#614 AI Can Replace Tasks, But Still Can’t Replace Judgment | Dilip Chetan

Mehmet: [00:00:00] Hello, and welcome back to a new episode of The CTO Show with Mehmet. Today, I'm very pleased, joining me, Dilip Chetan. He's the founder of DefensibleZone.ai. Dilip is a veteran in technology. He worked with major companies before, mainly, as you can guess, in the AI field. But today he's, you know, gonna discuss with me about what he's currently building, you know, some of the trends he's seeing.

I don't like to steal much from my guests' time, usually, Dilip, so what I do is I keep it to my guests to introduce themselves. So tell us a bit more about you, your background, your journey, and you know, what you're currently up to, and then we will start the discussion from there. So the floor is yours.

Dilip: All right. Um, well, my background. Uh, I started, uh, in the technology field about, I wanna say 21 years ago. Um, first at Oracle, and then I moved on through multiple big companies like Salesforce, Intuit, most recently Meta and Google. [00:01:00] Uh, I've been wearing multiple hats. So I have three degrees. I have, uh, ma- three master's degrees: an MBA, a master's in computer science, and a master's in human factors, which is a branch of psychology that deals with how human beings interact with different kinds of machines.

Uh, I've been wearing multiple hats. I have been an engineer. I have been a product manager, a product strategist, and of course, I've been quite entrenched in this field of user research, customer analysis, as well as customer insights. And bringing the whole thing together, uh, in, in one form, um, was, was actually this company that I'm working at right now called Defensible Zone.

And, uh, I will pause for a minute because you probably have a few questions. But, um, yeah, happy to talk m-much more about what this Defensible Zone is all about. 

Mehmet: Yeah, sure. Absolutely. Absolutely. [00:02:00] Now, let me start before that, uh, Dilip, because what-- before talking about the company, like I try to, you know, open the way, as they say, so we can reach what, you know, um, was the main cause for you to, to start what you're currently building.

Now, you have seen it all, you know, with these big companies, and you, you saw a lot of changes. So what convinced you that AI is not, you know, yet another technology cycle, but actually it's a structural shift? 

Dilip: Uh, that's absolutely correct. Uh, a lot of people compare AI with things like the Industrial Revolution and, oh, you know, when computers, computerization happened a few decades ago, a lot of people lost their jobs.

When the Industrial Revolution happened, a bunch of people lost their jobs. There are a few very key things that I believe make AI [00:03:00] incredibly different and something we have never seen in the history of our planet. Uh, the first thing is AI is self-learning. So far, you know, you could have any kind of a machinery, any kind of instrument, any kind of computer They wouldn't teach themselves to do new things.

Mehmet: Right. 

Dilip: AI is capable of that. It teaches itself to do new things. So the same AI, like you could, for example, you could have a, uh, a GPS. Um, the GPS still does one thing, and it'll tell you the way to a, to a place, and the best way to get there. Um, it, it can't read your books and summarize it. It cannot compose your emails for you.

It cannot do your marketing for you. It cannot analyze a protein structure for you. It only does one thing. Now, AI can do GPS, AI [00:04:00] can compose your emails, AI can do pretty much, teach itself pretty much to do anything. 

Mehmet: Right. 

Dilip: That is fundamentally different, and it's a double-edged sword because while it is also your friend, it can be your competitor at the same time.

Mehmet: You know, like, it's one of these moments which are, I think, rare moments in technology where things are, you know, the A- what AI brought us as to, as you mentioned, it's not like just another technology. Excuse me just one moment. Sorry. So it's not just another technology, it's just a, you know, a moment where fundamentally we're changing the way we work, actually.

And even the naming conventions, tele- like we started to talk different thing. We used to say API, now we talk MCP, and so on and so forth, right? Um, so while preparing, and this is some of the information I got, uh, beforehand, you [00:05:00] describe AI as workforce infrastructure rather than software, right? Can you elaborate a little bit more on this, and what does it mean for, let's say, a CTO planning maybe for the next couple of years, or maybe a engineering lead who's f- again, thinking what, what AI would, would change for my organization?

Like, if you can elaborate on that more, that would be great. 

Dilip: Fantastic. Yeah. Uh, what does it mean when I say it's, it's the wh- the whole workforce is integrated, and it's not just, like you said, um, just another technology. AI at its core is to be understood, um, as something that it-- when it came in, it was something that could take over your grunt work, your routine work, work that you did not like to do, and free you up to do better things.

That was the [00:06:00] initial premise of AI. And people could do that by augmenting themselves with AI. So the way to succeed with AI is for a person to get training on how to use it and then maybe use that AI just to do all their grunt work and then sit back and think about higher level things. That's not so easy, is it?

And the reason for that, and this is where you're-- I'm going to touch on the second part of your question on what- Sure ... the CTO can start thinking about when they're doing AI implementation. Uh, a lot of companies have already jumped in headfirst. Uh, number one, and I've had, uh, more than one, uh, head of a company talk to me about this, is, uh, they come to me and, "Dilip, uh, with the help of your defensible zone, we can understand where people stand, where different individual employees stand with regard to AI replacing them."

That is the operative [00:07:00] word here, replacing people. So- Right ... I want to get rid of, say, twenty percent of my workforce or forty percent of my workforce or, heaven forbid, more than fifty percent of my workforce, and can you help me, um, do that? Can you help me understand the different kinds of tasks that AI can do, uh, that no longer require people?

And I think the number one thing I wanna stress is that is the wrong approach. You never start with AI as I'm going to use it to replace people. The second thing that a lot of people ignore is, "Oh, I can just bring in AI, and, uh, everything's, you know, uh, uh, let's just, let's just figure out how it goes."

But the way you've defined your jobs, job responsibilities, job definitions, they can- they haven't changed. So when you tell somebody you-- "I'm giving, I'm giving AI to replace the grunt [00:08:00] part of your work, freeing you up to do higher things," your job responsibility is not gonna let you do higher things.

There are other people that are doing those higher things. How many strategists do you-- can you even employ, right? So, so when you start to look at that, it's not just about bringing in a technology and training people how to do it. You have to rethink the very manner in which your organization works and what it does.

It-- the very fabric of the organization has to change, and it is not about replacing people 

Mehmet: Right. Now let's elaborate more on this. And I think, you know, the defensible zone, you know, it's a framework as I understood, right? So h-how, you know, leaders can identify, you know, exactly, you know, what genuinely is defensible versus, you know, something that is protected by today's limitation of AI.

So they d- And [00:09:00] actually, you know, from my discussions with other guests and of course, outside, you know, the thing-- common thing that always used to happen is people jump fast and they think, yeah. Actually, I think I read something about Ford the other day that they had to rehire some of their engineers that they thought, you know, we can replace them with AI, and then they figured out, like, the talent, actually AI wasn't able to, to, to get the same level of these engineers.

So h- let's, let's try, Jalap, to explain this more in details if you, if you want and, you know, get us what is the defensible zone framework is all about. 

Dilip: Perfect. Uh, and I'll start with an example just to substantiate what you said. One of my close friends, uh, got laid off at a Fortune 500 company, and three weeks later she was rehired.

Uh, it only took them three weeks to understand that they couldn't replace her with AI. [00:10:00] So yes, it is happening. I live and work in the Bay Area. I've been there, uh, and this is my third decade there, but you can 100% see that happening. Uh, where the defensible zone comes from. Uh, let me explain this concept and even why- Sure

I call it defensible zone. So everybody, um, you have a set of skills, right? You have a set of, you have a set of capabilities, and within that you have a set of skills. Skills are the capabilities that you have that you can do well. Now, within those set of skills, I talk about this concept called natural affinity.

What it is, is, in, in simple terms, if you ask me what is natural affinity, I will answer by telling you that it is why you are put on this planet. It is what you are naturally wired and built to do. It is something you can do even as part of your sleep, right? Natural affinity is something that you can do in your sleep.

It is something that you're hardwired to do, and that, [00:11:00] uh, it- that is your natural affinity. An example I give is usually in the area of sports, where, uh, you see a really good sports, sports person, they're good on the field, and then they... When, when they are old enough to retire, they can no longer play as well, but they never leave the sport.

They either turn to coaches, uh, commentators, managers of teams. They're always hovering around the sport. Right. It is because that sport is their natural affinity. Now, similarly, everybody, um, it's my theory that everybody has a natural affinity, something they're wired to do. But the question is what part of that natural affinity the market will pay for?

And that's something you, you've got to really sit back and think about. Uh, and then what part of that is AI encroaching into? And that part that AI still has not touched, that part of your [00:12:00] natural affinity that is still in demand but AI hasn't touched yet, I call that the defensible zone. And remember, I don't call it the safe zone, I call it the defensible zone because it is, AI is growing so fast there is no safe zone.

This is what I believe. Currently, there is something called a defensible zone, and you have to be really aware of, of where your defensible zone is s- and you may need to pivot, uh, on a dime to ensure that AI does not come into that area and you still are able to provide value to the org. 

Mehmet: Right. Can we conclude from that, Dilip, that based on what you just mentioned about, you know, the affinity theory you just gave, so As a leader, I need to understand what kind of affinity my team members, you know, they have.

So are we saying that instead of asking, "Okay, [00:13:00] which jobs the AI can replace," should we ask, like, what the team are really good at that we think at the moment AI will not be able to replace that? Is this, like, the approach to, to, to do it, or they should be asking something else? Like, what, what, what are you seeing when you're interacting with companies, Dilip?

Dilip: I think that you absolutely are correct. I think we need to elevate ourselves from just asking what are the different skills that AI can replace. It is not something that sits on a spreadsheet. It is asking the broader question of what are the qualities of people that AI is not able to replace? Uh, and, and currently, you know, when you Google something, every, every, every LLM will tell you when you ask it, "What can't you do?"

It's like, "Oh, I can't do judgment. I can't-- I don't have taste. I don't have accountability." But there's one other thing that AI cannot do, which is it can't do the kind of context switching or holding multiple contexts at the same time in your head. Uh, [00:14:00] that only a human being can do. And, and hon- and I think CTOs as well as people that are planning AI implementation need to understand that we need to get really broad and deep at the same time in understanding not just the skill set, but the qualities that people are bringing into, uh, uh, their jobs that we are not seeing when we are laying them off.

That's, that's the critical question to be asked. 

Mehmet: Right. Now, of course, top of mind, Dilip, and, you know, personally, I tried at least on my, uh, you know, hobby side of the things to play with the agents, and I started to learn more about agents, and everyone is talking about agents now, and organizations are rushing to deploy them.

Now, where do you think autonomous systems create the most value today, really, and where do they introduce unacceptable risks, in your opinion? Because, you know, again, when [00:15:00] the agent, the way we... A-actually some, in some use cases, yes. But you know, the way it's, it's being presented to us, at least from marketing perspective, it's like that super smart autonomous, uh, system that can replace what we do usually.

So where it becomes unacceptable because of the risks, Dilip? 

Dilip: Any agent needs supervision. I think that's the key thing within, uh, when using any agent. Uh, I've heard, oh, I, Mehmet, we should sit down and talk about the number of times I heard people tell me how their agent messed things up. Um, whether it is sim- Let, let's take a simple example of just email marketing.

Sure. I can't tell you the number of times people have combi-come back and said, like any aspect of email marketing, when you give an agent the power to send out emails, um, it can get catastrophic very quickly. So, [00:16:00] uh, you need to sit back, and, and I know people that have more than 15, 20 agents doing different aspects of the business.

It's, media has also taken up, right? Oh, the, the, you know, you hear the story of the small business owner who suddenly, uh, they have no people. It's just one individual and 25 agents, and it's a company of 26 people now. But there's only one human, and these 25 agents do everything, and it's perfect. And this person made $80 million in six months and sold the company, and now has retired in a private island somewhere, right?

That, those stories are really good, but the reality that I'm seeing is quite different. Um, they realize that very quickly, whereas as you're talking about the risk it is, accountability is, is one of the biggest risks. Uh, uh, and AI does not have that accountability. Human being is. And what I mean by that is it's your head on the chopping block, and you need to be really careful about what you're letting your AI do and the systems that you're [00:17:00] letting it take over while you're building it.

And this is not just me standing on a soapbox and telling, this is like I watched, I experienced it, and I've talked to people who give me like, "You know, my agent is great. It can do fantastic things, but Oh, there's that one time that horror story happened. That one time I nearly lost my business. So, uh, I think it's very important to, the more accountability it needs, the less autonomy it should be given 

Mehmet: Uh, to your point, I think there was another story, I can't remember, you know, the name, where the agent deleted the whole repository o- of the company and, you know, they, they lost all the data.

They didn't have backup. Um, a- and, and yeah, like simply they could have been wiped out. I'm not, I'm not sure what happened to them. Uh, but, but yeah, like we, we h- we hear these stories all the time because, you know, these agents without supervision are very risky [00:18:00] if you put them in risky places. Like let's, let's put it this way.

Now, but at the same time, Dilip, like, you know, a- and it's a, it's not... I- I'm with you on one thing about the marketing stuff. You know, like there are some people, I see them sometime in, in social media talking about the AI bros and you know, the Jarvis bros and these guys that they show you that they have multi agents doing business for them.

Yeah, we know like this is probably they, they, they're paid to, to promote something or, you know, they're just there for impressions and so on. But on the other side, and again, I experienced this to some extent, where AI is really writing good code, uh, it's writing good content, it's analyzing data perfectly, right?

So here, like what becomes more valuable for us, like whether engineers, architect, leaders, is it the supervision and the ability actually to control [00:19:00] these agents? Is this like the skill that moving forward will be more important, or what, or is it something different? 

Dilip: I think the skill that'll be more important is, is twofold.

One is to understand exactly where an agent should be allowed to play and where it does well. To your point, agents actually do, I'm not dissing on agents, they do a lot of things very well. They actually have saved a lot of businesses tons of money and time, especially time. I myself, the, the amount of work I've been able to do in the last even one month, pretty staggering.

I couldn't have done it, um, with- without AI at all. I spent, uh, I was telling somebody I spent twelve to fourteen hours a day minimum with AI, and it's frustrating sometimes, but it's also intensely rewarding. So that is definitely there. Uh, the, the thing that it is in-- it's going to get really good at is it is going to s- do less of hallucinations [00:20:00] as time goes by.

Because one of the ways in which AI is improving is it is understanding to do more with less data. And, uh, it does not need as much... The, the thing about AI, I was having this conversation yesterday with, uh, with somebody who runs a yoga studio and has now, um, really integrated AI into, uh, their business.

Uh, is, is as you're, as you're implementing AI, you need to understand how much context you can give AI. Sometimes too much context is bad. You are confusing it. You're allowing it to do multiple things. You're kind of incentivizing it to think multiple things at the same time, which it's still not good at.

But too little context is also a problem because then it starts to make stuff up. And once you understand that, then, and exactly where you can play with it, more importantly, understand how it's advancing and developing, I think the time, [00:21:00] money, um, and, and the groundwork aspect and letting it think the normal thoughts, um, really, really going to be very beneficial for people and businesses.

Mehmet: Right. So Dilip, when we talk about, you know, this judgment, context, trust, right? So these are all qualities. Do you think they can be kind of operationalized inside enterprise, or do they remain individual human capabilities? Like, in other term, it's like, you know, a specific skill to one person versus, like, something that we can standardize in the whole enterprise 

Dilip: I think parts of it can be standardized, and the reason for that is these are such nebulous terms, right?

Let's take words like judgment or trust. 

Mehmet: Yes. 

Dilip: Uh, what does, what does judgment even mean? And it, it's such a nebulous word that means so many things depending on the situation, on the context, and on the consequence. So when [00:22:00] you say, uh, can it be automated? Yes, parts of judgment can be automated. Parts of trust can be automated.

Taste, parts of that can be automated. Uh, there are things that we enjoy doing. There are things that are where our accountability comes in. Those parts are the o- areas where we need to be a little more careful in where AI goes, but you're-- I completely agree with what you're saying. Uh, when you're breaking down judgment, for instance, you can actually give it metrics to understand what you mean by operationalizing a word like judgment.

Uh, let me give you an example. 

Mehmet: Sure. 

Dilip: When you're talking about quality, uh, quality is another nebulous word. Um, you, you need to specify to the AI what you mean by quality. And I worked, uh, at Meta. I was, um, part of the team that was trying to understand the word quality when it came to understanding reels and short-form videos.

Uh, quality can mean, for example, you can [00:23:00] operationalize the part of the quality that talks about performance. Is the video capable of, you know... Is, what is the size of that video? What is the, um, um, the pixelization? What is the, uh, uh, length of that video, and how does it appear on low bandwidth systems, and especially even on cell phones in, in parts of the world where people don't have that kind of bandwidth, right?

Things like that are one aspect of quality that can be operationalized. Another aspect of quality could be when you're talking about how does it make me feel? Is, uh, am I watching a funny video? Is that good-hearted funny that I can laugh at? Is it the kind of funny that I can only share with my friends?

Is it a kind of- Right ... funny that I can share with my children, with my parents? Uh, there are different kinds of jokes. So what does quality mean in that kind of an aspect? Those are areas where AI has trouble breaking that down. So even when you're [00:24:00] operationalizing judgment about the quality of a particular video, there are aspects that you can actually automate right now, and it'll do well.

There are aspects where AI is still not built to, to operationalize. All 

Mehmet: right. Now, Dilip, I'm sure you interact with a lot of, you know, um, company leaders. Uh, and I'm not sure if you know this In other way, I'm not sure how this is being, uh, developed currently or like h- how, how the, uh, I would say transformation is happening currently.

Because, you know, w- if we ask maybe two years ago what are like the, mm, AI success metrics or gains, you know, maybe we would receive different answers. But now with everything happening, the answers are different. So what do you believe actually the metrics that are the important the most when we do, quote-unquote, "AI transformation" [00:25:00] that we want to Because people, I think the mistake that's happening, and I'm trying to relate it to what you're doing with at Defensible Zone.

I think the main thing is we are, uh, kind of mixing tactical shifts or moves versus a company vision to... And back to the affiliation, you know, affinity thing that you mentioned previously, which is if I want to think about it from business perspective, why a company exists, right? So, so it's their vision, right?

Like, what they're trying to achieve. And, you know, how we can use AI maybe to have kind of a competitive advantage. So how these metrics today should be really measured if we want to stay more competitive and of course, to the productivity part you just mentioned, saving costs and all this. So how do I track this and what I sh- where I should be focusing the most in these metrics?

Dilip: The thing I really like about [00:26:00] your question is you're asking like, you know, tying it back to the vision. Uh, you'll be surprised, Mehmet, like how few companies or people even think about that, right? Is like how few people even... When they are starting to implement AI, I think the main things they are thinking about are actually just, how do I bring it in as a new technology?

How do I improve a few things? I always advise people to spend, you know, it's like measure twice, cut once. In this case, measure 10 times and cut half, half of it. And the reason for that is: What is the goal of your company? You know, when you're sitting back, before you get into the metrics aspect, first understand why you're bringing AI into your company.

What is the goal? If it is something like, "I wanna lay 30% of my workforce off," then you're even starting, uh, ex- in, in a very wrong part of, um, that, that particular [00:27:00] aspect. So you want to sit back and think about, is AI even necessary for my company? And why am I bringing it in? Is it just so I stay cool? I can tell people I have 10 agents running, doing, running around doing things for me, or is it, am I really looking at some sort of a productivity gain, some sort of a monetary gain, some sort of a combination of, you know, I'm, I'm freeing myself up and freeing my employees up to do these other things that currently they don't have time to do, but they have the natural affinity to do.

Uh, and then so when you, when you really think through those and then understand if it's aligning with the vision with which I started my company And then you get into, okay, now you're breaking it down, right? You're starting with the overall goal, and then you're getting into your vision, and then here's why I need AI, here's how I need AI to augment my company, and here's what I need my people [00:28:00] to be doing with that AI.

And then you sit back and think about, how do I measure? How do I know I am successful? If it's saving something, what is... how is it going to look like? What is it gonna look like in one month, three months, six months, one year, two years, five years? What is this... Five years is extremely long time when it comes to AI, but I'm just giving you an idea of that.

So then you start thinking about, okay, what are the metrics that really matter to me that I can assess? And the example I gave with judgment in terms of the quality of a reel, right? That alone is something that talks about, okay, so now what is important to me here? Is it just serving up more reels to people, or is it making people talk about the quality of reels and say, "Hey, I got to go to this particular application because the quality of videos on this is really amazing," whereas the quality on a competitor's, uh, you know, platform is subpar.

So what is the [00:29:00] goal that you wanna do? And depending on the goal, you tie the metrics. If the goal is something that's simply saving X amount of time or, or, you know, doing these functions within a shorter time or doing these functions in a better way so that we can achieve this particular, um, outcome, that's important.

How you tie your goal with the outcome and are you measuring it? This is a very, very key thing that you're talking about where a lot of companies are not good at doing it. Because they need to get really deep into understanding, how can I operationalize this? What am I measuring and why am I measuring it?

It's very important. I'll just leave you with one, uh, one small thing, which is just because you measure it doesn't mean it's right. So you need to understand why you're measuring something 

Mehmet: It's a good, uh, it's good, you know, thing to, to think about it [00:30:00] actually. Uh, well, a- a- and again, to, to your point, Philip, like it's about Maybe we're mentioning something a lot of people knows about, you know, starting with why, right?

Uh, Simon Sinek, like, he's, he's famous for this, like, asking why, why you do what you do, right? So, uh, I, I think it's so obvious sometime to me at least, and I'm sure it's for you also, Dilip, but people, I think they go with the hype and all the, you know, these fantasy stories that they see, back to what we were saying, which actually it's an, it's also the noise that happens around which here where the filtration need to be, uh, you know, part of this.

Now, if I want to ask Dilip, from leadership perspective, what can, uh, what, what kind of leaders will change with time and which one will, will, will, will stay, I would say? 

Dilip: Oh, I love the [00:31:00] question. Oh, this is such a beautiful question. I keep thinking about this a lot. Uh, y- you know, um, s- so interesting. Uh, the, a leader, the, the, the qualities of a leader, um, whenever I, I was, one of my close friends is a, is a venture capitalist, and I was asking him, you know, "When you invest in a company, uh, what, what do you really look for?"

And he said, "Oh, absolutely the quality of the CEO, the, uh, because we need to understand it's not about what we're investing in, it is who we are investing in, and we need to understand the person more than anything else, more than the business, more than..." You know, we've, he, he was telling me examples of somebody who had a really bad demo, who, uh, the pitch was not good, but they understood the tenacity of this individual, understood their vision.

So I would say a leader who is focusing, if, if, if you, any leader, anybody you talk to [00:32:00] is, like you said, caught up in this hype, because, you know, AI is the new sexy, AI is the new shiny, AI is the- Right ... new thing. Um, it's the new cool thing, the new cool toy that people think like, "Oh, this, it's just like any other toy."

It is not. It could replace you. It's, it's not a toy if it can replace you. So you need to understand what, what it is you're dealing with, and a leader who understands why they're in business, and what is this business all about, and s- who has this long-term vision on, you know, "I want to do this with my business," and that this is not answered by the exit.

It's not about the exit. It's not about, "I wanna sell my business for 80 million, 100 million, one billion," or, uh, or, "I want to get bought over by," um, uh... Without giving you too much information, I was interviewed by a company in Germany [00:33:00] who- whose one goal was to get bought over by one of the largest companies in the world.

They're like, "Oh, we had, you know, we know, we know these other engineers that started this company and they got bought over, so we want, we wanna be that." I'm like, "Is that the vision? Is that the goal? Like, you just want a big company to come and give you, like, $20 billion? That's your goal?" Yeah. "What are you even doing?"

Right? So it's, it's, it's really interesting and, and I think the, kind of the quality of the person is what is their vision? What is their goal? Why are they in business? What do they intend to do for people? And what is their reason for existing? What is their natural affinity? That's-- And, and to remember that anybody that's investing is investing in the person, not in anything else.

Mehmet: Hundred percent, and I will refer people to your point, uh, uh, old but gold, as they say. There is a short clip from Steve Jobs where, you know, he says, like, [00:34:00] when people comes to-- used to come to him and ask him, "Hey, we want to be entrepreneurs," he used to ask them why you want, and they, some, m-majority of them, they will say, "Yeah, we want to be rich," and, you know, like, uh, you know, retire early.

And he said, "Okay, you're, you're in the wrong, you are in the wrong business." Like, it's not like how it goes. Like, you need to, to really sometimes suffer. If you're lucky, maybe you will not pass through this, but, I mean, you need to have this mindset that nothing is easy, actually, and, you know, you need to have a purpose.

And the main thing, which I think AI is, uh, and, you know, the conversation is somehow related to this, and this is where AI, I think it's not able to change even, you know, um, back to the defensible zone also, Dilip. It's this purpose and being, uh, persistent, right? And, and having this perseverance also as well and, you know, just keep trying.

Resiliency also as well, right? So AI [00:35:00] can't, can't have these traits because AI, it's at the end of the day, I tell people, you know, like, AI at the end of the day, if you really study how it do- of course, it's perfect technology, but it's trying to take shortcuts. And, you know, in real life, people who tries to take shortcut, usually they don't succeed.

You need, you need really to, to work hard to do something, and the LLM is just guessing the next token and just it put on in front of you something that should look meaningful for you. But us humans, this is what makes us different. And this brings me, you know, to the, you know, kind of the end of this conversation.

If we want to take one practical, you know, takeaway about discovering our own def-defensible zone, uh, what question should we ask ourself tomorrow morning, Dilip? 

Dilip: Uh, I love it. Uh, thank you for asking this, uh, this very important, uh, question. What you need to ask yourself, what you need to do is you need to introspect.[00:36:00] 

And I usually recommend you do it like I, I have exercises that I do with people, um, and, and that exercise involves you sitting down and listing out what you are. And the reason why this is so important, and, and I'll explain what, what I mean by that. By what you are, it is not just a set of skills. You are not just a set of skills.

You are so much more in terms of the qualities you bring in. Like, what am I, for instance? Am I a manager? Am I an engineer? Am I a product strategist? Am I a researcher? I'm all of it. In addition to that, I'm also a tenacious individual. I'm, I'm a risk-taker. I also have an enjoyment in doing things, and, and just how I love to see where the trail leads.

I'm one of those people that will take the road less traveled just because it's road less traveled. [00:37:00] And yes, there are perils in doing that, but that is my nature. That is my, that is what drives me. Uh, I am a risk-taker. I do dangerous things. Um, and, and, but I get a kick out of doing them. So, what does that, how does that translate into business?

I really like to go deep, deep, deep into anything. Any topic that I take up, I, I will... Uh, for example, I wanted to start a distillery. I went and took courses, um, um, in Scotland about distilling, and got into this whole business of whiskey so deep that now people don't wanna talk to me about whiskey because I start talking agriculture and talking about quality of barley.

And they're, "I just want the taste of the," like, you know. "Let, uh, let me explain how this work," right? So, getting that, that is, that is me. That is my natural affinity. And that is why I started this company called Defensible Zone, because I wanna help other people discover themselves. So, in one word I can say introspection.

You need to [00:38:00] really sit back and introspect because I can guarantee you when you do that, you're going to discover new things about yourself that you didn't know exist. And the reason for that is because you've been pigeonholed all these years. When you work for a company and you ask people what they are, they will give you their job description.

And the larger the company you work for, the smaller is the sliver of work that you do. And that's all you think you're capable of. And, and then people pigeonhole you into that even outside. They're like, "Oh, this is this person. This is what this engineer does. They're, oh, this is a backend engineer. That's, that's all this individual is."

No, they're not. Uh, do they have the capability of running a company? They probably did at some point. They've forgotten that. So, when AI is coming, and when you want to discover your natural affinity, and to understand, you need to understand yourself first, then you worry about understanding market demand and AI 

Mehmet: Right.

Final traditional question, Dilip, where people can get in touch and find more? 

Dilip: Defensibledo- zone.ai [00:39:00] and dilipchetan.com. Dilipchetan.com is my personal website where I write about myself. Um, but defensiblezone.ai, if you go in there, um, you'll start seeing products, assessments for, for, uh, professionals, as well as I, I have links for, uh, businesses.

Um, I'll be happy to give them a, you know, they can mention you and they get a free, um, a free 30-minute to 45-minute call with me where we can walk through what needs to be done for 

Mehmet: you. Great. I appreciate that, Dilip. So for the audience, you will see the links in the show notes if you're listening on your favorite podcasting app.

Uh, if you're watching this on YouTube, you'll see it in the description. Uh, Dilip, I can't thank you enough for this discussion. It's, uh, you know, I like when... And this is, you know, the whole show is the idea is not we talk about bits and bytes, although, like, it's called The CTO Show. We try to bring the business side, we try to bring, which is more important in my opinion, the human side of the things.

And [00:40:00] today we discussed, you know, so the takeaway for me is finding, you know, the affinity. Finding, you know, what's your purpose, what you do, what you do, right? And actually try to leverage the tech- this awesome technology, which is AI today, to help you. Uh, and, uh, you know, I, I think this will make people more productive.

This will make people less, and most importantly, less stressful because we hear about all this noise and people get worried we're gonna lose our jobs. So I think, you know, what you're doing, Dilip, is something very, very, very, you know, uh, valuable. And thank you again for sharing this, uh, on my podcast today.

And this is how I, I end my podcast, my episodes. This is for the audience. If you just discovered us by luck, thank you for passing by. I hope you enjoyed. If you did so, give me a favor, subscribe and share it with your friends and colleagues. If you are one of the people who keep again and again, thank you for the support.

Thank you for the feedback. I'm trying my best, so thank you very much [00:41:00] for tuning in today, and we will see you again very soon. Thank you. Bye-bye.