SPEAKER_00 0:03

Dennis has spent more than three decades helping organizations navigate large scale technology transformations from privatized telecoms, airports, and retail operations in Argentina to cloud migrations, AI implementation, DevOps, and enterprise modernization project for some of the world's largest organization. Today, Tangonet Solution collaborates with over 200 technology professionals across the Americas and helps companies bridge the gap between business strategy and technical execution. In a world obsessed with AI, Dennis brings a refreshingly pragmatic perspective to what on what actually creates value and what Dennis will come to scale and clarity. Thank you for that, Maxine. It's my pleasure. Um we like to start the interviews with uh with a founder story. How did you come to do what you're currently doing?

SPEAKER_01 1:11

The long and winding road, as the song goes. Yeah. So really it started when I I started living overseas in my mid-20s, Maxime, but uh I'll fast forward from that point. But uh long story short, I started working at the UN doing some integration projects and many moons ago in the mid-80s, uh, integrating many computers and mainframes, things like that. And then uh diverse set of circumstances sent me to Latin America uh as part of a project with the UN also. But uh later that became something different. Um, it became an offer to go to work for a privatized company in the early 90s called Telecom Argentina. That company was a privatized version of a state-run telecoms company called Intel. So uh the privatization wave that hit the Americas or Latin America at that time, I got caught up in that. And uh my role there was to build uh or rebuild the technology infrastructure, which was either old, outdated, non-existent, or non-operable uh into something that uh was a better fit for a company that they really wanted to make a market uh-based company. So starting in Telegram, Argentina in the early 90s, and then uh through several other big turnarounds, retail, property management, airports that you mentioned, which was also privatization, my roles were always uh building teams, uh allocating large budgets to renew or rebuild technology infrastructure. Okay, so that was my role then, and because of that, I um uh specifically in Argentina, I became heavily embedded into the technology ecosystem there and became very familiar with you know the resource development from let's say the university stage onwards. And uh when I returned to the States in the early 2000s, uh was working for a company for a while that had recruited me to come back there, but then uh 2007 came along, I decided it was time to do something else, and that's when I founded Tango Net Solutions, which eventually took me back into building uh technology infrastructure and solutions for companies here in the US and uh North America in general, and uh using Argentine talent for that, as I had done when I lived in country. So it was kind of the outgrowth of my time in Argentina, where Tango Net Solutions really has its foundation, uh being embedded into that technology ecosystem, and then from there understanding it to the point that I felt comfortable enough that I knew that those folks that we had the talent to deliver some impressive solutions uh here in North America. That's the high-level view.

SPEAKER_00 4:17

Yeah, yeah, no, I appreciate um as you know, we have uh we have a team in Buenos Aires as well, and it's uh it's amazing the caliber of people that you can access in Argentina. Um, it's a highly educated workforce, and I mean the economy has been quagmire for quite some time and are trying to dig themselves out out of it, but uh it's it's a it's a beautiful country, lots of opportunities. Now, Dennis, where um where do you see the biggest opportunities currently in terms of helping companies kind of uh continue down the digital journey and evolve?

SPEAKER_01 5:02

You know, it's you know, we we have a lot of uncertainty or uh Maxime around where I mean I think we feel it in our own company too, but you know, with the impact of AI and so forth, but we are seeing some potential trends. I say potential because we're seeing some indications, but not it's too early to call them trends. You know, one of the things that I think that uh we're seeing, we're beginning to see, and we're seeing this in different areas, we're seeing it ourselves, we're seeing it with partners who we talk to, is the fact that you can build solutions so fast today with AI and AI assisted development. Uh some people are getting in over their heads, meaning you know, they're just developing too much too fast without the necessary guardrails, and it's just difficult to get uh production versions up and running. Yeah, you can build the code, you can build uh whatever it is, 30,000 lines of code in 3.5 seconds, but from building the code to operationalizing it, yeah, can be a big gap. So I think that's one of the things that companies may struggle with. And I think that I still think that the result is, you know, you're going to kind of come up with a positive outcome. The better you do the software engineering up front, the better the results are going to be. That doesn't mean it's going to be a complete disaster, but we're starting to see situations in which you know we may see complete teardowns and rebuilds, and we might just see minor tweaks before we can get those in production. But that's one area that we're seeing some some movement in.

SPEAKER_00 6:40

For sure. I mean, we like I your your comments resonate quite a bit because if you don't know where you're going, how the hell are you going to get there? Is is kind of the the question I have. And the other one is um at the moment, inference or AI coding is not priced to its market price. So once once you have to pay the true price of something, I wonder how many companies will continue to just churn through codes.

SPEAKER_01 7:09

Yeah, yeah. And I think that there are going to be those cases, for sure there are, you know, and I think that I wouldn't necessarily call that a bad thing or a negative thing, but I think if you just look at it and say, well, look, okay, we learned that we have to do it the right way, and the right way may require more of a software engineering foundation. Well, I think the net result is going to be a better result, you know, even though the I think we're hearing and we're seeing a lot of high incidences and high percentages of failed projects. That just might be uh, you know, a stage that we're going through right now, you know, that we just have to weed out what we're doing. You know, you mentioned something about, you know, if you don't know where you're going, you're gonna get there. You know, you know, how are you gonna get there, right? The thing is, today we have a Ferrari to get there, right? But the thing is we still have the same problem. If we don't know where we're going, we're just gonna go faster, but we're still not gonna get there, right? So I think that's where we're at right now. Until we learn how to handle that Ferrari, until we learn how to get our bearings and understand what it can do, where it can go, and what kind of drivers we need there, then yes, I think there is gonna be some struggle uh from some companies. And you know, we're beginning to see that. We'll have to see how it plays out over the next few months.

SPEAKER_00 8:28

So Dennis, you work with a ton of companies in terms of like helping them transform. Like what is the most common reason or common reasons as to why most projects fail?

SPEAKER_01 8:44

Wow, that's a loaded question. I I think that the the biggest reason is not understanding what the outcome is. You know, what is what what's the end game of this? You know, like if you have a clear enough idea, and you don't have to have a hundred percent idea, but I see companies, for example, one of the common terms you hear nowadays is product market fit, okay. And we've seen that with with some of our clients that they go down this path of building something, and many times we're building it for them, but they don't have the right fit, and they only find that out too late. And I think that's where the real problem is. And I think that nowadays maybe that problem is mitigated to a certain extent because now we have AI assisted development that can help us build faster, sort of fail faster. I don't like that term, but you know, at least to do the you know the proof of concept faster, get to the MVP faster, and to do that market fit analysis. But I think that that's one of the biggest problems is understanding you know what the real market is. And by the way, I think that's going to be a bigger problem now. Um, I think we're gonna have a variation of that problem, a saturation problem. Because now with AI and the fact that we can build so fast, we're gonna have a proliferation of products all over the place, like we already do have. And people can be looking at saying, Well, it's this one today, it's that one tomorrow, it's another one the next day. Yeah, you know, a confused mind says no. That's what's of course, confused no, or they can take another path, and that is saying, forget about all these products, I can just build it myself, I can build it faster, I can build it, you know, to my level of customization. Yeah, and I think that's a possibility, also, a big possibility.

SPEAKER_00 10:41

You know, we're seeing some of that as well. I mean, I I spent uh the reason why I'm laughing, I spent 13 years working with uh EY and Deloitte like with a big force, and and uh um SCP implementation, things like that, and then every company is special, every company has special workflows, and you're like, at the end of the day, all companies are the same. Workflows are workflows. Whether we're talking, order the cash, procure the pay, sells the receipts, like it's the process, the process. Maybe you have some nuances, but um, it you're like most companies are not all that different. The other thing that I um the reason why we laughing is um people underestimate how hard it is to build something, how long does it take? Um how like how much time does it take to build, acquire, aggregate, the clean the data that you would need to to to use this, like like like when on any kind of implementations we do is like data is usually the the slowest stream, and it always takes five times as long, and it costs five times as much money as you as you want it to cost, but that's just the reality. So I think that yes, it's tempting because people jump into like a raplet or a cursor and they build something, but it's very simple, like kind of moving from there to be enterprise grade tool, there's a lot of effort that's involved in that.

SPEAKER_01 12:17

Yeah, it can be it can be a lot of effort, and you know, there's you mentioned data, uh Maxine. Processes are another issue, you know. You talk about SAP implementation or any ERP implementation, right? What were the problems that we used to have with those? Data, processes, people, you know, those are the things that those are the same problems we have today when we think about AI, okay. You know, what processes can we automate using AI? What processes can we adapt to make AI automation really, really effective and really impactful, right? Because that's I think that's one of the things is that if you overlay a technology solution onto an inefficient process, you have an inefficient technology solution. I mean, that's just the basic that's the way we lived back in the days when we were implementing SAP or Oracle Financials or whatever it was, JD Edwards, you know, years and years ago. But we don't we we haven't shed ourselves with that problem, you know. You still have the data layers that you have to pay attention to, you still have those foundational pieces of process data, people. When I say people, I talk about you know, culture, mentality, adaptability to change. All of those things are still valid today. Uh and and I think they can be a great enabler or they can be a great impediment. You know, that depends on the culture, that depends on you know how a company approaches this, especially from the top down, you know. Yeah, or maybe even from the bottom up, you know, and you know, if the bottom can push that innovation upwards, that's that's definitely a solution. But maybe it also has to come from the top down.

SPEAKER_00 14:06

Well, it's uh I've done hundreds and hundreds of projects, and so we we've even developed our our own predictive project uh framework that kind of risk assesses project in real time. And um, you're talking about people, people are the weakest link in any project, and and not because the people are not smart or people are not uh governance you can set, uh uh processes you can build, technology would do whatever you you tell the technology to do. People and change management is the hardest element. And I have seen projects where the leadership is like, we're just gonna slam it in and people are just gonna use it. And people are like, I'm not gonna do that. Um, and so unless you have that active visible sponsorship at the top and get the binding and enrollment and go top down, bottom up, top down, bottom up until you kind of modulate and get to alignment as to why we're doing, doesn't matter what technology it is, it's not it doesn't have a stand, it doesn't stand a chance of success.

SPEAKER_01 15:17

Yeah. And you know, I think there's a lot of talk too, Maxime, around that. I mean, talking about the technology side and the impact on people, and particularly technology people like developers and that type of thing. I was talking to a recruiter a couple weeks ago, and I asked her because I think the recruitment people are very good to get a you know, to get a feel, uh, to get the temperature of the market. You know, where's the market? Are people hiring developers? Are they shedding developers? What's going on? And she told me, and she works at the enterprise level, she says, look, I've got a tremendous amount of demand for technology people, including developers. And I think, you know, maybe it's too early to say why. You know, she also mentioned, you know, there could be some of these AI rescue projects that people are starting to talk about that, you know, you can't vibe code your way to an enterprise as you were alluding to earlier, right? So I think the companies maybe maybe the pendulum is swinging, and maybe we're going to see these pendulums swings a lot more quickly now because things change so much more quickly than they did before. But, you know, as far as technology is concerned, and we hear a lot of news about the layoffs and all that sort of thing from the big companies, but I'm also hearing that you know there is demand for technology resources, in particular software developers. So I don't know that we're, you know, at this stage of Armageddon in terms of you know software development and uh services like ours, for example, which are third-party services that do that. Um, I just think we're in a in a sorting out phase. You know, people are gonna try, they're gonna, you know, they're gonna try to figure it out, they're gonna try to do it themselves, they're gonna try to you know see if you know what are the possibilities of AI. And I don't see that as a bad thing, you know. I mean, yeah, would we like to have more business and not lose it to AI at this time? But I think we're in a I don't I don't think we're in a in a bad spot. I think we're in a good spot. I think that there's going to be plenty of opportunities to build AI-enabled solutions and clean up the data layers and you know, get observability where it needs to be so those solutions can be properly monitored and operationalized in a way that people have full visibility into what's happening. So I think it's all good.

SPEAKER_00 17:41

I uh I agree. Um when it was with Deloitte, we had a term called um purple, purple unicorn. And what this was referring to is somebody like a person that has both a business mindset and skill set and tech and technical skill set, and kind of we merging the two disciplines. I find that AI has helped um make the technology people more business oriented and business people being more technology predilect in terms of kind of like what's possible, because I I mean I I find that with AI with technology being so ubiquitous and so easy to use, like I started to code two years ago. Um, I mean, mind you, in grade seven, that was like many, many years ago, I I I took coding classes, but I'm not a developer and I was never a developer, but I I've always been around technology, being a product man, and so it became a lot easier to access. And and to your point, around kind of like uh companies building things and not knowing where they go. And um, if you are a business person and you have that understanding in terms of what you're trying to build, and you you and and you know what problem you're solving, it's a lot easier to build a solution that you know solves the problem versus you kind of the traditional approach is you got business requirements, the business requirements, the business analyst does this, get translated to technical requirements, technical requirements get translated into code, and you every handover loses some of that signal. Whereas now that chain is compressed, it's a lot shorter.

SPEAKER_01 19:24

Yeah, it sure is, yeah, it sure is. And I think that you know that's a that's not necessarily a bad thing, you know. I I don't think it's a bad thing. I I think it's a thing that would we're just going to we're in the process of figuring out. I say we, I think that a lot of people are. I don't know that anybody, I mean, I well, maybe Claude, maybe the people at Anthropic have it figured out because my understanding is that Claude is building Claude itself, and that you know, that's a pretty amazing story. Uh, but then there are others that are, you know, I think it was Uber, I think that uh have already burned through all of their budget for token utilization through four months of the year or something like that. I mean, I some things like that, you know, that okay, you know, it's it's easy to it's it's easy to congratulate on the one hand, it's easy to criticize on the other hand, you know. But I think we're all learning through this, you know. We are learning, you know. We I'll give you a couple of stories. One, we we had a client that we we worked through um some specifications with. They actually said they had them. It was for an integration with different CRMs, and you know, we probably spent a little bit too much time with them on the on the pre-sale side, you know, helping them fine-tune their uh technical requirements. And when it came time to give them the proposal, they called us and they said, Look, you know, we've got this issue. You know, uh one of our developers took it upon himself to go out and and build this thing and in a four-day or three-day weekend, long weekend sprint, built 75% of it. And we were like, okay, that's fine. What do you want to do? You know, so this was about a month ago, and uh, like I said, I think you know, shame on us a bit for spending a little bit too much time in that in that uh specification stage, right? Assuming that they had, as they said, the the requirements, but on us, it's 100% on us that we should have validated that first. But we did help them through that. It was a it was somebody we'd done business with before, there's a lot of confidence there, so uh, you know, no blood, no foul, let's call it that. And um, so yeah, according to them, they did build this one developer built 75% of the system uh over a long weekend. So, our only ask to them they said, Look, we we appreciate all the work you've done, but we really feel like we need to go down this road. Our only ask to them was can we just stay in touch? Because we want to see how you get that last 25% built. That's all. And it'll be a case study one way or the other, Maxine. Yeah. One way or the other. You know, if they do the 25%, let's just say it takes them a month to do the 25%. Okay. They still come out ahead. That's fine. You know. But if they're still struggling and we're about at the point where we need to do the follow up, in fact, I think we are doing the follow-up with them. You know, we'll we'll find out, you know. Uh, so yeah, that's that's a real AI impact. Um, yeah, the other one is we did build an AI solution for our company up in Boston, you know, very innovative solution, a very innovative, forward-thinking CEO, and not even a CEO of a technology company, he's a CEO of an electronics marketplace, uh, recycling marketplace for electronics, but still a very forward-thinking technology enthusiast, let's call it. Yes. And we built this very unique and and uh and uh robust system that's a kind of a Jira knockoff, but with a lot more functionality, all based on AI, all based on AI, MCP servers, that type of thing. And um, you know, it's largely done, and we're gonna do some integrations now with some other companies, which is kind of interesting, also. But he said to me last week, you know, we had a conversation, I had a conversation with the CEO directly. He said, You know, what do you think I should do with this? Should I take it to market? Should I just use it internally? Should I, yeah, you know, should I just keep it amongst my my other CEO friends, you know, because I'm not sure that I'm the person he's a technology marketplace guy, he's not a tech technology guy, he's not a technology product person. So, you know, yeah, he's got something, and I look at it and I say, like I said to you before, my doubt is that there's so many tools now. How many of these tools are there out there? I don't know, yeah, right. So, you know, do you spend a lot of time, do you burn a lot of cycles trying to get something to market that doesn't have a market, or it does, but the that market is saturated with other things, or do you just say, look, this is my baby, I'm gonna I'm gonna make the best use of this because me and my group of companies that are that are working together on this can really we can supercharge our own companies, and I don't have to spend a lot of time going to market. That's where he's at, you know, and that's and then you know, we sort of talked through and you know, like a little bit of you know, psychological session on that, and I think that was the conclusion he came to. He says, You know, I don't really need to take this to market. We've done a great job on this as it is, and it can stay amongst a smaller circle, and they like it, I like it. Yeah, and who knows, you know, maybe he's losing a great opportunity. We don't know.

SPEAKER_00 24:59

Well, I mean, I spent a lot of my time uh uh coaching and advising founders around strategy and risk, and we we we talk about one-way door and two-way doors, and so I view this as a two-way door. Uh, it's like he decided not to walk through this door, but if he decides to, he can walk through that door in the future. And so, in situations like that, a lot of people want to like they build a product and want to take it to market, but take podcasting because we are on a podcast at the moment. 0.1, not even one percent, 0.1 percent of all podcasts generate 80% of all sponsorship revenue. So you have 99.9% of the podcasts that generate 20% of the revenue, and so um when you think about it, like like if you're bringing a product to market, these technology is no longer a mount because it's with AI coding, you can you can develop technology quickly, unless you do something proprietary that's like the better mousetrap. But in order to be successful in building a product to market, you either have to have distribution, you have to have proprietary data, or you have to be faster than any uh anybody else in terms of being able to build something that the market desires, and so unless you have one of those modes, why spend energy on trying to build something to market? You build a product for yourself, use it, make your business more efficient, make your business more productive, scale it up faster. That that that really becomes your competitive advantage.

SPEAKER_01 26:45

Yeah, I you're absolutely right, Maxime. I think the p I think where we're stuck right now, okay? We're stuck on speed, okay. We're stuck on speed because we're into these vanity metrics of you know, we can build so much code in so little time, right? What we talked about earlier. You can build, you know, you can create thousands and thousands of lines of code just by inputting, you know, some good prompts into CLOD or you know, whatever else you want to use, and but speed isn't the only thing. Like I said, you can get stuck on speed. Okay, okay. Okay, you you go so fast that you get ahead of your skis, and all of a sudden, you know, you're tumbling over them, right? You're trying to figure out okay, what do I do now? I can't, you know, I can't operate operationalize this, you know, I'm missing security, you know, I'm missing scalability or missing maintainability or whatever, you know, I'm I'm having trouble testing it, whatever, whatever the case may be, you know, but because we're stuck on sports and we're we're to the detriment of the other necessary elements to build a solid software engineering platform, right? So I think that you know, once we get you know, once we start thinking more about quality, about software engineering best practices, get back to that stuff as basically the the priority. I still we're still gonna have the speed, you know, we're still gonna have the speed. But you know, now it's just a race to get out there and get to the market. And you know, I'm talking to somebody else also in private equity last week that said you know, they're seeing you know millions of dollars infused into these companies into certain AI solution companies, and in months those companies are going under. Whereas before they may be, you know, they may have had a uh a years long runway. You know, now it's a few months that either they they you know they they sink or swim. You know, a lot of times they're sinking, you know, and you know, tens of millions of dollars are going in, nothing's coming out, they're here today, gone tomorrow. So I think you know, I think we need to, you know, maybe we need to slow down. We're still gonna be faster, but maybe we just need to slow down a bit. I don't know.

SPEAKER_00 29:11

I mean, um the the saying slow down to speed up exists for a reason. Uh, if if we if if we continue down a path of like the the team that's been exploring here around software development, if you don't know where you're going, you can build a product, you can you can build additional things around it all like over time, but you're bloating the product, and so a lot of companies don't take into account the total cost of ownership, particularly when they're building. They they're amazing at TCO if they're evaluating off like uh third-party solutions, like you get procurement and supply chain and legal, and like like everyone is involved and look at like okay, what what's my cost to implement? What's my cost to maintain? But when it comes to building the product, the more Frankenstein the product, the more time it takes to maintain or clean it up or you know, drive the quality.

SPEAKER_01 30:06

Yeah, yeah, and I think that's what we're gonna see. We're gonna see everything, you know, from these massive code bases, or maybe not even that massive, you know, just you know, the these rescue missions that we you know, we go in or others go in and they say, you know, look, this is a teardown and rebuild, you know, yeah, because let's face it, if we're honest with each with ourselves, you know, most developers when they take over somebody else's project, most of them will say, Can we just rebuild this, start over again? You know, that that's the developer mindset. It is, yeah, you know, absolutely, and a lot of times it's going to be the best solution. The best solution is gonna be tear it down, start over again. We still have the tools to go fast, but we need to inject that software engineering layer underneath. We need to build a foundation, a solid foundation underneath this thing in order to be able to go fast in a sustainable way, and that's I think where you know we're gonna find those lessons out. We're gonna find them out the hard way, the easy way, and anything in between.

SPEAKER_00 31:14

Um, Dennis, do you guys do um work in in AI uh in terms of like agendic workflows or any anything in particular particular around AI? Yeah, yeah, yeah where okay, so kind of like uh two related questions. One, where do you spend most of your time? Where are companies unlocking the biggest opportunities and and and and why? Like, like what are they doing to unlock these opportunities?

SPEAKER_01 31:45

You know, I gave you the case of the electronics marketplace um company, you know, that has a product that he doesn't know if he has a market, right? I mean, the opportunity is a question mark, okay? At the very least, he has something internally that he can use or he can use amongst a group of companies that you know work together, that collaborate together, and that's part of what we're working on, also to build to extend the agentic capabilities to integrate with other companies and pull in, you know, pull information from them, send information to them, and adapt to their workflow, or you know, bring their workflows into the mix and account for those. So I think that there's a lot to do there, uh, in terms of you know integrating uh maybe portfolio companies, that type of thing. We're seeing some work in that area in private equity. Um we're and then the other thing is just building what we're doing is just saying, look, we're gonna build these agentic workflows as part of the software development process, and all of that goes to the client. I mean, we build, you know, we're a force services company, right? You know, so we're building, you know, like we have been in the past, and we like we have done for years, we're building uh applications and other platforms. We hand over the IP, it belongs to our client. We hand over the agents, they belong to our client too. I mean, I think that's you know, they're going to have, you know, at least an hour from our perspective. Looking at it from a tangent solutions perspective, they're going to have all of the benefits of agentic AI if they build it with us, yeah, that they would have internally. You know, we're going to give them the agents, we'll give them the workflow, we'll give them the documentation, we'll give them you know the knowledge transfer and all of that. And then they can they can either let us keep it up or they can do it themselves. You know, it's okay. But I think that I think that that's just as much of an advantage doing it, let's say, with a third party like us. But I think that you know, getting back to your question is like, I still think there's a big question mark in the market. I I get these um I get these uh newsletters from these you know AI um uh companies and people, you know, that talk about all these new solutions, right? Uh okay, they they highlight some new solutions, and I and I'm starting to see this repetition of solutions, you know, like whether it's a JIRA knockoff or whether it's you know some other you know uh financial services type of thing, payment processing. And you start to think, you know, there's just so much stuff, and you and you wonder, you know, are we just saturating ourselves with with stuff to the point where we say, okay, I'm I'm saying no because I'm confused, and a confused mind says no. You know, that's where I think you know, there's a there's a there's a problem and an opportunity, and I just can't see it very clearly. You know, I know that we we take things when we're talking about internally, we're starting to we're starting to work together with our own internal processes to to look at AI and how we can optimize workflows and email management and you know marketing spend and marketing performance and all that. And it's it's a job, you know, we're not a huge company, but we still got to look at people, processes, data, tools, and and figure all that out. And it's no different than when we were implementing SAP or anything else.

SPEAKER_00 35:33

Okay, is there a technology, Dennis, that you're super bullish on? And and the inverse, is there technology that you're very skeptical on?

SPEAKER_01 35:43

I'm very bullish on, you know, we're we're doing a lot of stuff. I'm seeing some really cool solutions that we're building in the AI ops and observability space. We're taking you know information from so many data sources, monitoring tools, uh other infrastructure uh tools, uh operating systems everywhere. And we're just ingesting a lot of information about uh the entire technology landscape of a company and able to condense that and distill that down into very actionable, predictive, analytical type of solutions based on AI ops, based on AI. I'm very bullish on that. So that we can see not only at the infrastructure layer or the data layer, but also the application layer. So application data and infrastructure. We're we're building these solutions quickly under AI ops and under observability to be able to get a much more comprehensive view of the snapshot of the uh technology landscape, but also the you know, general health checks and be able to see in a very predictive way trends that start to develop before, long before, in fact, that they they become problems. So very bullish on on that and some of the stuff that we've done, particularly on the cloud, mainly AWS and and Azure. Uh the other things, the the thing that I'm not so not so keen upon, I think, is I think it's just you know it's a perception, right? Um, of these, you know, building so much code, using AI just to be, you know, uh you know, getting over the FOMO, getting over the fear of missing out thing of, you know, people are hearing that you know, we can build so much code in so little time, we don't need these developers. I'm very skeptical about that whole the whole perception around uh the vanity metrics of of AI. You know, I'm not of AI itself. I just want to be clear about that. You know, I think there's a huge place for it, you know. But I think that you know, we're starting to see that we cannot sacrifice uh solid software engineering fundamentals and the foundation of software engineering, especially if we're talking about larger systems, enterprise-wide systems, just because you know somebody says, yeah, we can build, you know, we can build that code in you know three hours, something like that. That's that's where I'm very skeptical, and I think we see some indications of that.

SPEAKER_00 38:38

What are your top three most favorite AI tools at the moment?

SPEAKER_01 38:44

Oh, geez, you know, like I'm gonna talk about the team, what the team tells me, you know, because uh I'll tell you that a year ago, uh first half of 2025, you know, we had clients insisting that we use cursor, you know, because that was kind of their development tool of choice. And I haven't heard a word about cursor so far this year. You know, everything is Cloud Code, it's Opus. I don't know what's happening with my thoughts. I think you know, some of the stuff is getting you know held up by the government. If I I heard something recently about that, but you know, cloud code seems to be you know the the the platinum standard for everything. What I will say about that is that you know, we we have, for example, a couple of clients that you know, we say we're using uh quad code and they might be using a different tool. So my advice to those clients would be look, let's use one tool, you know, let's use one tool across, you know, like if you want us to use a particular tool because that's your industry standard, you know, we can debate it or we can show you, you know, what we're doing with Claude that might, you know, tip the scales more towards Claude, but we should really be looking at you know, standardizing on one tool. But Claude Code is definitely one of them. Um is the one, I would say. Um and then beyond that, I don't know, Maxi. They may be using things I never heard of. I don't know, but I know that one's the thing.

SPEAKER_00 40:24

Yeah, I know. I I I echo your comment. Uh I mean, as it is, as I said earlier in the conversation, I'm not a developer, I'm a CPA, I'm a finance capital guy, risk guy. Um, but I am on pretty much on a daily basis in Visual Studio and Cloud Code is is running and I'm doing things tinkering, or in my case, because I'm at the call phase, I understand the problem. I I'll I'll develop the plan and turn it over to the team and say, okay, go review the plan. Does it make sense? Does it not make sense? If if it does make sense, go and execute. Um, but my like my my most favorite base tool at the moment is whisper flow. Um because it translates uh voice into text, but but doesn't just do the literal translation, it just cleans it up. They call it polish, it polishes your speech into into um um into coherent text. And so I use it for everything. I mean, I'm in the top one percent of whisper flow users globally, like in in the run of three months. I think I've done like 450,000 words. I use it for coding, it's amazing for that. I use it for emails, I use it for prompting because it it cuts you off at about five minutes. But uh, I'll be walking the dog and it'll be like coding, or be like sending out emails or just just just talk to that. So by far, this is my favorite tool at the moment.

SPEAKER_01 41:55

Yeah, I heard about that the other day. In fact, I I queried Gemini the other day. Um okay. Just wanted to know. I said, look, you know, like how, you know, because I was I was interested to know how they were tagging YouTube videos, you know, how they were identifying, because there's been a lot of talk about YouTube videos that are just AI slot, you know, that type of thing. So I was querying Gemini the other day about, you know, hey, how does YouTube uh tag and categorize AI content? I think it went into this long explanation, and you know, one of the things that it said was it said something about it identifies, it uses something called synth ID, which I can't explain what it does because I don't remember, but but it was something like um it's able to identify at the point of creation. That's kind of the explanation explanation it gave me. And he used Whisperflow as an example, and I think there's some tie-in there, but um, that's where I first heard about Whisperflow. But uh it was interesting in the sense that I said, Well, I just want to understand, you know, my prompt went by, you know, I want to understand policies and processes as far as YouTube is concerned and how they tag and how they identify AI content so I can clean up my feed, you know, all whatever those. And they they said, and I'm particularly interested in how you do this at the point of creation, assuming that we're not using a Google tool to do the creation, right? Yeah, and uh Gemini defended itself by saying, Oh, we don't do any spying, you know. So it was kind of funny, you know. I was not, you know, I didn't I didn't I might have implied it, might have been implicit, but I didn't specifically say that hey, are you spying on us? Yeah, but it was interesting that it it reacted that way. I got that reaction from the LLM that that said that you wouldn't do it. I went into a long explanation about how this is done and clarifying it. But whisper flow was mentioned, and uh yeah, I understand that that that might be very, but it's very, very interesting. This whole speech to text and um. You know, video translation type software, that's that's also a game changer.

SPEAKER_00 44:05

For sure, for sure. I mean, we we I think we have subscriptions of three different uh video or audio tools. Um we love 11 Labs, that's what we use for our voice agents and systems. Um we like it better than Telegram bot and WhatsApp bot. Um we have subscriptions to Kaplan and uh different video tools. It it's amazing what you can do with technologies these days, and and again, what tomorrow looks like, I don't know. Um, because like I mean, you saw early in the year, open AI was the clear leader in AI in terms of the the large language model, and now Claude is pulling ahead. Now they have this path with the federal government with the Pentagon with the jailbreaks and security and Mitos. And I get this, but but it's still like it's the preferred tool. And this all happened within the span of what like six months? It it shifted from open AI being on top to allow anthropic being on top.

SPEAKER_01 45:08

So, what's gonna happen over the next six months? That's the question. Is it gonna flip-flop again? Is somebody else gonna come up and you know challenge uh anthropic? Who knows? You know, like like I said, I mean, we would we were using different tools a year ago than what we're using now. A year isn't that long of time, you know, and but now the time frames for these tools are becoming so much more compressed, and you know, it's a race, you know, that's that's just getting faster and faster, but I'm saying it creates so much you know uncertainty. You know, do you go all in on this? You know, do you say, okay, look, I'm gonna go all in on this, and then you know, a couple weeks, a month, two months later, something else new comes out, and that's so much faster or so much better, or so much more comprehensive, or you know, incorporates so much more engineering, foundational software engineering principles into the development, whatever it is, you know. And I think that's where, you know, I think that's where we we you know, where we're suffering a bit, you know. Like, what do we do? You know, do we go all in? And I think there's, you know, I think you do go all in. If you're gonna go on in, I mean, what are you gonna do? You're gonna wait. I don't think it makes any sense, you know. I mean, that's the other thing. If you're gonna go all in, go all in. Don't worry about what's coming next. It'll come or it doesn't come, you know. And and that's what that's the way we've taken it with Claude and just say, let's go on in this and just not worry about anything else. Let's just not spread ourselves too thin. Let's become really good at this. You know, let's come good good enough so that you know one of our guys actually teaches uh coding principles, you know, using Claude and AI assisted development uh as part of you know, offline from what he does with us. So yeah, I think there's you know, uh, I I think you just go in with with something, learn it. If something else comes up better, take it as it comes.

SPEAKER_00 47:08

Makes sense. And Dennis, um, couple of last questions for you. What's the best advice you have ever received?

SPEAKER_01 47:17

Damn. The best advice I ever received. You know, I think the best thing I ever heard, and I'm gonna bring this back into the entrepreneurial world, is uh many years ago, you know, there's a lot of uh the big question about why businesses fail, right? Um, I was listening to, I don't know if it was a podcast interview, don't remember what it was. This was 10, 12 years ago, and we were just pivoting the business at that time, and it was a real, real struggle, you know. And um, you know, we were wondering, you know, are we gonna survive this or what? You know, um, you know, I remember, you know, when you're not making, you know, when you're not covering your expenses, that's a tough world to live in, right? You know, and you're gonna come out of that somehow. And I remember somebody on uh on an interview said, look, the biggest reason why businesses fail is because people don't hang in long enough. I hadn't heard that before. And it came at a perfect time, it came at a time when I was really questioning whether, you know, I you know, should I stay, should I, should I, you know, hold this up, do something on whatever it is. And um, you know, he wanted to say that sort of like you know, if you have if you're struggling in something of getting clients, you know, hang in there long enough to figure it out, you know. If you're struggling with operational cost or you know, getting the right people or whatever, hang in there, you'll figure it out. You know, if you if you hang in there, if you find the right people, ask the right questions, get the right information, you can figure out anything. You know, so his point was you know, people give up too soon in the business. And that at the time, I think, was the best advice that I heard. It wasn't directed at me, but it was the best thing that I heard at the time as far as what was going on in my own business.

SPEAKER_00 49:14

Makes sense. Uh, last question for you what's the kindest thing anybody has done for you?

SPEAKER_01 49:21

The kindest thing that anybody's ever done to me, um, she's been a lot of kind things, actually, I have to say, you know. Um, if I pick one right now, I'm sure I'll think of a better one five minutes from now. Um, but I think the the kindest thing, uh I should have thought about this before, shouldn't I? I think that um I don't know, Maxine. I can't come up with one right now. Can I owe you that one? That one's a tough one. You should be able to come up with a hundred of them because I know that I've had a lot of kind things happen to me. I just can't think of one that stands out above the others.

SPEAKER_00 50:11

Yeah, you mentally start to to do search through your brain, through your brain, and then you mentally start to start to rank rank and prioritize. But which one is it? Like, yeah, no, I understand that the dilemma that comes with this.

SPEAKER_01 50:23

I just can't find, you know, but you know, the good thing is that you know, the lot of good things have happened to me, and and uh I feel very fortunate. I got to live overseas, you know, just by chance. Uh, you know, when I was 25 years old, you know, I got a lot of great opportunities that sort of came my way, you know, between living in Switzerland, living in Argentina, uh, you know, traveling the world a bit, you know, and there was a lot of good things and a lot of good people that that I had to run into, you know, to you know to make those experiences in a living experience for sure, you know. And um there's just you know, probably in the thousands, really. Um I feel very fortunate in that perspective.

SPEAKER_00 51:12

Yeah. Gratitude is um is an important skill that people need to uh to learn how to cultivate because it keeps you grounded, it keeps you happy, and keeps you focused on on what really matters.

SPEAKER_01 51:27

Yeah, it sure does. You know, we talked in our CEO group last week about that, um, how we can implement empathy into our companies, you know, with our people and and be empathetic to you know their situations, you know, because you know, you have an employee or whatever that you know acts strange for whatever reason, you know, out of character, you know, you don't know what's going on in their lives. You know, it's easy, you know, to say, hey, you know, take it easy or whatever, but yeah, you know, sometimes you just you know, when you find out, when you look you know underneath the covers of what's going on in people's lives, you know, um, you know, a dose of empathy goes a long way. Uh yeah, absolutely.

SPEAKER_00 52:09

Businesses at the end of the day are human.

SPEAKER_01 52:12

Uh you can have you don't at the end of the day, you don't do business with companies, you do business with people. Right? We're all human beings, we all have our our strengths, our weaknesses, our flaws, you know, whatever it is, you know. But you know, uh, I found that you know a small dose of empathy can go a long way.

SPEAKER_00 52:30

100%. That's um a great note to end this conversation on, Dennis. I really enjoyed um your thoughts and perspectives. Um, so thank you so much for coming on to Scaling with Clarity Show.

SPEAKER_01 52:44

My pleasure, Maxime. Enjoyed it. Thanks very much. My pleasure.