just want them I just want that employee on my team to come in every Tuesday morning and do their thing that they always do on Tuesday morning, and I just know that it's getting done.
Brian Casel:And so, like, that's that's the kind of thing that hasn't quite been possible with the Claude's and the and the ChatGPT's. Up until now, there's some new features coming out. But, like so I don't even use OpenClaw. I don't use it for to build stuff. I still use Claude code because I like to build stuff.
Brian Casel:I don't use it for I've got this one off complex task that I only need to do once today. I'm not really gonna use OpenCloud for that. I'm just gonna sit in my desk and use Claude to help me knock this out once. Right? And I'm not gonna use it to help me pick a restaurant.
Brian Casel:I happen to wanna pick the restaurant for me and my wife to go to. Thank you very much. I don't need an agent to help me do that. But I do wanna make sure that every single day I have a new set of eight ideas that I could develop into content Just pitched to me. And then I can react off of those.
Brian Casel:And then I can say, yeah, let's green light these three. Go go off and draft those. I'll review those tomorrow. Like, that's the kind of automated recurring stuff that so for me, it's like mostly marketing stuff. So what's been interesting over the last, I would say month, the team at Claude, man, is on fire.
Brian Casel:Anthropic.
Jordan:Yeah. Incredible stuff.
Brian Casel:They see this okay. So I'll rattle off some of the features, but I think the bigger story is product management here and roadmap, which I'm mind boggled by how they do this. I don't know what's happening behind the scenes there. But just from a product standpoint, Claude has released Okay. In Claude sorry.
Brian Casel:In Claude Cowork
Jordan:Which is new itself.
Brian Casel:It's relatively new. That's been around for a few months. But now they just finally added a feature called scheduled recurring tasks. So so you can have Claude CoWork on your machine, and I just get I I have moved that content pitching task out of OpenClaw, and I have put it into Cloud CoWork because like, why not just have OpenClaw do that?
Justin:Yes.
Brian Casel:Well, there was a big brouhaha about a month ago, two months ago, when OpenClaw came out. Technically, it's easy enough to just connect your Cloudmax plan to OpenClaw. Like, I I pay $200 a month for Cloudmax. I wanna use I wanna use Opus on on OpenCloud.
Jordan:Mhmm.
Brian Casel:You can't. Against their terms of service.
Jordan:Yes. A little TOS jujitsu there. Very nice.
Brian Casel:Yeah. It's technically easy. You can literally a couple clicks and you can connect your you can OAuth your subscription into OpenCloud. It's possible. But then you hear all these stories of Cloud shutting down accounts, and and and they're and they have some behind the scenes tracking of, like, they think that a that a OpenCloud is operating your Cloud account.
Brian Casel:They'll shut you down. Right? And so there's debate over whether that's true or not. There's been a lot granted, I would say that the Anthropic team has not done a great job of communicating, and they've been a little bit dodging the question. That was the story about a month or two ago.
Brian Casel:But the reality is, like, they don't want you using OpenCloud. They want you using Cloud products. But the other reality was, well, the Cloud products can't actually do all the things that OpenCloud can in terms of communication and autonomy and recurring tasks.
Jordan:So So the clock was ticking for them to hurry up.
Brian Casel:And hurry up, they did. Right? That's the story to me is is like they they have scheduled tasks in Cloud Cowork. Over in Cloud Code, they released the ability to say, first they shipped, okay, I start this Claude Code session on my computer. I can remote it over to my mobile and I can leave for the day and continue the task on my mobile.
Brian Casel:Okay, that's interesting, but I still have to initiate it on my desktop. Two weeks later, they ship, oh, now you can spawn it the other way. You can be out on mobile and now you can spawn a new Cloud Code session from mobiles. So that's really interesting. There's still some like little technical gotchas that add friction
Jordan:like- Temporary though.
Brian Casel:Exactly, temporary. And
Jordan:Steinberger joins OpenAI and Meta acquires Manus and now it feels like the race is on. The the visual analogy I got in my head as you kinda finally explained or got through to me on on on the difference between something using clawed versus open claw. It's like the Fantasia Disney cartoon where if you have a wand, a magic wand that can do things and gives you amazing leverage with magic, that's using clawed. But Open Claw is bringing everything to life that cleans for you, that goes off and does it for you without you. And that concept, as soon as that hit and everyone understood how big of an innovation it was, it was just a matter of time before that got absorbed into the existing products.
Jordan:I have to give credit to to Anthropic to basically say, we're actually going to enforce our terms of service terms of service, at least to the point where the rumor out there will be that you can't do it, and then we're gonna sprint and get people this power in their hands as quickly as possible.
Brian Casel:Yeah. And I actually don't I agree with you. I I actually don't blame them. Like, obviously, they should enforce their terms, their their tweets were, like, intentionally misleading a little bit.
Jordan:Well, they don't wanna be the bad guy. You you can't be you can't be cool and shut down the coolest product in tech at the same time.
Brian Casel:But it's also really interesting how it's like, okay. There was there must have been some internal decision making process that looked something like this. Like, oh, shit. OpenClaw is here. OpenAI just bought them.
Brian Casel:We we missed that opportunity. People are getting angry at these misleading tweets we're putting out. What do we do about that?
Jordan:Fix it now.
Brian Casel:Let's just deal with the with with the heat that we're gonna receive for a few months. Meanwhile, go heads down and ship the features we need
Justin:to ship.
Brian Casel:Right. Scheduling, remote, squad, co work. And, like, literally within six weeks, they're shipping these features. So what wanna know, I don't like, Claw team is is really transparent on Twitter. You can follow multiple team members over there and and watch them drop these features, which is fascinating.
Brian Casel:But there there is some behind the scenes product management story here where it's like they are operating on another level. How do they prioritize and decide to ship these features and get them into a shippable state? We know that they can build stuff quickly. Get that. They're obviously using Cloud Code, right?
Jordan:The best version.
Brian Casel:Because Jordan, we've talked about you and your team and Justin too. How do we adapt and change the way that we ship SaaS products? Right?
Justin:Mhmm.
Brian Casel:To ship these features that have this massive worldwide scale of of user bases.
Jordan:They're adding a million users a day. At all at the same time as as this is all happening, there are a million sign ups a day.
Brian Casel:Like, people see the the the the speed of building and shipping, which is obvious, but it's to me, what's more impressive and more mysterious is like the decision making behind it. Like, who's who's saying like, we have to prioritize these features and hold off on those other 100 features that we have, you know?
Jordan:I I think look. I have interacted with some of these teams in SF a little bit, and my general feeling is that the level of seriousness and commitment is well, well, well beyond what we are willing to accept in our lives. I leave this desk at about 05:45 and six. I go downstairs, and I cook dinner with my family, and I help them with homework. And then I'll watch a little TV and I'll go on the laptop.
Jordan:Like, I'm not living the same life as these people right now. They are 100% dialed, locked in 247, every Saturday, Sunday, every night, and it's an incredible thing to kind of see and benefit from. But these people are not like us right now. They're they deserve the progress they're making and the money that they're gonna make and all that because they are way, way, way out on the edge on that. It's the talent, the tech, the commitment, all of it in SF right now is something to admire.
Jordan:Mhmm. Yeah. Yeah.
Brian Casel:For sure.
Justin:Yeah. The other thing you didn't mention is this Dispatch thing they just launched as well, which
Jordan:is Yeah.
Justin:Yeah. Basically, you can use co work on the go.
Brian Casel:The the one Yeah. Another way, like, to to from your mobile device to to to what what's interesting is, like, you're not communicating into the cloud. I'm communicating to my computer. I could be out or I could be somewhere else, and I'm on my mobile device and I'm talking to the computer that is on my desk. That's the unlock because on my file system is where my skills live.
Brian Casel:They don't live in the cloud.
Jordan:Oh, that's why.
Justin:Yeah. It still feels a little bit janky to me because like people need to keep their computers running basically and out of sleep mode. I heard Yeah. What? Can't you put
Jordan:it into Dropbox or something?
Brian Casel:That's been the technical thing I've been working through today. It's like, how do I make sure that's Alright.
Jordan:So that's next. So they'll handle that next.
Justin:Okay. Yeah. Well, people are like force it. There's like this thing called there's a way to force your laptop to stay awake even when it's closed. There's an app to do that, I think.
Justin:Okay. So people are walking around with their laptops.
Jordan:That's absurd. That's so temporary. Yeah.
Brian Casel:And I'm accessing it on my Mac Mini, which doesn't have a monitor. It's just the Mac Mini. You know? Yeah. I just remote into it, and and but but I had to, like, put set up some settings that prevent it from going asleep so that I can always access the the tools.
Jordan:Yeah. Yeah. Yeah. So it's been, like, an hour and a half. We have to stop.
Jordan:I was gonna ask about spring break and how we're doing, but I think for the respect of our audience, we're gonna call it we're gonna say thank you to everyone. Great conversation, guys. Yeah. This was great. Enjoy the weekend.
Brian Casel:Good hosting, Jordan.
Justin:Yeah. Thanks, Jordan. Thanks, everybody. Later.