#414 Tech Due Diligence Mastery with Joe Conley: M&A, PropTech, and Gen AI Integration

[00:00:00]

Mehmet: Hello and welcome back to a new episode of the CTO Show with Mehmet. Today I'm very pleased joining me, John Conley, CTO of TechSight. Joe, the way I love to do it is I keep it to my guests to introduce themselves. So tell us a little bit more about [00:01:00] you, your background and what you're currently up to and then we can take it from there.

Joe: Okay, great. Yeah. First of all, thanks for having me on. Yeah, so my name is Joe Conley. I'm a CTO based in the greater Philadelphia region. I've been in software for over 16 years now, you know, started out kind of full stack full stack engineering focus on kind of data integration, scaling data platforms, things like that.

Joe: And the latter half of my career, I've been very interested in working with startups. So I've kind of naturally gravitated towards the CTO role. And so I've been CTO for a couple of different prop tech firms in the area. And now I'm doing kind of more for Fractional CTO work under my company tech site.

Joe: You know, working for some private M& A companies, doing some M& A due diligence, working for a few different startups at the various phases, whether it's pre revenue kind of scaling. So, you know, I, I think I've really embraced and liked that challenge of being able to, you Spend one day doing something that's, you know, M and a tech due diligence, spend the next day helping a company grow from literally zero to one.

Joe: And then, you know, another day helping a company that's already had product market fit and already kind [00:02:00] of shown the growth and kind of trying to figure out, okay, how can I help these guys really scale and really start. Going from walking to running. So I really enjoy that challenge of kind of, you know, every day is different challenge every day is kind of a different set of set of problems and opportunities.

Joe: And, you know, so I've really kind of embraced that. And I think that's kind of what my personality is. I kind of like to, to, to solve different problems. I, you know, I, I don't like to be kind of stuck doing one thing and that's, that's kind of how I got into startups. You know, anyone who does startups notes, you know, you're always wearing a lot of different hats.

Joe: And so I've spent my time there in startup land.

Mehmet: Yeah, absolutely. It's, it's a, it's a nice place to be now. You know, while preparing for, for the episode today, Joe, like I've seen, like, you have really a very rich and mix also blend of verticals that you have worked with, like from telecom, healthcare, logistics.

Mehmet: So But when it comes to be a technology leader and be a CTO, obviously, whether in a full fledged capacity or like as a fractional what you can tell us like that some [00:03:00] common thread, I would say in building, you know, robust Systems that can serve the aim of that startup or if even if it's a full fledged, you know running business scale scale up business.

Mehmet: So what is like the common? you know, I would say Threat you saw in building these efficient systems and you know, making sure that everything is working perfectly across these industries

Joe: Well, the first thing that comes to mind is really having a good understanding of the business in the market that you're in, you know, certainly between prop tech or telco or health care, you know, there's very specific dynamics that sort of govern these different marketplaces and these different markets and companies and, you know, that can be very idiosyncratic, it can be very, you know, taking prop tech, for example, that can be a very hyper local domain where even you know, County by county things, people do things different ways and they format data different ways.

Joe: So I think having a [00:04:00] grounding in that, I think that's probably a good sort of general point for anyone who wants to be a CTO is to really. You know, I found the best CTOs tend to kind of have come up as engineers, but they also have developed just a strong business sense, like, okay, you know, if I'm going to build this system, what is the business value of the system?

Joe: Is it going to provide an ROI for the business, you know, or is it a science fair project? You know, I've, I've certainly developed a. Sort of a distaste for kind of spending a month and doing something and have it be completely disconnected from the business. Like there has to be some, some connection to the actual the actual domain that you're in and.

Joe: And again, I think that's, that's sort of why I guess I got involved in and gravitated towards startups because I do find the business aspect interesting and it does dictate a lot of what you can do. And, and, you know, at the end of the day, especially if you're a startup, if you're not, if you're not keeping that, the business priorities front and center, then at some point you're going to run out of runway, you're going to, you know, you're, you're going to run, you know, you're going to burn through through your capital reserve.

Joe: So, so, so that, that, that to [00:05:00] me has been the most interesting, I think challenge for me is. You know, I've certainly spent the first half of my career just focused purely on the technical aspect of things and how to build systems and how to build systems and scale. But then, then kind of adding that extra dimension of like, okay, you can build a system and scale like, okay, well, how is it going to be?

Joe: Tailored to your target market, you know, you know, especially for startups that maybe aren't going to, you know, explode in the next 12 months that need to kind of scale gracefully, it's like, okay, I need to, I know how to build a system, but how can I build it in a way that aligns with that kind of 12 to 18 month timeline and how can I scale gracefully over the period and, but again, still be reactive to the fact that, okay, we could see upticks in volume and transactions and things like that.

Mehmet: Absolutely. Fantastic. Now you, you know, mentioned also something introduction about M& A and, you know, like this also, like there's a, the CTOI when it comes to, to doing due diligences, right? So, I like to start with the red flags, right? So what are like some of the red flags that you [00:06:00] always try to spot whenever you approach a new technical due diligence in an M& A deal?

Joe: So, in terms of red flags, and for most of the engagements I've done, it's typically involved sort of an interview with sort of the, the target, the targets like head of technology, whether, whether it's a CTO or just a, you know, a founding engineer type approach. And so I typically, and again, it's, it's a, it's, it's a, it's a tough needle of thread because, you know, you, you want to find out what's true.

Joe: You want to find out what the system, you know, how the system was built, but you want to do it in a way that doesn't really over critique them. Cause again, you know, anyone who's built a startup knows that, you know, nothing's perfect, you know, you're going to take, you're going to incur tech debt in certain areas.

Joe: And, you know, certainly, you know, you can get defensive if five, five years later, you're trying to sell the software and you're trying to put a nice kind of red ribbon on top of it. And You know, you have someone like me come in and kind of grow and you're like, okay, well, you chose this stack. Why'd you choose a stack?

Joe: Why don't you choose a, B or C that sort of thing. [00:07:00] So, where I see red flags and again, going beyond kind of the basic sort of, cause there's tools you can use to help automate a lot of this process. But, but I think where I had the most value is having a human conversation with, with the target and I'm really understanding.

Joe: Why they did something. And certainly if they get to a point where they're super defensive or they're, or if they don't have a strong justification for what they've done in terms of an architectural decision, that usually means, okay, I need to dig a lot deeper on that area. Cause there's probably some, there's probably something there that, you know, could bite us if we, if we acquire this company.

Joe: And then I'll also, if they're also, if they're very reluctant to talk about a certain aspect, that's probably common, you know, if I'm asking very simple questions about like, where do you store your data or things like that I And I get very kind of one word answers or very evasive sort of responses.

Joe: Then, you know, that sort of triggers my curiosity, like, okay, so, you know, not what I expected to hear. And then I'll try to dig deeper either explicitly. With that resource or just on my own kind of due diligence as well. So, so yeah, and [00:08:00] again, it's, like I said, it's a tricky process where, you know, you don't, you have to really have to thread a needle there in terms of, you know, trying not to be too insulting, but at the same time, it's my job as a due diligence expert to really understand the truth and then understand what we're getting into and understanding.

Joe: If we acquire and operate this thing, is it going to blow up in our face in six months? That sort of thing

Mehmet: Right. So joe one also You know, i'm not sure you you are the expert here and that's why i'm asking this question Let's say, you know, there are no red flags, right? But let's say i'm i'm i'm the one who gonna acquire this new startup, right?

Mehmet: They built fantastic technology They they they have everything, you know as it should be by book, but let's say You There are like differences between the technologies that we have built in house, let's say, or we have acquired over the years and what these guys they have built, or maybe, [00:09:00] you know, they built something fantastic.

Mehmet: Everything is again done very, very good, but it's not designed to be serving like large scale customers. So in these situations, like what Have you seen the best approach to solve like I just came up with these two scenarios, which I heard from people They they're kind of common sometimes. So what's your point of view on on such scenarios?

Joe: Okay, so taking the first scenario, which I guess was the integration challenge that's certainly a big aspect of you know, sort of like the pre acquisition analysis that I do and really it's it's sort of dependent on the acquirer what their kind of overall philosophy is in terms of Do we want to centralize all those functions, all the engineering functions, or do you want to keep things more decentralized and maybe we're just acquiring this.

Joe: This product for say cashflow or things like that. We're not trying to, maybe we're not trying to achieve economies of scale. We just want to kind of acquire this product and maybe fold it in with others that are in a similar industry, but not necessarily [00:10:00] do it, do a pure integration, just kind of keep the product as is.

Joe: I certainly prefer the latter. Like I'm a huge fan of constellation software and an investor in them. And They are very much in the latter camp of just kind of keeping things very decentralized and they have very little in, in, in terms of kind of central functions and sent and enforcing folks to use kind of central services.

Joe: They're very much take this decentralized approach where they acquire a company, you know, that's very much sort of acquire and operate. And, and again, they, they do fold things under brands, but that's more kind of a. More of a market type of approach that, you know, they don't do a ton of sort of central integration.

Joe: So, you know, software is hard. So, so I certainly for the clients I've worked with, I've certainly encourage more of the latter path in terms of. You know, having acquiring a stable product and, you know, more often than not trying to achieve economies of scale that have any really significant long term value can be, can be really challenging and there's always going to be challenges trying to, trying to, you know, refactor, you know, entire products, just, [00:11:00] just to put them on your stack, you know, you know, it has to be a very special, special situation for me in terms of.

Joe: You know, is there really a lot of value? Are we really at the scale where integrating all of these products that we're acquiring really helps us achieve economies of scale. And again, there are exceptions. So certainly if you're, if you're a massive fund and, you know, let's say you have, you know, big contracts with maybe some of the bigger cloud providers and you just can realize a lot of discounts and savings from there, then probably at that scale, yes, then maybe integration makes sense.

Joe: But certainly if you're maybe more of a middle, a small market player, where you just want to kind of build a portfolio of maybe, you know, six to 10 different companies and maybe one or two different verticals, then, you know, it's, it's just so much, it's such a headache to try to integrate those folks to, to use one stack.

Joe: And, you know, I'd say most of the time it's, it's not worth the, the. The technical cost and then just the manpower to get there and then just to make sure that nothing breaks is you know That would certainly kind of keep me up at night So, so that was the first scenario I think integration and the second I think you mentioned was kind of dealing with scale [00:12:00] Honestly that the biggest the biggest way to really attack that is really to make sure that we're acquiring a company That can be set up for scale Most of the acquisitions I've seen tend to be on a cloud provider.

Joe: So I haven't really seen many recently where that we've had to do something where we're on some custom custom metal or some custom on prem kind of deployment. It does seem like most of the folks have gotten a memo that cloud is the way to go. So that's been very encouraging and certainly made my life easier.

Joe: But I mean, what's nice is I think most folks do tend to use kind of the big three in terms of cloud providers. Now, I tended to specialize in AWS, but I've worked with others as well. You know, there's definitely things that you can do to kind of help in terms of scale. And again, AWS provides some nice migration tools and others like there's the AWS DMS, like a data migration service that kind of makes that migration process easier.

Joe: So maybe if you want to take data that's in, say, maybe a non AWS database and move it to yours, that's great. So, so really I think the challenge is kind of, one, finding targets that are already cloud compatible. And if not, maybe find, you know, and [00:13:00] again, if it's still an acquisition that you want to make, maybe finding someone like myself or someone who's, Who can specialize in getting them to be cloud compatible.

Joe: So it's almost to my mind, it's more, it's more kind of preparing the target company to be in a certain state that you can have it in the cloud. Cause once you have it in something like AWS, you know, AWS has hundreds, like literally hundreds of services that can really make this process easier between migration and, and scaling up data and defining, you know, defining scaling policies and things like that.

Joe: So it's almost more of like, once you can get it in the ecosystem and get it to be Effective in that ecosystem. Then that's kind of the biggest battle from there. And honestly, beyond that, if you're looking at acquiring something that is super complex and would be a huge effort to rewrite, to get the scale, then from, you know, from my experience, a lot of times we just, that's a pass.

Joe: It's like, okay, you know, the juice isn't worth the squeeze there. You know, it's going to take us all this effort just to get it to scale. And maybe there's other levels of risk at different areas, whether it's at the product or the market area. And we just kind of do a pass there.

Mehmet: Okay, great. [00:14:00] Now, just quick question.

Mehmet: Is there any vertical where the technical due diligence is harder than other ones?

Joe: I would say so. Certainly, I haven't really done much done much in verticals that are like super hard. You know, really deep in terms of like a niche, say in technology, like maybe a super like hardware niche, for example. But I mean, certainly what I've seen, you know, I've done a couple now in like the real estate sector.

Joe: And I wouldn't say it's hard. Technically it's just the, the, the biggest problem in real estate is data is just everywhere. There's providers on the brokerage level. There's providers like the Zillows of the world, you know, they provide things like metrics data. There's you know, the MLS is the multiple listing services.

Joe: They, there's, there's roughly 500 plus of those across the country. And while there is an effort to standardize their data, you know, their data can still be pretty disjointed. So, so the challenge there is more of a trying to get all the data to speak the same language, to be in the same [00:15:00] format and to be processed in real time or close to real time and understanding how that works, understanding all the business logic that happens.

Joe: And again, for certain. Real estate companies who are trying to collect all this data, trying to understand how they process it and making sure they're doing it in an effective way. But certainly, yeah, I wouldn't describe that as really a deep technical challenge. It's more, just more like herding cats.

Joe: Like, you know, we have all these different processes and making sure we have full line of sight into how those processes are designed and, you know, how they're designed from a, Failover perspective, how they handle things like if an API goes down you know, that sort of thing. So it's more, that's where more of the due diligence sets in is just more of just making sure that from an operational perspective, we have a good understanding of, you know, is this API well covered?

Joe: How do we handle failure? How do we handle downtime? You know, are we, you know, things like that.

Mehmet: Yeah, it makes sense. Absolutely joe now because you mentioned prop tech couple of times like And prop tech is I I think like something I didn't cover it [00:16:00] much, you know on on the show You know, maybe one time last year but you know, because I'm based in Dubai.

Mehmet: So anything related to properties is very, and PropTech is, is, is on the rise, I would say. So there are a couple of startups. So what trends, you know, have, you know, you seen like really they are transforming the real estate industry these days?

Joe: Yeah, it's a great question is certainly I think data standardization is one.

Joe: I sort of already talked about that just in terms of really trying to find ways for folks to develop a data standard. And there's a there's an organization called Rizzo who's built and developed this standard over time. And, you know, I found their work extremely helpful just in terms of managing data integrations from from a bunch of different sources.

Joe: Certainly seeing a lot of Jenny as well, just in terms of you know, with my first real estate startup, we had a very kind of basic machine learning process to, to rank seller leads. So it was more than residential real estate space where, you know, if you're a, if you're a real estate agent, you want to develop kind of a whether it's in [00:17:00] your CRM, wherever it is, just a list of kind of seller leads that you can go through.

Joe: And we developed some, some technology there to help rank those several leads based on all the data that we had access to. But now you're seeing, and again, that was more of a traditional machine learning process, but nowadays you're seeing a lot of those functions being kind of taken over by Gen AI in terms of just basic recommendations in terms of financial analysis and financial health, you know, taking a look at potential home buyers and understanding their profile and understand, you know, what they can afford in the current market and things like that.

Joe: And so, so I'm definitely starting to see more and more usage of that. You know, having a, a Jenny, I co pilot help walk you through the home buying process understanding what your needs are, where you're looking and helping to improve that. But again, just like, just like in all industries, it's sort of like a, you know, there's a popular term now called the jagged frontier where, you know, Jenny, I can make a lot of progress in one specific direction.

Joe: And then maybe in other areas, it's not as helpful. So, you know, we're still in this age right now where, especially with Jenny, I it's very much in it. experimentation age where we just want to find out what we can [00:18:00] do, what works, what doesn't. And, you know, as these models get better and better, and certainly the focus recently with these models is more on reasoning.

Joe: I do think that we'll start to see better and better results just in terms of being able to meet those kind of core business functions in PropTech and elsewhere.

Mehmet: Right. You mentioned, you know, Gen AI and, you know, I'm sure, you know, Like myself, you've seen many tech waves, you know, over, over the years.

Mehmet: And you are mentioning like, it's still kind of, it's in, it's infancy, right? From, from Gen I perspective. But if you want to compare within this short time, you know, the, of course there is a lot of doom and gloom. Everyone knows about it. But if you want to compare, for example, the previous waves off of these emerging technologies with what Jenny I have done so far.

Mehmet: How do you see, you know, the impact on transforming the technology, including not only I mean, for example, From the sector's perspective, we're just talking about, for example, m and a. [00:19:00] So for example, utilizing even gen AI for MM and A per and so on. Of course, I had a, a fantastic episode with, with mad Van Italy I think couple of months ago, and he, he needs a company that helps actually in, you know, discovering the code and you know, if there's something wrong with the code.

Mehmet: But from your perspective, Joe. What are you seeing, you know, the most promising aspects of utilizing Gen AI in general, not in a specific sector?

Joe: Sure. So just in terms of technology waves, I've really only seen up front, probably the more recent one before this, which was probably, you know, the move to the cloud and cloud infrastructure.

Joe: So when I first started, we, you know, we had kind of the bare metal servers in the back room. And we, you know, if we wanted to, if we wanted to increase the scale of our platform, then we had to go buy another server and get that set up and things like that. So. I'd say cloud was the biggest wave that I've seen and actually been a part of kind of firsthand, just kind of seeing the impact of it, and it certainly feels like Jenny eyes.

Joe: This is a similar level of impact where [00:20:00] it literally changes the way we work. It literally makes us more efficient. It does drive greater productivity. And really, the biggest way I'm seeing that is. Is sort of having AI enhance existing engineers, enhance their work and make them more productive. So you talked about M& A.

Joe: I've been doing M& A, M& A kind of advisory for almost two years now. And probably for the last nine months, I've been much more active in using Gen AI for helping me look at code, for helping me sort of analyze a market. So if I'm looking at a brand new product, brand new market, I'm completely coming in cold, having just talking to AI for 10 minutes and kind of doing iterative.

Joe: a lot of iterative work with it, not just relying wholly on it, because I think anyone who's done enough of work with Genii knows that it can't just be a one way street where you just rely on it for everything. It's got to be very iterative where you kind of do a piece of work, then Try AI for a bit, have it maybe even do the work in parallel, maybe compare and kind of take the best of both.

Joe: But certainly [00:21:00] in M& A can really help tackle a huge code base. You know, there's great tools out there like cursor, where you can basically have your code base on your local machine and have this great tool like cursor. Be able to analyze either specific chunks of the code or the entire code base and ask very basic questions and then drill in to very specific things as needed.

Joe: So that's, that's been a huge help. That definitely has saved me a ton of time compared to more manual ways of looking at code. So that's just even on the M& A side. So even just for basic engineering productivity, it's definitely at least a two to three X multiplier. I think. Again, using tools like Cursor and things like that, where you know, the biggest value to a tool like Cursor is that it's so ingrained in the context of the way you work.

Joe: And I think that's where, I think that's where we still have the biggest opportunity in terms of leveraging Gen AI is to be much more integrated into, The current ways of work and have it be much more seamless. I know there's tools out there like whisper where you can basically, as you're doing work, you can basically just talk out loud and it can record [00:22:00] things and, and kind of almost be like, almost like a, like a, like a coach that basically like a pair of programmer in a way that's kind of sitting right next to you.

Joe: But curse is another great example where you don't even have to leave your code base. Like you can literally be looking at your code, either type a question in line for the code to get some code changes or use a chat window. That's in the same. Kind of context as well. So that's where I see the biggest value.

Joe: You know, certainly, I think the ability to chat is nice. It's simple and easy. And again, one of the greatest things about Jenny eyes is how great it does with natural language. Like it can literally can type any question. I can type in typos, you know, things like that. And it gets still very much because it's using That very powerful kind of statistical statistical inference, even even though I do typos or different languages or whatever, it can still kind of parse that out and understand what I'm asking and at least try to predict a reasonable response.

Joe: And again, it's not perfect, but but I definitely see a ton of value for engineers to. To help them do their work. But again, like you said, it's, yes, I've definitely seen situations where [00:23:00] folks just aren't really embracing it. You know, they either are kind of doom and gloom and they just think it's coming to take their jobs or they've tried it once and got a bad outcome and just stop trying it again.

Joe: So that's where anyone I talked to, I definitely encourage them to, to, to be patient, to, to really, to try it in a couple of different use cases. And I think if they put enough effort and have enough of an open mind, they're going I think they will see the benefits of it and they will see how, how can really kind of turbocharge their own work.

Mehmet: Yeah. I always, you know, it's not always like recently I start to tell people like, I'm not sure if it's, you know, it's a coincidence of course. So when the search engines start to appear and of course like Google was leading the way. So that was before, you know, the social media, the Instagrams and the TikToks and where everyone was trying to show like, Hey, like I know, you know, like this technology.

Mehmet: So Genentech AI came up. When this wave where everyone is a, and I'm not against that. So there are great content out there and, you know, people started to have these different [00:24:00] ideas and criteria and about you know, how the best way to use the gen ai. Gen AI is good for that. But for this, but to your point, Joe, and this is something I experimented personally.

Mehmet: So not only like, of course not relying a hundred percent on the, on the Gen ai, of course, like. acting as a copilot as Microsoft, they call it. But another thing which I started recently to, you know, really get benefit out of it is using at least two gen A. I. Tools for getting better results. So mainly I use chat GPT from OpenAI and I use Cloud from Anthropic and the results like really are mind blowing because you know, I think the output from there, I put it in the other tool, vice versa, by the way, it's not like one is better than the other.

Mehmet: And then, you know, I get really a very, very, very refined result. I say, wow, like I couldn't even think about it. And I think both with the, you know, they start to include this reasoning. Within, [00:25:00] within the model. So it's really fascinating. Now we talked about, you know, like on, on, you know, from, from technology perspective, the value, if I want, and of course, like as, as a, you know, CTO, you know, whatever you're working, like a, like a You know, on the organization, which is larger organization or as a startup.

Mehmet: So the customer is in mind, right? So from your perspective, what's like the, in your opinion, the main value proposition to customers in industries you worked in or you're currently maybe supporting for, for the value that really the customer worry about

Joe: in terms of gen AI or in terms of just, yes, gen

Mehmet: AI.

Joe: Okay. Yeah, certainly I think the value, and again, I guess what I've seen the most is just kind of general productivity gains, so certainly when I, when I'm talking to CEOs and founders of startups who are just kind of dipping their toes into the pool of gen AI, you know, they certainly, the, the biggest hasn't, you know, I don't think cost hasn't been as much of an issue [00:26:00] recently.

Joe: I think some initial conversation I had, there was definitely concerned of like this cost way too much and it's wrong half the time. So why would I bother? So I haven't seen cost being as much of a barrier. To entry for these folks. Recently I think, I think more of the concerns still tend to be around data privacy worrying about giving these companies acts potentially access to their data and the, and again, there's, there's solutions for that.

Joe: I think that are becoming more and more popular. We don't have to give a data per se, and what's great with open source models, like, like, like, Metaslama And other great open source models where you don't necessarily have to send data off to a third party. So, so again, I don't think data privacy is becoming as much of a concern as it used to be.

Joe: But so, so I think what I've seen lately is that folks have definitely seen the value of having a chat GBT like, or clog like chat bot embedded in their products. They've, they've seen kind of the base, like, okay, You know, having, just having that basic kind of interface [00:27:00] and having, having it be more dynamic than say, sort of a more traditional deterministic chatbot.

Joe: I think they see the value in that, but I think there's sort of that, they're at kind of like a plateau right now where they're like, okay, we see a chatbot. It works most of the time. That's great. Kind of asking what's next, like, like, what's, what is really the longterm goal here? Cause, cause again, chatbots are nice, but that's not, that's not really the promise I think of, of this revolution that, that is And the hype that we're starting to see.

Joe: So, so, so I think that's, it's a very good question that we're trying to answer right now is what is the long term direction of these GNI tools? For me, I think, you know, companies like Palantir, I think are doing a great job of showing us what that vision looks like. And, and again, I think it's much more grounded in sort of operational intelligence and understanding how your enterprise functions and being able to use.

Joe: Sort of in very kind of almost surgical ways, being able to use Jenny, I very specific choke points of your kind of normal enterprise process to help enhance it, or even just do it, you know, a hundred percent [00:28:00] autonomously. So I think that's kind of where the biggest value is heading is in terms of having very specific agents with very specific jobs.

Joe: Being fed kind of private enterprise level data and, and being fed like, okay, here's, you know, or being trained on or fine tuned on how, you know, an enterprise usually does things or how they execute workflows and giving the agent enough interface. To be able to do most, if not all of those kind of typical workflows.

Joe: So again, it kind of does circle back to that idea of kind of enabling higher productivity. And again, and I'm still optimistic and maybe I'm too optimistic in this sense, but I'm still optimistic that this is still a tool to, to enable the current workforce to be more productive. I don't see it replacing folks per se, but again, like.

Joe: With any sort of technology revolution in any sort of wave like this, I'm sure there's gonna be some displacement somewhere along along the stack. But I, I do think with today's tools and, I mean, you can learn anything today. I mean, literally just go on a website and you can [00:29:00] be a, you know, with enough discipline and enough structured courses, you can learn engineering in no time.

Joe: So, I'm certainly encouraged by the way we're set up today, where even if you do have some level of displacement, I think you can find, you Ways to reskill and potentially join forces with Jenny. I'm not, not, not think of it as a, as a, you know, as an antagonistic force, but more joint forces with it.

Mehmet: Yeah, absolutely.

Mehmet: And I'm optimistic like yourself, Joe, and I believe. What people needs to do is like, okay, to accept that. It's here to stay. It, as we were mentioning, and you mentioned also as well, it's still early stages, but it proved that it can do something, right? So it can solve some problems and you mentioned about the agents and the specialized agents and, you know, like, I'm happy to see, because when I first saw the early tries of these open source agents that communicate with each other.

Mehmet: I guess that, yeah, [00:30:00] like this is the direction. And now I'm seeing, you know, the, the big players in the space saying, yeah, like we're going to develop these agents. And, you know, like even to your point, People can do it themselves because you can train, you know, on your own data, you know, the, the open source LLM on your own finances and you train it on your own, like how you hire people and you train it on your, to do this.

Mehmet: And then you let them communicate with each other. So we have now, of course, still human supervision is required. And this is why I tell people because you are there to actually, you know, you still kind of the. The project manager, the lead, whatever you want to call that position. And you give this task and you know, like this is, I believe, you know, technology and not only technology, like this is how humanity have been always.

Mehmet: We have something new that replaced the old, of course there will be sounds, Oh no, this is destroying jobs and this is gonna cause poverty and it's going to cause cows. Yeah, of course we, we will, we will have these people all the time, but [00:31:00] yeah, so we have to, we have to live with it now. One, one question, you know, before I jump to something else related to gen AI, when you talk gen AI, so we're talking a lot about innovation, right?

Mehmet: So, and you just mentioned about the companies that they just, you know, put into sort of a chat bot and so on. And I know like you have You know, your perspective that you advocate for, which is simple, elegant solutions. So how to not, you know, exaggerate the use of AI at the same time and keep the things simple rather than over complicating by just putting gen AI for the sake of putting gen AI.

Joe: Right. And, and that's a great point because I certainly think a lot of folks, and I'm certainly guilty of this as well, certainly have that sort of shiny object syndrome where they see this brand new toy and they want Use it everywhere. And, you know, there's that old saying, you know, if if all you have is a hammer, then everything looks like a nail.

Joe: And so certainly there's a [00:32:00] temptation by, by, you know, product owners and founders who just want to just use it everywhere, everywhere they can and be able to talk about it and put it on their investor decks and things like that, but I think that's where it goes back to having it, having a real grounding in the, in the.

Joe: Space that you're in and, and, you know, I'm lucky. A lot of the founders I've worked with have kind of been in their own industry niche for like 20 plus years. And they just know like the back of their hand, they know the problems you're trying to solve. They've been through it. They've been through these, these battles and trying to solve these problems.

Joe: And they now have kind of. Had this epiphany or had this idea of how to, how to make these problems either go away or make them a lot easier to deal with. And so, and so I think I've just been lucky enough to work with folks like that, who are just so grounded in the business that they're in and they just, they know they want to solve a specific problem.

Joe: And these folks just tend to, you know, they're open minded to Gen AI, but they're not just going to throw a chat bot on there and just hope it solves a problem. Like they're very much grounded in like, okay, like what is the business we're in? Like, let's stay focused on what. What is the [00:33:00] value we're driving?

Joe: And in a lot of cases, yeah, chatbot is more of a nice to have. It's not really a, we're not really solving that job to be done type approach and understanding what the true pain point is of users. Cause. You know, you know, take data analytics, for example, you know, every data analytics platform, you know, and you've probably seen a lot, you probably know of a lot, you know, they basically take their, they take your data, they, they reformat the way you want to give you a nice visual, whether it's a nice dashboard or whether it's a nice export into CSV or Excel We could certainly build a chatbot that could talk to that data and maybe just tell you some basic facts about the data, right?

Joe: But a lot of times that might not be useful. You know, a lot of times that what the chatbot could say is probably already up in the dashboard. It's probably, or it's very kind of surface level insights that aren't really going to tell you much. So we could spend You know, six months or whatever it is to build a chatbot that generates SQL queries and kind of does what the dashboard already does.

Joe: But at the end of the day, is that really going to drive ROI and business value? You [00:34:00] know, there's a chance it might not. So that, that's where it really is important that if we're, if we're building something with Gen AI or whatever it is, Gen AI or anything, we're really super clear about what, what value, like, what are we doing different that hasn't been done?

Joe: Like where, like the true insights going to come, where's the true Business value going to come from, you know, cause we don't want to go down a rabbit hole for a few months and build some chat bot and then find out, Oh, it, it, you know, users aren't even using it. They don't really even care about it. So I I'm definitely very skeptical of, of situations like that.

Joe: And that's why I certainly preach to the folks I work with, whether it's engineering teams or more of the business folks, to really have rapid iteration, especially for projects like Gen AI. I think we can very quickly build a prototype, but I think it really is important to get it in the hands of users and really just get their feedback and Would you actually use this thing?

Joe: If it said something different, would you use it? And, and, and that's, that's important across all software, not just gen AI, but to really get that customer feedback, have them be like almost like a first class citizen, almost like [00:35:00] they're sitting right next to you while you're writing the code and the customer saying, Oh no, I wouldn't use that.

Joe: Or no, that's not useful. Or, Oh, Hey, that's really cool. I wish I had something like that. It's, it's super important to have that customer kind of be front and center.

Mehmet: So basically we need to know, you know, the business benefits and I think this doesn't apply only to Gen AI. This applies.

Joe: Yeah. Yeah. I mean, this is, that's part of just writing great software is really just having that, that customer perspective front and center.

Joe: And cause it's really easy to get lost in the weeds of writing software. You can build up all these sprints in JIRA and, and build up backlogs and things like that. Yeah. Especially if you're an engineer and that's all you see, then you really kind of lose that really important business context of why we're building what we're building, you know, what it is that's most important.

Joe: So it's really important to try to do things that keep that front and center. So I certainly like to do things like have really strong visuals. I tend to be more visual learner. So I tend to gravitate towards, you know, drawing some nice diagrams, but also. Also, also kind of mapping out the user flow and just having that kind of have a nice picture of [00:36:00] that.

Joe: So, so everyone who's touching the product has a good mental model of, of how it's going to be used and where the value is coming from.

Mehmet: Right. And I think, you know,

Joe: I,

Mehmet: I imagine, you know, it's not, again, not only Gen AI, so when, when we do, when we do such Let's call them initiatives, whatever, to try to just, you know, enrich you know, the tech stack to give better outcomes.

Mehmet: So I think we are adding cost also as well and and maybe sometimes we need to do kind of a pivot, right? And this where we will have the challenges of, you know, moving, you know, capital instead of building this. Now we are going and building that, right? So, do you agree with me on that, Joe? Like we have, you know, a challenge in allocating, you know, capital around which tech stack we should currently be, you know, Putting so we move, you know, the, the things forward or like, you know, it's, it's like just a problem that always [00:37:00] have been existing.

Joe: Yeah. I mean, especially I've done a lot of work with startups and if I had to say one thing that was my superpower, I think it'd be building systems, you know, with that are very cost efficient, we'll say, I mean, I think I'm very frugal by nature, just, just how I am, I guess. And so I'm, I'm very, I'm very conscious of.

Joe: Anytime we're building any sort of system. I'm very conscious of how to build it in a way that we can build the first version in a very specific way that doesn't harm the future version, but it's still very cost effective. But yeah, so, and again, that's. That's probably another reason why I started gravitated towards startups is just that, that, cause that's just part of any other business, any businesses being mindful of costs and understanding, understanding how those dollars are spent and, and trying to make a dollar spend, spend for free.

Joe: So I've always enjoyed that challenge. So like, I, I see having that constraint as, as almost a, making it more fun. Like, it's like, okay, like if you had, if you had millions of dollars to play with, like, yeah, sure you could build whatever you want, but like, having like a, like a, having a set [00:38:00] runway that you have to operate under, it's I think it's a much more interesting and fun challenge to kind of design through because, because again, it's not just designing a technical piece.

Joe: It's okay, how can I design it so it's still delivering business value? And by the way, let's make sure we stay in budget because our runway is nine months or something.

Mehmet: Right. You know, as we are almost like you know, coming to the end, so I got to try to combine these two questions into the one. So for folks who want to transition from technical role to an executive leadership role, such as CTO, whether as a, again, full time or like fractional what kind of advice you can give them?

Mehmet: And this is where I can emerge it with the other questions. And, you know, from your experience, what is the, I would call it the leadership philosophy that you have always used and you believe it was one of the main reasons for your success.

Joe: So for the first point in terms of folks who are interesting in transitioning to more executive leadership I'd say first and [00:39:00] foremost, just talk to folks who have done it.

Joe: You'll probably hear 10 different stories, but I think it is important to understand what folks have gone through and understanding the reasons for, for why they wanted to transition. I will say if you're doing it just for hoping to make more money or hoping The status of having that cool C level title next to your name.

Joe: I don't think that's a sustainable, sustainable reason for doing it. And I certainly fell into that trap as well. You know, the first time I was able to put that title. And my LinkedIn, I thought it was cool. And then over the years, like, yeah, it's just a title though. Like, it's just, at the end of the day, it's like, you know, it's, it's more about the work you're doing and the value that you can bring to society and, and things like that.

Joe: And, you know, titles or titles don't really mean, mean a ton. So I think first and foremost, talking to folks, understanding what they've gone through and understanding what, like, like really, like, what is your, What is your reasoning for doing it? Cause not everyone, I don't think everyone's wired to, to, to be at that level or, or, you know, I think, I think most folks, I think some folks might be happy just, you know, writing code and just [00:40:00] staying focused, just purely in the technical aspect of things and not have to worry about, can we pay the bills next month or have to worry about market dynamics or, you know.

Joe: You know, being on quarterly investor calls and speaking to the market and things like that. So, I think that's first and foremost, understanding yourself and understanding your own personality and your own ways of work, and, you know, it, is it a good fit and having a really honest conversation with yourself and ideally getting one or two mentors, you know, I've certainly been lucky to have a couple of mentors in my life.

Joe: And they, and they, they provided insights that I still reflect on to this day, even just kind of just a simple sentence here or there that they, they just kind of mentioned offhand. I still think about and still kind of, still kind of helps me and kind of helps, helps my day to day and helps me kind of keep perspective.

Joe: So, yeah, so I definitely think introspection is very important and just understanding why you're doing things and just, you know, and if possible, just see if you can try it out, see if you can, you know, If there's some way that you can maybe do a quick side gig or something, if your life can, if your [00:41:00] current life situation can allow that, you know, just try it out for 90 days on a side gig and see if it makes sense, you know, so anything you can do to kind of, to, to test your theory about, do I want to move up to this level and validate that or, or invalidate it?

Joe: I think that's important. So that's in terms of transition in terms of, I guess, my own leadership style. I think, I think really having. Simple, effective systems, whether it's technical or whether it's communication is, is, is probably the most important thing to me. And maybe it's just the way I'm wired.

Joe: Like if things start to look too complex or feel too complex, I immediately just sort of take a step back and say, no, this is, this is, this is ridiculous. We have to make things a lot simpler. So I'm almost like allergic to things that are just too, like overly complex or just have an excess of things. So that's probably why I ended up being a good engineer because I think one of the things I've done well is.

Joe: Take existing code that maybe was a hundred lines and maybe turn it into like 10. So I've always been drawn to technology and frameworks like Scala is a good [00:42:00] one, or Python, for example, where you can, you can really have a minimal amount of code and be able to express things in like the most simplest manner and still work.

Joe: And I think that's sort of translated to how I sort of manage teams and manage communication as well. You know, I don't like to have, I don't like to have a ton of, you know, burdensome systems where we have to have daily check ins and you have to, you know, do a bunch of things that are not related to the core job that we hired you to do.

Joe: So I prefer to have the engineers focused on doing engineering work and having just communication being quick, simple, clear, and effective.

Mehmet: Fantastic. Joe, like, you know, I think these are all spot on points. And what I liked is your transparency also as well. And I think it takes like a little bit of humbleness, which indeed you have Joe.

Mehmet: And, you know, to your point, like I was chatting with someone the other day and said like, okay, you know, do you think this guy can take a leadership role? And you know, this [00:43:00] happens to some people, they say, look he said, I love to write code and I think I'm going to stay doing it for a long time. I'm not into, you know, and, and the guy was very straightforward saying, no, I, I don't want to become, not because I can't because, but I want to do something that I enjoy and currently I enjoy doing software engineering.

Mehmet: I don't want to go through this this leadership or like kind of management first, of course, and, and so on, which is really fantastic because also like people and this is why I like when you said like, it's not about the title, right? So honestly, I tell all people now, guys, you claim your own title.

Mehmet: You don't need someone to give you the title to, to have it. So brilliant, brilliant points you brought Joe finally where people can get in touch with you and know more about what you currently do.

Joe: So you can find me on LinkedIn. My own personal website is called jpc2. org. And you know, I used to be more active in writing things there, but lately just haven't had much time.

Joe: And then my, my kind of my [00:44:00] company website is called techsight. dev. And it just gives a brief explanation of kind of the fraction CTO work that I do, whether it's M& A, whether it's kind of helping startups scale or and one thing I've been doing recently as well as in a couple of use cases where, Part of the M& A was, okay, we wanted to operate this product, but then we also need a team to operate it.

Joe: So in one or two instances there, I've kind of, kind of been serving as sort of the fractional CTO for like the acquired product and that, and I have a couple of dev agencies that I partner with that I can help help operate the products kind of post acquisition as well. So a text like that should explain all that

Mehmet: cool for the folks who are listening to us On your favorite podcasting platform so you can find the links in the show notes.

Mehmet: You don't need to you know, just you know Replay this for for getting that and if you are watching, of course this on youtube, you will find them into the description Joe, really, I enjoyed the chat with you today. And this way it was, you know, explained to you before we started. So I call it kind of a casual chat because this is where, you [00:45:00] know, we can get the best out of every guest experience.

Mehmet: And today you shared a lot of insights, whether like related to the PropTech, MNA, Gen AI and the leadership and also like the transition. So these are like really good points that I think we were able to discuss today. So. Thank you again for giving us the time and this is how usually I end my episodes So this is for the folks who are either watching or listening to us if you just discovered this podcast by luck Thank you for passing by.

Mehmet: I hope you enjoyed if you did So, please give us a thumb up share it with your friends and colleagues and bye Come again. And if you are one of the people who are always following and always sending me their feedback and their messages, thank you for doing so. Please keep doing that. I enjoy reading all your feedbacks.

Mehmet: If I have something to enhance, happy to discuss it, happy to take it forward. Thank you very much for tuning in and we will meet again very soon. Thank you. Bye bye.

[00:46:00]