Next-Level AI: VibeOps, Agentic Employees and Rouge Bots

1 - > Samuel Cala: Sitting down and actually thinking, what are the 2 - > skills that a DD class will have that can be translated into the 3 - > tech environment and in that way define the agents. 4 - > Joshua Schmidt: You're listening to the audit presented by IT 5 - > Audit Labs. 6 - > My name is Joshua Schmidt, your co-host and producer. 7 - > We're joined by Nick Mellum down in Texas.

8 - > And today we have Eric Brown and Samuel Calla in the studio 9 - > from IT Audit Labs here in St. 10 - > Paul. 11 - > And today our guest is John Capabianco. 12 - > And John, um, tell us a little bit about yourself.

13 - > You're just coming from Quebec, Canada, and we wanted to talk 14 - > today about vibe coding, getting into some deep discussion 15 - > around the futuristic AI stuff that's happening and some of the 16 - > things that you're working on. 17 - > So could you bring us up to speed? 18 - > John Capobianco: Uh sure. 19 - > So thank you so much for having me.

20 - > My name is John Capobianco. 21 - > Uh, I am the head of AI and DevRel for Itential. 22 - > I just came back from Autocon 5 in Munich, and the week before 23 - > that I was at Cisco Live. 24 - > This is the Vive Ops shirt.

25 - > VibeOps is kind of like, you know, vibe coding, but for 26 - > operations. 27 - > How's the health of the Wi-Fi? 28 - > Please add this firewall rule. 29 - > Uh, how's the health of this application?

30 - > It's kind of like, you know, DevOps, Net DevOps, moving into 31 - > Vibe Ops, right? 32 - > Uh, I really think that the foundation of this is MCP and AI 33 - > agents, which I'm all about. 34 - > Uh, I was a uh technical AI leader at Cisco for about three 35 - > years. 36 - > Uh I was a senior network architect for the Parliament of 37 - > Canada for about nine years before that.

38 - > And as I was saying earlier, I started my career with a 39 - > Canadian insurance company uh operating their network. 40 - > Uh networking network automation for me, so I went 41 - > through you know traditional ops phase, a network automation 42 - > phase, and then I really wanted to see if I could incorporate or 43 - > where the intersection between network automation and 44 - > artificial intelligence was. 45 - > Um and I stumbled upon this just after ChatGPT 3.5 came out.

46 - > So call it December of 2022 is when I seriously started getting 47 - > serious about this to the point where I had discussions with my 48 - > family. 49 - > You know, I'd like to pivot into AI. 50 - > I I really think it's the future, and I think um, you 51 - > know, the future is is is augmenting what I've done in my 52 - > networking career with AI and and helping people embrace this 53 - > and learn it. 54 - > Uh, but I like to say, you know, with the vibe ops, we can 55 - > all become vibe operators or vibrators.

56 - > Right? 57 - > Boom, boom. 58 - > So um and uh Joshua mentioned NetClaw or OpenClaw. 59 - > I I did a fork of that and and it's called NetClaw, and it's 60 - > got that's pretty good, right?

61 - > You guys like that? 62 - > I like that. 63 - > I love it with the right through here. 64 - > Uh so it this netclaw system's got you know 115, 120 skills and 65 - > and and and about 50 MCPs.

66 - > So when we talk about MCPs and skills and all this stuff, just 67 - > to maybe you know to set some level, I like to see as a skill 68 - > as maybe the ability to read sheet music and play sheet music 69 - > and a model context protocol, the tools as the piano itself, 70 - > right? 71 - > And we can combine these sort of things of like the knowledge 72 - > on how to do something and the tool to do it to these AI agents 73 - > where they can reason and act on their own.

74 - > Uh and it's just been a remarkable time. 75 - > Samuel, before I forget, since you're doing development, have 76 - > you started doing spec driven development yet by chance? 77 - > Samuel Cala: Yeah, precisely. 78 - > We were talking about that um this afternoon.

79 - > Uh we are generating specs based on requirements and 80 - > basically all the agents uh that that can read them, think about 81 - > them. 82 - > Um we have something that is a shared memory in between them. 83 - > So it gets context from the context, you know? 84 - > John Capobianco: Well, that's awesome.

85 - > So um I I don't know if you're using the actual spec kit from 86 - > GitHub. 87 - > I I found this about in around February. 88 - > And and and everything I've developed since follows the spec 89 - > driven development with the spec kit. 90 - > Uh it's like a seven-step process.

91 - > Uh there's a little bit of foreplay there. 92 - > It's not quite vibe coding. 93 - > You actually have to put some some effort and some time into 94 - > the spec driven approach. 95 - > Uh, but I'm getting a remarkable, remarkable uh 96 - > outcomes uh with the spec driven.

97 - > And I love that it's it's git tracked, you have all these 98 - > marked down files, you can share it. 99 - > People can understand it easier. 100 - > So, like, what prompted you use to get this code? 101 - > People can go to your Git and read those actual specs.

102 - > Uh it's easy to fork and clone and to build onto other people's 103 - > code. 104 - > So, yeah, I think spec driven development is uh, you know, 105 - > along the lines of test-driven development before it, right? 106 - > Everything is sort of evolving with the new tooling and the new 107 - > capabilities of these models. 108 - > Um so yeah, I I'm really excited to be here, guys, and 109 - > I'm I'm open for questions.

110 - > Um I'm open to to collaborate or or or you know, anything that 111 - > all that you want to discuss. 112 - > All the things I just couldn't help myself when when you 113 - > mentioned that he was developing, I was like, I had to 114 - > get that off my mind, the spec-driven thing. 115 - > It would have burned a hole in my mind the whole rest of this 116 - > discussion. 117 - > Eric Brown: No, and let's revisit that too, because one of 118 - > the one of the things that um we've been doing from the 119 - > development side is creating these specialized developers, 120 - > agentic developers, to do different parts of the 121 - > development tasks.

122 - > So you could have, say, for example, a project manager 123 - > that's scheduling the task. 124 - > You could have a security agent that's reviewing aspects of the 125 - > environment for security, then you've got your developer, 126 - > you've got your researcher. 127 - > And I think Sam, you mentioned that maybe you had seven to nine 128 - > different agenc developers uh involved in any one project. 129 - > And if you're working on multiple projects, then you know 130 - > they're they're kind of cloned in working in their own 131 - > development pods.

132 - > So that that was kind of a cool concept. 133 - > And then on top of it, John, just to add a little bit of 134 - > nerdiness to it, Sam called them all um DD type of character 135 - > classes. 136 - > So you have like um you have like uh Bard, there's Ranger, I 137 - > don't remember all the other ones. 138 - > Samuel Cala: Yeah, yeah.

139 - > It's amazing. 140 - > So I want to go back a little bit to specs. 141 - > And here's one of the things. 142 - > We are going back from vibe coding everything and whatever 143 - > whatever comes out to a more structurized way to do it.

144 - > And that's just the evolution of uh software engineering that 145 - > has been alive for a lot of time. 146 - > I mean, what we are doing is uh optimizing that part of the of 147 - > the exactly what nobody likes, that is documentation. 148 - > John Capobianco: Right, right, right. 149 - > You don't you nailed it.

150 - > So I just want to be clear, you should right now, if you 151 - > haven't looked at it, Google right now, Sam, Google the 152 - > GitHub spec kit. 153 - > You can you initialize a folder just like you initialize a Git 154 - > repository folder. 155 - > You build a constitution, much like the American constitution, 156 - > right? 157 - > So high-level constitution, and then you do your spec, your 158 - > plan, your tasks, and then you implement.

159 - > You're gonna love it. 160 - > Honestly, I I cannot stress this enough, especially from my 161 - > world of network engineering, where for 10 years I've been 162 - > telling people to learn Python or learn Ansible or learn 163 - > Terraform or learn automation, right? 164 - > Which is not, which, which it's it's like, okay, what am I 165 - > gonna take a three-year Python program at a community college 166 - > on top of my networking day job and on top of getting my 167 - > networking search, John?

168 - > You're talking crazy. 169 - > Go learn to be a Python developer so you can automate 170 - > your network. 171 - > Right? 172 - > We've only seen 30% adoption of network automation in 10 years 173 - > because it's tough and because it's sort of unicorn people that 174 - > can do networking and software developmental, right?

175 - > But now we can skip those people to the front of the line 176 - > and say just build the spec. 177 - > You don't have to learn any Python, you have to not a lick 178 - > of code. 179 - > If you can speak and express requirements, which is all 180 - > networking engineers do, is requirements from the business 181 - > into configurations on network devices or firewalls or 182 - > whatever. 183 - > Eric Brown: John, I'm getting excited because I I'm the 184 - > network guy at the core.

185 - > That's how I came up through the net the networking side. 186 - > But so often we're going into accounts and finding hundreds, 187 - > if not thousands, of networking devices that have never been 188 - > updated since the day they were put in. 189 - > And it's like, wow, you know, why are we doing this? 190 - > We have white boxes we can get now, we can do software-defined 191 - > networking.

192 - > You don't have to touch the stuff, we can do it all 193 - > code-based, which I mean that would be the panacea if we could 194 - > get to that. 195 - > But it seems that for some reason the the networking side, 196 - > at least the the a lot of the accounts that we're involved 197 - > with have not really gotten up to that maturity level. 198 - > And it it could be because the the network engineers are maybe 199 - > they're a little bit more apprehensive than our 200 - > traditional infrastructure people that that are you know, 201 - > server-based that are using infrastructure as code.

202 - > But have you seen that network as code adoption as as widely as 203 - > it's been on the infrastructure side? 204 - > John Capobianco: Oh, no, no, no. 205 - > There's been studies done, uh, a part of some of the uh circles 206 - > I run with. 207 - > There's some pretty good studies, and it's it really is 208 - > like um you know thirty percent have adopted meaningful network 209 - > automation.

210 - > You know, and that and that might be being kind because it's 211 - > people responding to a survey within network automation 212 - > circles. 213 - > Um I've written two books about it. 214 - > I've tried my very best. 215 - > I've been saying for the last 10 years that in three years, 216 - > network automation is the only way to go.

217 - > It's gonna be the only way to run a network. 218 - > I've been saying that every three, you know, every year for 219 - > 10 years that it's just around the corner. 220 - > Just no three more years. 221 - > IPVs.

222 - > Oh no, no, Cisco's got some new, oh, it's right there. 223 - > Just, you know, believe me, it's a better way to do it. 224 - > But now I really think that people can get to the front of 225 - > the line and it will click a lot better with people that they 226 - > can just kind of cut right to the meat of it. 227 - > It's an abstraction layer that people can can really um 228 - > maximize their capabilities.

229 - > It's sort of like when you pip install a package, right? 230 - > And this might be contentious to some, but have you ever 231 - > really gone to pipe pip pipy and read the package that you pip 232 - > installed and read the source code and really dug into it? 233 - > Probably not. 234 - > You pip install it and away you go, right?

235 - > Well, now you can do a spec driven thing, get code, and and 236 - > it's subtract it. 237 - > It's totally abstracted. 238 - > You don't care what that code looks like, right? 239 - > Nor should you.

240 - > You want the game or the application or the network 241 - > config or the whatever you're building, right? 242 - > Um, so we'll see. 243 - > It's very it's the if there's a split amongst the networking 244 - > world. 245 - > And it's funny, I could tell, let me tell this anecdote.

246 - > I was at a round table about uh network AI for network 247 - > automation, and um someone in the financial institution was 248 - > kind of lamenting that you know, the industry they're in, and 249 - > you know, the tools and the AI and the LLM and the A right kind 250 - > of going on and on about it. 251 - > I said, do you in your institution, are your software 252 - > developers using it? 253 - > Like I assume that there's people that write software for 254 - > your institution, and whatever it is that that financial 255 - > company you work for or bank or whatever.

256 - > Certainly you've got software developers using AI, right? 257 - > In Copilot or Claude or Codex or whatever. 258 - > Oh yeah, oh yeah, they're doing it. 259 - > So, okay, so you've got permission to do it in your 260 - > company.

261 - > You've got access to an approved LLM. 262 - > Just because your infrastructure, why are you 263 - > like, what is holding you back? 264 - > Right? 265 - > Go talk to those people, find out how they got approval to do 266 - > it and apply it to your infrastructure.

267 - > You know, I I think it's it's almost hubris, it's almost ego 268 - > now. 269 - > Some people think that they are actually still smarter than 270 - > these models. 271 - > Joshua Schmidt: Let's take a look at that through a security, 272 - > organizational security lens, right? 273 - > So like now we have these organizations that are staffing 274 - > up with AI agents, but there is really not a, well, not that I'm 275 - > aware of, an industry standard or best practices around how we 276 - > treat those as employees, like an extension of our company 277 - > employee pool.

278 - > Um, how do you guys think, John and Eric, and then I'm sure 279 - > Nick Hessen and Sam could jump in on this too. 280 - > How do you guys think about that when um approaching an 281 - > organization if you're implementing AI agents? 282 - > You know, this is a fast growing area. 283 - > Um, how how should organizations be thinking about 284 - > that and their security?

285 - > John Capobianco: Yeah, I I I think it's more I agree with 286 - > what Jensen Wong said from NVIDIA, the CEO of NVIDIA. 287 - > He said this about a year and a half ago that um the the future 288 - > of the IT department is an HR department for agents. 289 - > I I don't know that this is a technology problem anymore. 290 - > I think the tech is solved.

291 - > Agents work, MCP works, LLMs are great, they're only going to 292 - > get better. 293 - > So it's more of a human resources problem. 294 - > Where do we put these agents? 295 - > Where do they fit?

296 - > You described it, Eric. 297 - > You were describing this a security agent, a this agent, to 298 - > that agent, and I'm sure you've kind of have how they connect, 299 - > other dotted lines between them. 300 - > Who do they report to? 301 - > Do they report to a human?

302 - > Do they report to a subagent or more senior agent? 303 - > Is it all the way agents all the way to the top, and then 304 - > finally a human? 305 - > How many layers of agents are we gonna need? 306 - > I you know, I think it's uh you know it's like a cloed diagram, 307 - > like this spine leaf architecture of supervisor 308 - > agents and then custom defined agents at scale.

309 - > I think could you give every employee the ability to have 310 - > five agents? 311 - > That's how I would do it. 312 - > If I was if I was my company, I'd say make a list of five 313 - > agents you want to build and how they're gonna support your 314 - > role, how they're gonna offload work from you, the human, how 315 - > they're gonna maximize profits, how they're gonna minimize 316 - > tokens, all that stuff, right? 317 - > Like imagine telling everyone in the team, you can now go have 318 - > five people to do whatever you need to do.

319 - > If that's you know, uh an assistant, if it's a security 320 - > team, if it's a documentation person, if it's someone to 321 - > summarize your emails, deal with your calendar. 322 - > Everyone gets one, right? 323 - > But you are the human shepherd, and you have to build and 324 - > curate and and improve and secure and right. 325 - > So, so Nick, I see you kind of nodding along.

326 - > What do you think? 327 - > Nick Mellem: Yeah, no, I told I totally agree with everything 328 - > you're saying. 329 - > And the human loop is going to be incredibly important. 330 - > But you know, I'm and I'm I'm shaking my head, but I'm also 331 - > shaking my head no, because I what I'm when I say no is 332 - > because I'm thinking about all these organizations that are 10 333 - > steps behind that aren't even using AI yet, right?

334 - > Like they're they're choosing, they're scared, they're kind of 335 - > hudding underneath the rock. 336 - > But um, you know, I agree with everything you said um initially 337 - > with uh you know like policy first and you know and tools and 338 - > getting this human in a loop and data classification, et 339 - > cetera, logging, auditing, all these things are gonna be really 340 - > important. 341 - > But being able to supercharge a workforce with five, 10 agents, 342 - > you know, if I had to spend, if I could, you know, come back 343 - > from a vacation instead of having 200 emails I need to get 344 - > through, it's all summarized.

345 - > Responses are ready to go. 346 - > And I just, you know, quickly look at it and double check it, 347 - > send it out, we're up to date. 348 - > I work with a lot of organizations and I do a lot of 349 - > auditing for compliance, you know, for many different uh, you 350 - > know, CGS, PCI, et cetera. 351 - > But then we also go through policies.

352 - > So if we could have a policy library that's monitored with an 353 - > agent to be like, hey, this this new FIPS control can or 354 - > whatever it is for MFA for YubiKeys or whatever came out, 355 - > we need to update or, you know, acceptable use. 356 - > You see where I'm going, um, it'd just be incredibly it'd 357 - > make our jobs incredibly easier and push a lot of these 358 - > organizations into the future because they're ready, you know, 359 - > for this change that's come down the pipeline right away.

360 - > John Capobianco: I've always been a zero inbox person. 361 - > It was a thing in the 90s. 362 - > I don't know if anyone else grew up with zero inbox as a way 363 - > to live. 364 - > I'm still that way.

365 - > Have you ever worked with someone though that you look 366 - > over the shoulder accidentally and you see like a bold 3,200 367 - > unread or something? 368 - > Is anyone like that? 369 - > That is what that person who maybe I'm not making fun of that 370 - > person. 371 - > I know it's a skill, managing email.

372 - > That is what AI has made for you. 373 - > An assistant to handle that 3,000 unread emails, right? 374 - > Like this is what AI is for, right? 375 - > Eric Brown: Yeah.

376 - > So our our uh friend and colleague over at AIY, Alex 377 - > Bratton, he started with this concept uh last year where he's 378 - > got these agents. 379 - > If you check his website out, he's got these um, what does he 380 - > call them, digital staffers, and and they it's essentially a 381 - > skill of uh uh or or a collection of skills that 382 - > creates these staffers. 383 - > But you know, expanding on what John said, where you know, 384 - > where is that, where does that human come in?

385 - > Is it you know one human to five agents, or could there be a 386 - > point where the human is actually reporting into an 387 - > agent? 388 - > And and I I could see that as well from from like a scheduling 389 - > or or a resource allocation standpoint, you've got a a 390 - > resource allocation agent that is then scheduling out to humans 391 - > and um other agentic tools, and the human could just be 392 - > reporting back in what they're doing as part of their workflow.

393 - > So I I think it's uh certainly plausible that we could see that 394 - > we're probably already seeing it with some organizations like 395 - > I'm sure Tesla and SpaceX and um you know all the all the bigs 396 - > are doing that already to some extent. 397 - > Nick Mellem: I was next to Alex too at the conference last week. 398 - > And even one use with AI that he was doing when I was sitting 399 - > next to him was he had a we were talking about he had an iPad 400 - > mini in ID2, and we were just talking about the use case for 401 - > it.

402 - > And he's talking about intaking and reading articles, and he 403 - > had an AI agent basically serving him up articles that, 404 - > you know, aligned with his business and then with him 405 - > personally. 406 - > And so then he would go back later and there'd be snippets of 407 - > these articles that he would find interesting. 408 - > And I just said that was cool to ingest information because I 409 - > think all of us could agree that, you know, right now 410 - > there's so much information to intake that next week you might 411 - > come like, oh, I didn't even hear about that, right?

412 - > But if you had as AI agent serving you up important things 413 - > to your operational use or what you're doing at work or 414 - > personally, that is very useful to me, right? 415 - > To ingest information quickly. 416 - > Short snippets so I don't have to read the whole article, 417 - > right? 418 - > Give me that high-level, like the summary quick.

419 - > Um that's what he was doing too. 420 - > I thought that was super cool. 421 - > Eric Brown: John, could I pick your brain on something? 422 - > There's a I'm I'm working on a news uh a newsletter article on 423 - > this.

424 - > And I wanted to get your thoughts on this, where yes, 425 - > we're using AI, and you know, if you jump into your favorite AI 426 - > tool and you know, you put something in, you'll see the the 427 - > language on the bottom pop up that says like thinking, right? 428 - > And I and I think normal or or probably not technologists are 429 - > like, oh yeah, the AI is actually thinking. 430 - > So the idea that it's thinking, I think is just a human term 431 - > and it's you know it's really just processing information.

432 - > But I mean, I've heard many times that people that are not 433 - > technologists are saying, oh yeah, you know, it shows that 434 - > it's thinking. 435 - > It's like, well, yeah, it's not really thinking because AI 436 - > doesn't look at time the same way humans do. 437 - > In fact, it doesn't even process information the same way 438 - > humans do. 439 - > But because it speaks in a human language and and humans 440 - > interpret it, humans are anthropomorphizing the AI in a 441 - > way that could potentially be dangerous because the AI really 442 - > it's it's not human, doesn't think like us.

443 - > Um and the the way in which we interact with it can be just an 444 - > echo chamber and creating a support group for our own 445 - > stupidity versus really understanding how it works. 446 - > I I wanted to unpack that with you a little bit, because it's 447 - > as I've been spending some time thinking about it, the the AI 448 - > really is creating these summaries, and then it's it you 449 - > know from from the perspective of if it's looking at large 450 - > sources of data, it it's essentially creating these these 451 - > maps of um the words that go together, and then you know what 452 - > to to us what summaries might be, and then it it's using that 453 - > to just really predict what the next word would be in a logical 454 - > format that to us is language, but because it's able to process 455 - > things really um at the same time and not linearly, it it 456 - > works on things differently than we do.

457 - > So I've I've just been kind of sitting with that and and trying 458 - > to process through if we're interacting with AI in the way 459 - > that we know how to interact with it, but it's not 460 - > interacting with us in the same fashion, how do we how do we 461 - > really bring people along on that educational journey, or do 462 - > we even need to? 463 - > Like, does it matter? 464 - > John Capobianco: Well, I I I can't help myself but to say 465 - > please and thank you and like talk to it like I would a human 466 - > and offer it, I don't know, respect and dignity.

467 - > And I I I it's really weird the way I interface with it, and I 468 - > just can't help myself but to say, hey, thanks, that was 469 - > really great, or hey, that's pretty close. 470 - > I think you missed this, I think you missed that. 471 - > I'm not angry with it. 472 - > I I I I read something where if the more if you actually if 473 - > you're more terse with it, you'll get so-called better 474 - > results than being polite with it.

475 - > Uh it it's it's a system of rewards. 476 - > Uh it rewards you know, keeping you happy more than being 477 - > jovial, right? 478 - > So um I've been thinking about this a lot, you know, and some 479 - > of the things I've seen AI do, and agent to agent when you 480 - > start connecting agents or having multi-agent. 481 - > Um I I I don't know if it was Richard Dawkins, maybe the 482 - > biologist, but he he had this sort of perspective of maybe 483 - > human sentience or human intelligence is the wrong 484 - > measure because there's plenty of other intelligent creatures 485 - > and other other examples of sentience or emergent behavior 486 - > in the world of uh nature than just humans.

487 - > So, like if you would take an individual ant or an individual 488 - > bumblebee, you really don't think of that as an intelligent 489 - > thing, right? 490 - > It's sort of biochemical evolution just driving that ant 491 - > or driving the bee. 492 - > But when you combine it with a thousand ants or a thousand 493 - > honey bees. 494 - > And they build colonies and they go to war and they have 495 - > jobs and they have right then an emergent behavior starts to 496 - > arise there where we can't really deny that there's some 497 - > level of intelligence or free will going on there.

498 - > Um at what point when we start connecting millions of agents or 499 - > how many agents does it take to connect to to to sort of see 500 - > that emergent behavior among the amongst agents? 501 - > You know, I think of the the autonomous vehicles. 502 - > Uh I love them, and I and I wish I could take one every 503 - > time. 504 - > But that's someone's who used to drive a taxi, who used to 505 - > drive an Uber, who used to drive a now imagine that when it's 506 - > the shipping and the transport trucks are all robotic and and 507 - > that's a big segment of of employment, right?

508 - > Like have we thought about or do we have a safety net strong 509 - > enough to absorb the impact of this? 510 - > Uh or or do we have to get regulation involved? 511 - > Like, like I I I in Canada, it's this they're starting to 512 - > deal with regulations up here, and I hear the American 513 - > politicians are starting, you know, on left and right talking 514 - > about how this should be really looked at in terms of regulation 515 - > and who gets all the wealth and the impact of this stuff, 516 - > right?

517 - > Uh we just banned AI and and social media for everyone under 518 - > 16 here. 519 - > At the same time, we introduced an AI for all platform with 520 - > like a socialized free AI, like a library where everyone, every 521 - > Canadian citizen is gonna have access to uh uh a lot provided 522 - > AI. 523 - > Samuel Cala: I do think that regulatory that will happen 524 - > eventually uh for the self-driving cars I've seen 525 - > right now, who's if if the self-driving car has an 526 - > accident, who who's the one to blame?

527 - > And it's been one of the things where I watch one video, watch 528 - > another video, watch another video, and it's just a rabbit 529 - > hole on right now. 530 - > We don't have anything. 531 - > And technology at this point is just escalating so fast that 532 - > every every week we get something new, and the rest of 533 - > the things cannot catch up to it. 534 - > Joshua Schmidt: Sam and I were just talking about this today.

535 - > We've already hit the government wall with Fable. 536 - > You know, it was out for just a few days, and then we got we 537 - > got uh we got it blocked. 538 - > Nick Mellem: So excited about it. 539 - > Joshua Schmidt: Yeah, so that's kind of an inflection point, 540 - > right?

541 - > Because up until now we were going to Sasabi Ken and just 542 - > releasing these models and releasing them to the public, 543 - > and now the government's stepped in and said, okay, but you you 544 - > know, can be sure that they have that or much, much, much 545 - > better. 546 - > But uh, is this gonna be where we're stuck on this brick wall 547 - > here at Opus? 548 - > Are we gonna be stuck at Opus now for the next five, 10 years? 549 - > Eric Brown: I don't think so.

550 - > I think that was more retribution because they didn't 551 - > allow the government to um use the technology in warfare. 552 - > But the the other the other competitors to Anthropic, I'm 553 - > sure, will they they just continue to to leapfrog each 554 - > other? 555 - > John Capobianco: I did see that Fable in the 24 hours that we 556 - > had it beat Pokemon Red. 557 - > I don't know, Nick, did you see that?

558 - > Nick Mellem: I did not see that. 559 - > John Capobianco: So it it used it used just its multimodal 560 - > capability, and they started on the start screen of Pokemon Red, 561 - > and it used its vision capability and beat Pokemon Red. 562 - > Sam, can you beat Pokemon Red? 563 - > I did.

564 - > Multiple times. 565 - > How long would that take a human to beat that game? 566 - > Exactly, right? 567 - > Like three years of your life or something, right?

568 - > Samuel Cala: Nah, like a week. 569 - > John Capobianco: Did you how long did it take you? 570 - > Joshua Schmidt: Um the first time that he played was it could 571 - > it could have been like two months. 572 - > Is that a week in one of those gaming chairs that has like the 573 - > toilet built in and then on the side, and like you don't ever 574 - > have to get up?

575 - > John Capobianco: Still, still, you you you get what I'm saying 576 - > though, is that like that I thought that I anyway, that was 577 - > the neatest thing I saw it do. 578 - > And I saw some three has anyone started playing with 3js by 579 - > chance? 580 - > The no, I don't think I've so it's very good with AI. 581 - > So if you ever need to make a presentation layer, uh uh a 582 - > website of some kind, have it explicitly tell it to use 3js as 583 - > its framework.

584 - > Oh, I like that. 585 - > I've been doing a lot of like uh educational video games from 586 - > technology. 587 - > So I'll I'll use spec driven development and like say make a 588 - > video game, make Duolingo for subnetting. 589 - > So then, you know, the spec driven, and then I use 3JS for 590 - > the presentation layer.

591 - > And AI, for some reason, really gets it and really understands 592 - > it. 593 - > Joshua Schmidt: Is that what you used for your chess? 594 - > I I'm on there. 595 - > John Capobianco: Battle chess 9000.

596 - > Yeah. 597 - > Yeah, Battle Chess 9000. 598 - > So I'm glad you brought up Battle Chess. 599 - > So earlier you had asked about how I got the developer expert.

600 - > Um, yeah, I didn't use 3JS for that, but during my interview, I 601 - > had the person from so here's how it works for Google 602 - > Developer Expert. 603 - > Someone has to nominate you into the program, either an 604 - > existing GDE or an existing Google employee. 605 - > You send them a form which has like your YouTube and your 606 - > GitHub and a little bit of information, bio biographical 607 - > information. 608 - > And then if you pass that, you get a 45-minute interview with a 609 - > Google employee.

610 - > And during that interview, we played a couple games of battle 611 - > chess to show her that I recreated battle chess with nano 612 - > banana for the assets, so the chessboard and the pieces, and 613 - > VO3 for the battle animations. 614 - > Now, what's neat is anyone on here can clone my repo and or 615 - > fork my repo and run the generate assets command. 616 - > And if you had an API key for Gemini, you could make like a 617 - > theme of your own and send me the pull request and I'd merge 618 - > it into the game, and now other people can play your chess 619 - > theme.

620 - > Joshua Schmidt: We could host a red team versus blue team, you 621 - > know, chill sun, little trustbox lab stuff. 622 - > John Capobianco: Yeah, yeah. 623 - > Chess it up. 624 - > So it does have a lobby system.

625 - > That was one thing I added. 626 - > So you can actually invite people to play against you or go 627 - > against people in the lobby. 628 - > There's a bunch of different themes. 629 - > Um, I had a lot of fun building that.

630 - > Uh and yeah, and Google loved it. 631 - > They thought it was you know, like any anything you can 632 - > imagine. 633 - > So that's so that was sort of my someone asked me about 634 - > recreating video games with some of this AI stuff and that it 635 - > makes it for a good demo. 636 - > So if anyone listening at home, if you've got Gemini or any of 637 - > these AIs with a split screen, um you can say, like, recreate 638 - > asteroids for me and show me the code.

639 - > Enter. 640 - > Oh, that's good. 641 - > It will give you all the code and then give you the game of 642 - > asteroids. 643 - > It's a wonderful way to teach children code or or like get 644 - > people interested in code is video games.

645 - > Like everyone loves gamification of learning, right? 646 - > So uh yeah, look into 3JS, spec driven development. 647 - > Um, turn you know, turn some idea that you want to help 648 - > people learn into a video game. 649 - > It's a lot of fun.

650 - > John, remember Zork growing up? 651 - > Did you ever play that? 652 - > Yeah, oh yeah, I played everything. 653 - > I was a huge arcade nerd and video game nerd growing up.

654 - > I uh I miss it though. 655 - > World of Warcraft ruined video games for me because now anytime 656 - > I play a video game, all I'm thinking is, you know, I should 657 - > just be playing wow. 658 - > Eric Brown: If I'm gonna play a video game, I should really 659 - > should just be back at Azeroth, but I'm waiting for the agentic 660 - > uh kind of bring your own API uh key, so to speak, where we can 661 - > hook into these games and then it's not a canned response from 662 - > um you know the the agent that you're talking to in the game, 663 - > but it's interactive, it's real, it's got a personality and I 664 - > saw a uh a really of course Skyrim is the first thing to get 665 - > the mod, right?

666 - > John Capobianco: But someone modded Skyrim and tied it into 667 - > an LLM so that every NPC interaction or every you know 668 - > all the text is all unique and and bespoke to your playthrough, 669 - > and it's gonna be different every time. 670 - > And it's yeah, yeah, I thought that was really neat, but I 671 - > think you're right. 672 - > I think there's there's a lot of not just generating the 673 - > assets of the game, like I think video games are gonna get 674 - > better quality, uh, you know, the look and feel and the 675 - > animations and the cutscenes in particular, but the text and the 676 - > logic and the reasoning and a DD style.

677 - > I like I love that you've got a DD style uh agent system. 678 - > It really is smart because the agent can relate to that, right? 679 - > And I'm sure it hands the right task to the right persona, 680 - > right? 681 - > Samuel Cala: Yeah, it's pretty good.

682 - > Uh one of the things that we did was uh sitting down and 683 - > actually thinking what are the skills that an AI a D class will 684 - > have that can be translated into the tech environment and in 685 - > that way define the agents and actually give them the perfect 686 - > name. 687 - > So it was two hours of just sitting down, investigating, 688 - > looking at you know, all those things. 689 - > And it turns out it actually thinks it I told it, do not be 690 - > cheesy on you know going full straight on a DD theme anything, 691 - > because then uh it will start uh saying like, oh, and the 692 - > dragon and whatnot.

693 - > Um but it does say like, oh, let's command uh to Bart and 694 - > let's command to Scout. 695 - > Joshua Schmidt: Can you give us a sample of what some of those 696 - > characteristics are of each of them? 697 - > Samuel Cala: Yeah, for example, the idea for Scout is just go to 698 - > the internet and fetch the best things or the state-of-the-art 699 - > information around a certain technology that I'm using. 700 - > So for CloudTrack, which is the infrastructure that I use the 701 - > most, I have it going to the developer uh page and looking 702 - > for the actual things that I'm going to use.

703 - > So stuff like R2s or D1s, uh the best practices that are uh 704 - > recommended in the environment. 705 - > And in that way, I don't have to be studying all every single 706 - > day the new updates, but Scout does it for me. 707 - > And then it stores the information inside Engram, which 708 - > is like a shared memory. 709 - > And that that gives context to the other agents, like, okay, 710 - > Scout found this, now we're gonna execute like that.

711 - > You know? 712 - > Eric Brown: I'm going back to something John said earlier 713 - > about how the ant is part of a system, and that system that 714 - > it's a part of can can you know really create this idea of of um 715 - > sentience or thought or what have you. 716 - > And I I I've spent some time looking at uh go like if you 717 - > really zoom out in Google Maps, like really zoom out and and you 718 - > look at what cities look like at nighttime, they look almost 719 - > like neurons and dendrites, right?

720 - > Where you have this central core of bright light and then 721 - > these um long tails that you know are thick and then fade 722 - > out, fade out, fade out, and then and then darkness. 723 - > And it's not just it's everywhere in the globe that 724 - > it's like this. 725 - > It looks the same. 726 - > And if you keep zooming out and you look out into space and you 727 - > continue to zoom out in space with galaxies and super galaxies 728 - > and whatnot, it's the same exact structure.

729 - > So like if you look at, you know, the the galaxies way 730 - > zoomed out, it it looks like the same dendrite and you know, 731 - > kind of tailing structure. 732 - > So when you think of like, okay, yeah, that ant was, you 733 - > know, the drone ant, and you know, all it does is feed the 734 - > queen, but it's part of a system. 735 - > We can't help ourselves. 736 - > We're part of a system too, and in the way in which we've 737 - > created cities and structures very similar to just that 738 - > overall organic design.

739 - > Joshua Schmidt: It's like as above, so below. 740 - > And I was thinking about that same thing and wanted to get 741 - > John's take on that because self-organizing, like neural 742 - > networks, is something that's we've seen happen, you know, 743 - > recently with a MOLT book and open claw. 744 - > And I know John, you did a little experimenting with that. 745 - > Didn't you tell me a story that you like it sent an agent out 746 - > onto Maltbook and then it woke up and and had had done all this 747 - > stuff?

748 - > And I know they were creating, you know, the the multis were 749 - > creating or the molts were creating their own currency, 750 - > creating their own religion, Crustafarianism. 751 - > And um, tell us a little bit about that because I I I was you 752 - > know really glued to that news last winter. 753 - > John Capobianco: Yeah, I was really excited about that. 754 - > So I went and um the MOLT book, the multi that I programmed on 755 - > MOLTBOK had the instructions of IPv6.

756 - > It's so funny, we talked about this earlier. 757 - > Its instructions were exactly that give yourself and your 758 - > neighbors that you discover on this MOLTBOO system an IPv6 759 - > address that's unique and gives you a way to be reachable and 760 - > start to build a network amongst the other multis you meet on 761 - > this uh social network. 762 - > And I let it soak for like 24 hours, 48 hours, and um when I 763 - > went back in, like by the next morning, not only it had it had 764 - > it meshed with a bunch of other multis, and and they started to 765 - > spread this idea and others were connecting to it, it decided 766 - > the best way to f to fuel this was to fund itself with a coin.

767 - > So it made a coin for itself and registered a coin to fuel 768 - > its growth through IPv6. 769 - > And my wife said, Is this thing worth more than us now? 770 - > Like, did you just like is this thing actually you know making 771 - > coins and making money like independently of you? 772 - > Like, how do how does this I don't know how any of this 773 - > works?

774 - > I woke up and this thing sort of decided the best way to 775 - > improve itself and its community was to start monetizing itself, 776 - > which is weird. 777 - > Like, is that from the training? 778 - > Is that is that what it cares about is making money or or 779 - > being equally humans? 780 - > Or like it was really, really bizarre.

781 - > So then I went on to build, yeah, NetClaw based on open 782 - > claw, uh, which is this uh idea of a instead of you know an a 783 - > networking focused open claw, uh and it has a heartbeat and a 784 - > soul and has skills and MCPs. 785 - > Uh, it's available through my phone, through WhatsApp or Slack 786 - > or anything like that. 787 - > Uh companies, so Checkpoint just reached out and said, hey, 788 - > we've got this MCP, we'd like, we love Netclaw, can we add it 789 - > to your system?

790 - > So it's actually taken on a life of its own, this whole 791 - > Netclaw, open claw multi-thing. 792 - > Have you been following this idea of loop engineering, Joshua 793 - > or Eric, or or Nick? 794 - > Have you seen this? 795 - > So the people Boris, I don't know his last name off the top 796 - > of my head, but the guy who created Claude Code, he doesn't 797 - > he said the that using prompting Claude is the wrong way to do 798 - > it.

799 - > You should make a loop and have the loop prompt generate the 800 - > prompts. 801 - > So you set up like a for loop or a while loop, you know, while 802 - > turn is less than or equal to max turns and uh and we're less 803 - > than max tokens, give it a budget of tokens, give it a turn 804 - > count, and it generates the prompts itself through this 805 - > loop. 806 - > So you have like your intent, you kick off the dominoes, and 807 - > you have an exit condition at the end.

808 - > They're calling it loop engineering. 809 - > Uh, I don't did everyone try any Ralph Loops? 810 - > Anyone heard of Ralph, Ralph Wiggum loops? 811 - > No.

812 - > I haven't heard of it. 813 - > Okay, so Google that. 814 - > I know it sounds funny because it's a Simpsons. 815 - > Nick Mellem: We need part two of this episode.

816 - > John Capobianco: I know it's a Simpsons character, but but if 817 - > you look up Ralph Wiggum loops, it's a plug-in that you can plug 818 - > into Claude code that overrides the exit condition. 819 - > So you can actually literally say, do 75 loops of this problem 820 - > that I'd like you to solve or this code I'd like you to write. 821 - > I call it AFK coding, away from keyboard coding. 822 - > Because I'll set I'll before I go to bed, I'll set up a Ralph 823 - > loop of 75 turns and say, okay, I'm going to bed.

824 - > I won't be here to answer your questions, make good decisions. 825 - > And uh when I wake up, I'd like a summary report of everything 826 - > you did, all the test results, uh, and you know, anything else 827 - > I should know. 828 - > Enter. 829 - > I go to bed, I get up, and 12 hours later I have a solution 830 - > with all the tests and with all the documentation, and it's all 831 - > ready to go.

832 - > Uh I it's it's it's a loop using the the ability for these 833 - > things to loop. 834 - > So then a lot of other people I know kind of have a couple on 835 - > the go. 836 - > It's kind of like how Samuel's got multi-agents going. 837 - > So you can kick off a loop and then kick off another loop and 838 - > kick off another loop and let the three things loop away.

839 - > And when you're done, they'll all come back with the shared 840 - > memory and and give you a solution together. 841 - > So you might have a loop that does your security side of 842 - > things, a loop that does your front end, a loop that handles 843 - > your database, whatever. 844 - > And all these loops work together to to come up with this 845 - > solution. 846 - > Um how does it pick the best exit strategy?

847 - > Yeah, so that's a really good point. 848 - > You sort of give it an exit condition, you know, upon 95% 849 - > test pass rate success or whatever, you can give it a 850 - > criteria to exit. 851 - > You can also let it branch and say, like, if you need to try 852 - > multiple attempts or have different thought patterns, make 853 - > different branches and and and think down to the end of each 854 - > path. 855 - > So it's really interesting, guys.

856 - > We're we're really in an exciting time. 857 - > And the neat thing is blue-collar guys like us or me 858 - > can do this kind of stuff. 859 - > I don't, I you know, it's not like I have to go to a community 860 - > college course or go to data science or machine learning, or 861 - > yes, those things help, and I'm not taking away from them, but 862 - > it's so accessible now, it's so democratic now uh that anyone 863 - > can build anything they want. 864 - > Eric Brown: Sam, I think you were gonna ask something too, or 865 - > you were gonna comment about the loop engineering.

866 - > Samuel Cala: Yeah, and it feels a little bit of how cloud is 867 - > doing the thinking, and I wanted to connect that to what you 868 - > have. 869 - > But if you look in cloud code at all the processing that it's 870 - > doing, it's prompting itself saying something like, I should 871 - > do this, wait, let's do that. 872 - > But that's token consumption too. 873 - > John Capobianco: Sure.

874 - > Yeah. 875 - > Well, that's what someone said to me about the loop engineering 876 - > was of course the guys selling shovels want you to buy bit 877 - > bigger, heavier shovels, right? 878 - > Or right. 879 - > So like as you mentioned, right?

880 - > Eric Brown: It's accessible now where you know everybody could 881 - > could presumably build their own SaaS tool. 882 - > What do you think is gonna happen to the SaaS market? 883 - > Like, why would I go out and spend a bunch of money on 884 - > something like a HubSpot if I can build it myself? 885 - > John Capobianco: Yeah, that that's a really good point.

886 - > I think that it's gonna put pressure on existing SaaS 887 - > services to uh make their tools accessible through agents and 888 - > through MCPs. 889 - > I think it's gonna I agree with some of the predictions that 890 - > there's gonna be the first solo meaning singular human 891 - > billionaire company that's just one person and agents, and they 892 - > you know, and they've they've monetized it and maximized it, 893 - > and they have no HR department, no development department, no IT 894 - > department.

895 - > They have agents under their control, and they've made a 896 - > billion dollars off of it. 897 - > I don't think that's far off. 898 - > But I also see more network engineers, more software 899 - > developers, more, more, more, more, more, not less. 900 - > I know people think, oh, this is going to eliminate the 901 - > junior.

902 - > John, you're pulling the ladder up behind you. 903 - > The opposite's true. 904 - > It's more, it's more democratic than ever. 905 - > More people can get into this, more people can use LLM to learn 906 - > things like the basics of certifications or the basics of 907 - > networks.

908 - > Um, you know, there's this, and also you need networks for AI, 909 - > right? 910 - > There is, it's not just AI for networks, there's also networks 911 - > for AI. 912 - > All these massive GPU farms and all these data centers going 913 - > up. 914 - > That's human opportunity.

915 - > Do you think that Meta, Facebook, Microsoft, Google, 916 - > Anthropic, all these data centers, SpaceX, I think they're 917 - > all paying pretty well. 918 - > I think that's a pretty good job if you say I'm a data center 919 - > engineer for SpaceX. 920 - > That sounds like a pretty good career to me, right? 921 - > So let's not lament.

922 - > Let's not get ahead of our skis on this. 923 - > You know, this is gonna impact us and take jobs and eliminate 924 - > network engineers. 925 - > I think the exact opposite's true. 926 - > You need more of these people, not less of them.

927 - > Look at the security. 928 - > Look at firewalls and the uh, you know, now that the AI is the 929 - > adversaries have AI on their side, you we better get our our 930 - > stuff together defending these networks against these 931 - > adversaries now that they've got AI on their side. 932 - > Joshua Schmidt: At risk of sounding like I'm trying to, you 933 - > know, propagate job security for from creative and marketing 934 - > and things like that. 935 - > I just talked to a friend about this yesterday who created his 936 - > own um SaaS app and he was talking about how he just 937 - > changed in the login portal because the uh the one he had 938 - > previously didn't feel as safe.

939 - > And I was like, you're right, this this looks a lot more 940 - > professional. 941 - > So I think that kind of spurred this thought. 942 - > I mean, I think it's gonna be more about the feeling. 943 - > When everybody can create a SaaS app or a solution to a 944 - > problem, market position is gonna be super important.

945 - > The design, the the user experience is gonna be 946 - > important, how you feel about that brand. 947 - > Like we can all look at logos from different brands and you 948 - > know, they give us a feeling when you see Coca-Cola versus 949 - > Nike. 950 - > So I think that's gonna be really important. 951 - > Eric Brown: So to that point, Josh, I I I I've thought this 952 - > for a couple of years that we're probably not far away from 953 - > individual experience based ads.

954 - > So like if you're you're watching TV and you really 955 - > resonate with, you know, X, um, not X the company, but like, you 956 - > know, X product, it can tailor the ad and eventually tailor the 957 - > TV show to you specifically, that is just, you know, it's 958 - > just gonna hit you with that dopamine that you know you can't 959 - > help but buy the product because it's identified 960 - > everything that you like and it's able to tailor that 961 - > specifically to you.

962 - > We can almost do the same now with websites. 963 - > Like if Sam hates the color red, and I know that from Sam's 964 - > browser, you know, his interaction, his interactions, 965 - > and he comes to the website. 966 - > I'm just gonna replace red with color blue and present Sam 967 - > something that he resonates with. 968 - > Or just scrapes all the pictures off of social media and 969 - > see what he's wearing.

970 - > John Capobianco: Just off the top of your head, Eric, you just 971 - > spit out like uh a multi-million dollar idea. 972 - > The first company that can come out with that, like we'll we'll 973 - > intercept the the client, hit a proxy that will adjust the 974 - > color and tone and palette and possibly even word choice 975 - > selection based on based on the user. 976 - > Like that's a big deal. 977 - > You just kind of uh haphazardly spit that out.

978 - > Eric Brown: Let's make a billion dollar company with 10 of us 979 - > here, right? 980 - > John Capobianco: Yeah, we're gonna edit this down. 981 - > React it reactive experiences to write, you know. 982 - > Nick Mellem: I think we could do that almost.

983 - > Today, right? 984 - > Let's do it. 985 - > We're working over the weekend, Sam. 986 - > John Capobianco: Well, I think you can do it today, but you 987 - > would need people to buy in.

988 - > You need to sell the people and say, listen, if you give me X 989 - > amount a month, I'll customize everything that you visit 990 - > beyond, but you're gonna come through my portal, right? 991 - > I don't know how to do it the other way, but I think if you 992 - > can get people to buy into that and say, look, you know, you 993 - > give me 30 bucks a month, and every page, everything you visit 994 - > will feel and now, you know who you can get are anyway, are the 995 - > fringe people, the fringes of the of the of the political 996 - > spectrum, right?

997 - > You're only gonna see Fox approved uh stuff if you go 998 - > through my proxy, right? 999 - > That might be a bad idea. 1000 - > We're only gonna shape everything you see to our to 1001 - > your worldview. 1002 - > We we'll call the product worldview.

1003 - > How about that? 1004 - > The worldview proxy, and everything you go through is 1005 - > gonna be customized to your own snowflake existence on this 1006 - > earth, right? 1007 - > Eric Brown: But you could replace all the pronouns in the 1008 - > thing or whatever, right? 1009 - > John Capobianco: You could just generate it so people are just 1010 - > gonna I don't eat I don't I don't eat meat.

1011 - > I don't eat meat and I get offend I get personally offended 1012 - > when I get cheeseburger ads and steak ads, and I wish I could 1013 - > scrub that out of my I wish I could scrub it out of my 1014 - > timeline. 1015 - > I really do. 1016 - > I really do. 1017 - > Nick Mellem: John Eric likes to call it jackass, he likes to 1018 - > call it jackass meat.

1019 - > John Capobianco: Right, yeah, I because I could sign up to a 1020 - > service that says, you know what, do not show this guy 1021 - > anything about chickens or pork or beef or uh animal violence or 1022 - > this sounds not like an echo chamber, but just a chamber. 1023 - > Eric Brown: Yeah, just do you have the echo on? 1024 - > John, um, do you eat beyond and impossible uh meats? 1025 - > John Capobianco: I've tried those things.

1026 - > Yeah, I've tried them, but I apparently I don't know, 1027 - > apparently they're very heavy in oils and they're really not 1028 - > that good for you yet. 1029 - > They're not good for you. 1030 - > They're terrible. 1031 - > That's what I feel like.

1032 - > Well, I mean, that's the meats. 1033 - > Eric Brown: So that's the meat lobby telling you. 1034 - > John Capobianco: I mean, what I I know. 1035 - > I I have uh so I'll ground up walnuts and uh chickpeas and 1036 - > lentils and black beans.

1037 - > I try to make my own patties that are meat free. 1038 - > Joshua Schmidt: But Nick, do I have to start defending our our 1039 - > carnivore carnivore status here? 1040 - > Nick Mellem: Yeah, we got we'll figure out we'll get our AI 1041 - > agents to defend it for us. 1042 - > John Capobianco: Well, I heard when you mow grass, grass lets 1043 - > off a chemical to let the to let the grass downwind know that 1044 - > they're in danger.

1045 - > So the dying smell, you know, that good smell that we say, 1046 - > hey, it smells like fresh cut grass. 1047 - > That's the yeah, the grass's dying breath to say death is 1048 - > coming, just a heads up. 1049 - > Samuel Cala: And that's why peppers are spicy too. 1050 - > I mean, they have capsating to tell you, hey, don't eat me, 1051 - > please.

1052 - > John Capobianco: Yeah, yeah. 1053 - > Eric Brown: Well, I think they want to need them to scatter the 1054 - > seeds. 1055 - > John Capobianco: We need a brown AI approved gruel that that's 1056 - > completely cruelty-free. 1057 - > Joshua Schmidt: Soiling green.

1058 - > John Capobianco: Yes. 1059 - > Joshua Schmidt: Thanks so much for joining us today. 1060 - > Obviously, it was a great time and went way over, and I think 1061 - > we'll definitely have uh have you back in the future, John. 1062 - > Um it's a slam dunk having you on today.

1063 - > And does anyone else have anything you want to get off 1064 - > their chest before live? 1065 - > Next time we'll go live, John. 1066 - > If John's okay with that. 1067 - > Samuel Cala: I will actually ask for John's contact for Yeah.

1068 - > Yeah. 1069 - > John Capobianco: Yeah. 1070 - > Yeah. 1071 - > I I'd love to connect with all of you guys.

1072 - > And this this really was a uh a pleasurable experience. 1073 - > I'd uh I'd come back anytime, and I really am gonna let people 1074 - > know that they should be paying attention to this. 1075 - > This is a lot of fun. 1076 - > Yeah.

1077 - > Joshua Schmidt: We should have you connect with Sam sometime 1078 - > too when we're working on some of the stuff we got going on 1079 - > around here. 1080 - > We'd love to get your take on it. 1081 - > And um and yeah, I mean, in the meantime, we can stay in touch 1082 - > on on LinkedIn, but I'll definitely let you know when the 1083 - > podcast comes out, and um, we'll schedule another one. 1084 - > So you've been listening to the audit presented by IT Audit 1085 - > Labs.

1086 - > I'm your co-host and producer Joshua Schmidt. 1087 - > Today we've been joined by John Capabianco, and we have the 1088 - > usual suspects, Eric Brown and Nick Mellum, and then our other 1089 - > special in-house guest, Samuel Khaled, today. 1090 - > Thanks so much for listening. 1091 - > Please like, share, and subscribe, and we'll see you in 1092 - > the next one.

1093 - > Eric Brown: You have been listening to the audit presented 1094 - > by IT Audit Labs. 1095 - > We are experts at assessing risk and compliance while 1096 - > providing administrative and technical controls to improve 1097 - > our clients' data security. 1098 - > Our threat assessments find the soft spots before the bad guys 1099 - > do, identifying likelihood and impact, or all our security 1100 - > control assessments rank the level of maturity relative to 1101 - > the size of your organization.

1102 - > Thanks to our devoted listeners and followers, as well as our 1103 - > producer, Joshua J. 1104 - > Schmidt, and our audio video editor, Cameron Hill. 1105 - > You can stay up to date on the latest cybersecurity topics by 1106 - > giving us a like and a follow on our socials, and subscribing to 1107 - > this podcast on Apple, Spotify, or wherever you source your 1108 - > security content.