Why AI Won’t Replace Software Engineers Anytime Soon | The Disruptors #10 - Intuji

Show: The Disruptors Episode: 10

Publish Date: 01 September 2025

AI makes coding easier than ever — but at what cost?

In this episode of The Disruptors, Ritesh Shah and Julian Wallis unpack the hype around AI in software development. They explore why AI can speed up tasks but can’t replace the judgment, discipline, and experience of real software engineers.

If your business is betting on AI to build scalable, secure systems without proper oversight, this episode is your wake-up call.

Topics Covered

00:00 – Intro

00:45 – Why AI is everywhere in coding today

02:30 – Garbage in, garbage out: the problem with AI training data

04:30 – Augmentation vs replacement: where businesses get it wrong

07:00 – Security and scalability risks of AI-generated applications

10:00 – Developer vs Engineer: why experience and strategy matter

13:00 – The danger of chasing short-term wins with AI apps

16:00 – Why critical thinking is still the differentiator

19:00 – How businesses should actually be preparing for AI

24:00 – The PSV Thinking™ lens for using AI effectively

29:00 – Closing message: AI is a tool, not a brain

Key Takeaways

AI speeds up development but can’t replace critical thinking.
Businesses risk scalability, security, and IP if they rely blindly on AI.
The real differentiator is how you use AI — within guardrails.
Engineers provide judgment, strategy, and architecture that AI cannot.

Transcript

Julian Wallis, CEO at Intuji (00:00:00:01 – 00:00:09:03)
If you know what the good output is, you can use AI to your advantage to get things done quicker. If you don’t know what a good output is, how do you know if you’re producing garbage?

Ritesh Shah, CTO at Intuji (00:00:09:05 – 00:00:13:01)
With the apps built with AI right now it’s risk with the data.

Julian Wallis, CEO at Intuji (00:00:13:06 – 00:00:24:14)
Everything on the surface looked great. The people that are in it and are using it like we are. Know that it is not bulletproof. It is a tool to be more efficient within guardrails.

Ritesh Shah, CTO at Intuji (00:00:24:18 – 00:00:30:23)
AI can write code. I can’t deny that. But what is the quality of the code? Is the engineer’s decision.

Julian Wallis, CEO at Intuji (00:00:31:04 – 00:00:36:11)
When you look at it as a replacement, not an augmentation, you’re going to have problems.

Ritesh Shah, CTO at Intuji (00:00:36:13 – 00:00:40:23)
Hey, Julian, welcome back to yet another episode of the Disruptors.

Julian Wallis, CEO at Intuji (00:00:41:01 – 00:00:43:07)
Thanks for having me.

Ritesh Shah, CTO at Intuji (00:00:43:09 – 00:00:47:03)
How was your weekend? How is your days going on, everything?

Julian Wallis, CEO at Intuji (00:00:47:04 – 00:00:55:10)
Good, can’t complain. Always got more to do and always got things going on don’t we so, but, it’s been been good, looking forward to this. This is a good chance for us to have a good chat.

Ritesh Shah, CTO at Intuji (00:00:55:10 – 00:01:03:12)
So yeah, same likewise, mate, it’s a best chance to have, like a good meeting, a good conversation, I love it.

Julian Wallis, CEO at Intuji (00:01:03:12 – 00:01:04:14)
Yeah. It’s good, it’s good.

Ritesh Shah, CTO at Intuji (00:01:04:15 – 00:01:31:02)
Yeah. So today we will be discussing around AI. AI as a developer and, or developer with an AI or engineering difference. We will be discussing around the use of AI and being a developer and all that kind of thing. So let’s get it started. Let’s dive into it, to what do you think about today, exactly? Anyone can be a developer.

Ritesh Shah, CTO at Intuji (00:01:31:04 – 00:01:37:00)
Write me a code for this. AI writes it, writes it. What do you think about it?

Julian Wallis, CEO at Intuji (00:01:37:01 – 00:02:01:08)
It’s interesting times. Obviously, it’s only getting more and more. There was a new model for Claude released yesterday, and they have to do with their coding model. And yeah, it’s just coming out more and more. ChatGPT are releasing more and more models, etc. it’s happening more and more. We’re seeing products like Lovable and Cursor etcetera that are.

Julian Wallis, CEO at Intuji (00:02:01:14 – 00:02:33:03)
You know, out there as these AI generated applications theres so many hype stories about overnight SaaS products and all this kind of stuff, but there’s so many I don’t know, there’s, there’s so much potential there. But I the way I look at it as is, if you don’t know what you’re doing, you don’t know what’s good or bad, and it comes back to everything, which is, if you know what the outputs meant to be, if you know what you meant to be doing, you know how it’s meant to be built, whether that’s technically whether that’s product strategy, whatever it is, could be anything.

Julian Wallis, CEO at Intuji (00:02:33:05 – 00:03:04:00)
If you know what the good output is, you can use AI to your advantage to, to get things done quicker. If you don’t know what a good output is, how do you know if you’re producing garbage or rubbish? That to me is the difference. And I’m seeing a lot of businesses, not just businesses, but I’m seeing a lot of developers, but also then businesses implementing things or doing things with, they’re just doing things because it looks good on the surface, and underneath there’s massive issues.

Ritesh Shah, CTO at Intuji (00:03:04:02 – 00:03:41:19)
Yeah. Recently we have experienced a lot of, codes that have, AI has generated. What my, my initial problem is, AI is generating codes that AI is using for it’s, as a sturdy material or data for it, its, its, its like feeding itself with its own food and like so, now the, now I see already started seeing is is taking garbage in and then getting out the garbage out actually.

Ritesh Shah, CTO at Intuji (00:03:41:21 – 00:04:02:19)
So, and, then yeah basically that is what AI, AI is has improved a lot. There’s no question about that. I totally see the value in using AI. We can’t complain about it. We have to use it. And, but developing a proper application out of it, I still question that.

Julian Wallis, CEO at Intuji (00:04:02:21 – 00:04:22:15)
Yeah. And I think this is where, sorry, just on that to comment on it, just like, it’s good like if we use it as an assistant or as a helper here and there, within reason, within measure, when you start relying on it as like the primary source, right, of like the work or whatever, and it’s as a replacement. That’s where we start to see the issues crop up.

Julian Wallis, CEO at Intuji (00:04:22:15 – 00:04:41:00)
I think that there’s no doubt there’s like you can generate smaller-scale applications, you, one-page websites and small, even smaller-scale applications, code here and there to, for example, you can, what would have previously had like been able to be would have to be done by for example, a developer, maybe a small Google app script or a

Julian Wallis, CEO at Intuji (00:04:41:00 – 00:05:15:03)
NetSuite script or whatever, smaller things like that, then, yeah, you can do it. But if you’ve had no experience doing it, you’ve got a possibility of running a script that deletes the whole database or whatever else. Like I’m saying, if you don’t know what it’s producing to a degree, then you’ve got issues. When you’re doing smaller work or like a one script or a small thing, obviously, there’s less at risk, like it’s, there’s less complexities, less moving parts when you start using it as a replacement for like, okay, where now going to be doing this large-scale project and going to replace everybody with AI.

Julian Wallis, CEO at Intuji (00:05:15:05 – 00:05:20:08)
That’s the problem. I look at it as augmentation, not replacement. There’s a big difference.

Ritesh Shah, CTO at Intuji (00:05:20:10 – 00:05:47:02)
Exactly. Basically in market also right now what’s, theres over the internet, this news going on, people, software engineers getting laid, laid off, laid off because of this, because of AI, 40% Microsoft laid off something. This kind of news is all around the internet, I see that every day. They are laying off 40%, but still 60% are there, that is the use case of efficiency.

Ritesh Shah, CTO at Intuji (00:05:47:02 – 00:06:20:03)
They have improved the efficiency. That’s what I see. But they still need engineers to code. Basic, but looking at that, people are other people who are the smaller companies and then trying to build out products here and there, they are trying to do, replace, firstly replacing those engineers itself with the AI. If you are replacing the engineering team, then it’s, then it’s a vast difference there.

Ritesh Shah, CTO at Intuji (00:06:20:03 – 00:06:44:09)
Vastly different, if that makes sense because because like yeah the ground fundamentals of programing, engineering fundamentals of programing, the direction. It’s the human that gives that direction. I, I last couple of days ago we were discussing the same thing. Right. Anyone can be a developer. Anyone can be engineer from the day one.

Julian Wallis, CEO at Intuji (00:06:44:15 – 00:07:06:03)
I agree. And there’s a few memes and funny funny things but it is the irony of it is it does have some truth in it. Is that like, I saw someone posted something a while ago, Guys like, don’t worry about AI, no stress. In order for it to work effectively, stakeholders will actually have to give proper requirements.

Julian Wallis, CEO at Intuji (00:07:06:05 – 00:07:31:16)
And it was like quite ironic. It’s funny, but like, it is like another point is AI can’t read your mind and it can’t read between the lines even necessarily. Right. Now, if you give it enough context and stuff like that there, it can see some things. But again, I’m a big advocate, it sounds like we’re hating on AI here, but I think the most important thing and I had a discussion with multiple businesses just this week across a couple of points.

Julian Wallis, CEO at Intuji (00:07:31:16 – 00:07:52:10)
Firstly, being businesses that have used, you know, AI in their products to actually replace engineers or work with, with engineers and caused, ended up causing more issues than solving. Right. And caused more problems with the product, going backwards, than actually moving forwards. Other businesses are looking at how they implement AI in their business. So looking at how they can actually do it.

Julian Wallis, CEO at Intuji (00:07:52:12 – 00:08:10:13)
And just across the board, fundamentally, for me, it comes back to having the right systems and the right processes in place and the right data. And for example, what I mean by that there is, you can go and get someone to even, let’s say, for example, take a manufacturer here, they might want to do a small script on.

Julian Wallis, CEO at Intuji (00:08:10:13 – 00:08:26:06)
NetSuite. Fine. They might be able to get that one of the in-house IT guys to generate a script and it works. All right. That’s fine. Not, not a big deal. They then want to go and generate an, let’s say, an integration or a small web app that integrates with NetSuite. This is just a practical example I’m using.

Julian Wallis, CEO at Intuji (00:08:26:09 – 00:08:44:06)
They use a product out there to just generate the actual application. They vibe code it so called vibe code it, connects up with NetSuite, all this stuff. They find out six months later all of their data from NetSuites’ been publicly exposed for the last six months. That’s just a just a random example. They didn’t know what.

Julian Wallis, CEO at Intuji (00:08:44:06 – 00:09:04:07)
They didn’t know. They were doing to the best of their ability. Everything on the surface looked great. Everything looked like it was working. And this is the point where, a problem where I have with all of this hype and stuff, the people that are in it and are using it, like we are, a lot, know that it is not bulletproof, right?

Julian Wallis, CEO at Intuji (00:09:04:09 – 00:09:30:08)
It is a tool to be more efficient within guardrails. It is not a replacement, it is an augmentation. I sound like a broken record, but you take the practical example I just gave for like a manufacturer, wholesale, whatever. And you look at the T-app story that we’re talking about that was ‘vibe coded’ by a, non-technical founder who was very proud of, you know, being a non-technical founder.

Julian Wallis, CEO at Intuji (00:09:30:10 – 00:09:52:13)
I mean, and look what happened. Now, it’s not to say that, there’s been other stories where there’s been really big teams that have had those issues as well. It’s not to say that is the only, like or just because of that. Like, there’s, there is issues and there’s apps that can be built incredibly well that still have vulnerabilities, like it’s not bulletproof, but it’s more about the point to say, it’s not just a game kind of thing, right?

Julian Wallis, CEO at Intuji (00:09:52:13 – 00:10:10:19)
Like as in, there is actually serious stuff at play here and that, this is going to happen more and more. And it’s just going to be, I personally think, going to continue if people keep, keep, using it for, for production stuff.

Ritesh Shah, CTO at Intuji (00:10:10:21 – 00:10:35:20)
Exactly. Ideally, what I think about these AI tools is, these tools are has helped me a lot, basically, a couple of years ago, I used to have to go to Stack Overflow. I had to has to research for hours and hours to get one solution, and then I would try it, implement it and then see what’s the visual and.

Ritesh Shah, CTO at Intuji (00:10:35:21 – 00:11:08:00)
Does it even fix for me, fit me or not? That kind of decision. It was, it used to take like hours, right? Today it takes like minutes. That has helped me. But using this, I need to know what input, what context, what kind of output I need to get. I need to know all this kind of thing. If I don’t know what output I’m expecting, then it takes me.

Ritesh Shah, CTO at Intuji (00:11:08:02 – 00:11:33:07)
It could, it has taken me out of context and then went far away then the result what I want it to go. And then if that kind of solution goes to the production as far away from where you’re looking for, and then engineering actually is something different to development. Developer, developer and an engineer are completely different. They want anyone to they can be a developer, but engineer.

Ritesh Shah, CTO at Intuji (00:11:33:07 – 00:11:58:06)
It takes years and years of experience. It’s all about planning. It’s all about scalability. It’s, it’s all about maintainability. It’s all about everything. It’s not just one solution that we are like, it’s, you are not building a hut. Are, you are building or trying to build today, you are building, trying to build or really trying to planning to build a skyscraper, right?

Ritesh Shah, CTO at Intuji (00:11:58:08 – 00:12:00:14)
You’re not building a small house or something.

Julian Wallis, CEO at Intuji (00:12:00:16 – 00:12:24:11)
I think, I don’t know if this has been verified as well, but with the specifically with the T situation, I’m not, it’s alleged, in regards to I, have read it in a few different places, but I haven’t seen it verified. But it wouldn’t surprise me if this is the case that as they started to get some growth, they had to completely well, they tried to re-architect the whole thing because of the issues that they were having with scaling.

Julian Wallis, CEO at Intuji (00:12:24:13 – 00:12:49:19)
And, but they had the base code base, the fundamentally with the issue that I had, they had basically a firebase and I think it was their firebase data, but their object storage or whatever was public. I don’t remember exactly what it was, but, pretty fundamental flaw, whether that was made by a human, I would imagine it would have been because I know with Fire Base, you have to get about 16 warnings to turn to actually turn it from private to public.

Julian Wallis, CEO at Intuji (00:12:49:19 – 00:13:09:14)
Right. That whole thing aside, to me, it’s more about the fact that think of it like this. Even today or even three years ago, anyone could then write a book with AI, if you know what I mean. Like, you could write a book, or you could write blog posts and there’s some people that will get some quick wins out of it, if you know what I mean.

Julian Wallis, CEO at Intuji (00:13:09:14 – 00:13:38:08)
They will make some quick bucks or some quick, quick wins, etc. but there’s no long-term win necessarily, right? It’s a short term win. It’s a short term thing that it gets. Anything that comes, I guess, is that I think anything that comes easy to a degree is not worth having. And I guess it kind of applies in this regard, because when you think about it as this becomes the new normal, the new average, let’s say, for example, we can spin up a couple of page web app that does a certain thing.

Julian Wallis, CEO at Intuji (00:13:38:08 – 00:14:01:00)
If anyone can do that, then we move on to the next level. Do you know what I mean? Like, it becomes the new normal. It becomes like, okay, what’s the next level that we’ve got to achieve, right. And yes, there’s going to be changes in, you know, people have been saying for a while now it’s not going to be software as a service, going to be AI as a service.

Julian Wallis, CEO at Intuji (00:14:01:00 – 00:14:29:15)
So you’re going to have like a layered, you’re just going to have conversational AI that does stuff in the background. Yeah, like there will be ways in, that’s a UX thing to a degree of like how you actually interact with the applications and the technology. I’m not even talking about that here, but I guess fundamentally I’m ranting around a bit, but is fundamentally like talking about the fact that AI, when it becomes, if there’s no differentiation, there is what I’m trying to say, then what is actually your differentiation?

Julian Wallis, CEO at Intuji (00:14:29:15 – 00:14:49:15)
Right. What are you, you’re just generating a generic app that everyone else has. And I won’t even get into the fact of what we’re talking about here, developer versus engineer. But then you’ve got all your intellectual property, all your data ownership issues, all these other things. But fundamentally, yes, you’ve got to know what good output is to use AI effectively.

Ritesh Shah, CTO at Intuji (00:14:49:17 – 00:15:35:19)
Imagine if self-driving car was launched on the road the first day they built it. It took years to get there, right? Its, it still is still not definitely right? It because like there was a race of life, life and human lives and everything. So they didn’t do it still. But similarly with the apps built with AI right now, it’s risk with the data, its risk with the scalability, it’s risk with the everything you’re losing yeah. Still, it needs to be properly planned out and everything there. If you, like for a startup, there, it would just a T app actually, like this was just an idea.

Ritesh Shah, CTO at Intuji (00:15:35:19 – 00:15:49:17)
He’d clicked it, he did it, he or she, whoever did it and launched it, but it caused a lot of like, the lawsuit might cost him a lot, whatever or anything like that. But yeah.

Julian Wallis, CEO at Intuji (00:15:49:17 – 00:16:09:03)
Because there’s obviously significant issues with it. And I guess they didn’t start out thinking like that. And again, I’ve said this before, the AI stuff for ideating prototypes and maybe even for working on production stuff as like an assistant, as a helper, as an augment. It’s brilliant, right? It’s not to say AIs bad, but when you look at it as a replacement, I think that’s just the simple message.

Julian Wallis, CEO at Intuji (00:16:09:03 – 00:16:29:04)
Yet when you look at it as a replacement, not an augmentation, you’re going to have problems. Imagine a, like let’s put it this way. Okay. This is actually an interesting question for you. And more about the like, do you ever see in ten years time where like just say a bank, a large bank, AI is basically doing all of their work and there’s hardly any human oversight at all.

Julian Wallis, CEO at Intuji (00:16:29:06 – 00:16:36:18)
Right? So it’s it’s making deployments, it’s making production changes. It’s yeah, there’s no oversight. There’s no nothing.

Ritesh Shah, CTO at Intuji (00:16:36:20 – 00:16:49:21)
You got it wrong. Because I don’t know, I don’t see a wrong example. But it will it will because like in ten years time do you see money there or cash money or not? Like that is also one of the biggest question.

Julian Wallis, CEO at Intuji (00:16:49:23 – 00:17:27:09)
In the bank. Yeah. No I would, I will I would say yes, they will absolutely still be, still be cash. That would be my opinion on it. But let’s put that aside. The point of it is, is that no matter what it is, a bank could be digital currency, it could be physical currency, whatever. Are we going to trust the fact that we’re going to have like no oversight and that production deployments, changes like database changes, disaster recovery, who’s, who’s handling or who’s doing all of that, who’s actually making sure like, oh, when we scale now, when we do this, or we add this new service or who’s making those product decisions.

Ritesh Shah, CTO at Intuji (00:17:27:11 – 00:17:29:02)
Like.

Julian Wallis, CEO at Intuji (00:17:29:04 – 00:17:32:02)
It’s not even feasible to have all of that.

Ritesh Shah, CTO at Intuji (00:17:32:04 – 00:17:59:16)
Marketing decisions like, yeah, this, basically what I see is like, sure and certainly the resource used right now, the efficiency will increase for sure. We won’t require human, won’t require that many number of people, that’s for sure. But decisions are to be made by the human. It’s right or wrong that AI can’t take the responsibilities with that decision making.

Ritesh Shah, CTO at Intuji (00:17:59:16 – 00:18:30:06)
They can command give the commands. Okay, think about this. This is the right decision. Do this, do this, do that. That’s going to happen for sure with software as well, with everywhere. That’s what I see. But still the brains behind won’t be removed. The fundamentals won’t be removed because it works on the fundamentals. That still the computer. There’s, still to, in today’s day it’s still in binary.

Ritesh Shah, CTO at Intuji (00:18:30:06 – 00:19:00:14)
It’s still zero and one working, so it won’t be removed. This idea of the, quad computing thing coming in, that’s a different story. But still, right now, today, the power will be enough, the efficiency will improve. The whatever the computer is doing right now can do more efficiently with less power and all that. The efficiency with computer as well, the power of the computer will be.

Julian Wallis, CEO at Intuji (00:19:00:16 – 00:19:22:20)
Essentially it’s going to be able to do more in the, in less time really. So it’s just taking the same thing more in less time. And I think for anyone listening, businesses listening, there’s two key points from from an AI perspective, they need to be thinking about. The first is, how are our systems and our structures set out for AI and how we’re going to actually augment our processes?

Julian Wallis, CEO at Intuji (00:19:22:22 – 00:19:45:04)
Most businesses that we work with are nowhere down the line of even digitising properly, their processes let alone augmenting AI into them. And it’s a lot harder when you actually start doing it than people realise. It sounds easy and you can do your simple things easy. Like you might be able to do a simple task that automate something with, you know, a summary or something like that there pretty easily, which is, yeah, even that’s incredible.

Julian Wallis, CEO at Intuji (00:19:45:05 – 00:20:18:23)
It could have been done five years ago, kind of ten years ago. But to really take advantage of it at scale, it really needs how you’re looking at your data, how you’re setting things up. And then the second point with what we’re talking about is we need to be vigilant and very intelligent about how we actually use it to generate our outputs and what that looks like. When we’re doing content and written text, which is, let’s be honest, the vast majority of what AI use for at the moment is written stuff emails, blogs, proposals, whatever it is, it’s it’s written, it’s it’s writing.

Julian Wallis, CEO at Intuji (00:20:19:03 – 00:20:56:23)
It’s. Yeah. Now a lot of, in the I’ll take that, will clarify that in the general population is what I’m saying. The vast majority using it for that writing, generating content, generating images, whatever it’s for, that’s the idea generation, etc.. And then there are obviously a lot of then come to that code and everything. I guess the point of it is to the businesses though, is what are we doing to make sure that we’re not releasing AI and developing apps or connections and whatever integrations, etc., that are probably developed internally and and not up to spec.

Ritesh Shah, CTO at Intuji (00:20:57:01 – 00:21:09:19)
One thing I would ask you is do you see anytime soon in the future AI is spending your money like, that’s a good question.

Julian Wallis, CEO at Intuji (00:21:09:21 – 00:21:12:17)
For me personally, are you asking?

Ritesh Shah, CTO at Intuji (00:21:12:19 – 00:21:14:08)
Yeah, do.

Julian Wallis, CEO at Intuji (00:21:14:10 – 00:21:15:14)
Would I let it?

Ritesh Shah, CTO at Intuji (00:21:15:16 – 00:21:17:03)
Yeah.

Julian Wallis, CEO at Intuji (00:21:17:05 – 00:21:19:16)
At this point I would say no.

Ritesh Shah, CTO at Intuji (00:21:19:18 – 00:21:47:12)
Yeah. It’s, it’s your decision right where you want to invest, where you want to spend, how you want to enjoy. And with businesses also like, with coding, also with, oh okay, I want this quality of code or I want this one to be you want this to do this much or anything like that. That is it’s your decision, right?

Ritesh Shah, CTO at Intuji (00:21:47:14 – 00:21:49:09)
It will be your decision.

Julian Wallis, CEO at Intuji (00:21:49:09 – 00:22:05:16)
I guess the way I would look at it would be no, at this point, possibly in the future as in I could potentially be open to it. But another point I would say is that I can see it changing the way of buying behaviour and everything else where its like, I have actually done all of this research and I’ve done this, I’ve done that.

Julian Wallis, CEO at Intuji (00:22:05:16 – 00:22:32:10)
Here’s your options. Click this to buy it and I’ll buy it for you. Yeah, right. But you’re still making that decision which is what you’re expecting. But we have to be aware of the changes that will make to get to that, if that makes any sense. And then on top of that, then it starts introduce a whole another question of like, this is where now we’re getting probably way off our original topic, but it is a point that probably relates to it, is like, where are people going to start tuning out from this?

Julian Wallis, CEO at Intuji (00:22:32:10 – 00:22:58:06)
Right. Because we’ve had SEO go to these and people somewhat became, I guess, desensitized to SEO early in the early days, because people were gaming the system. Even still, there was SEO, you can still rank differently, but Google got much better at recommending the right stuff, right? But what happens when, like AI at scale is recommending products and people and then advertising becomes part of it, right?

Julian Wallis, CEO at Intuji (00:22:58:06 – 00:23:14:03)
It’s pay to play a game, right? Suddenly is not actually based on factual information. It’s just a pay to play it. Where does the point where people, you know, we’re on the rise at the moment, right? It’s going to peak and then we’re going to drop off because.

Julian Wallis, CEO at Intuji (00:23:14:05 – 00:23:31:07)
Not through capability, I don’t feel with like efficiency and everything. I think there’s going to be so much noise in general. There already is like so much noise of like it’s too much, if you know what I mean. Like it’s too much and I’m done with it kind of thing or, I don’t know, it just an interesting for me.

Julian Wallis, CEO at Intuji (00:23:31:07 – 00:23:46:02)
I feel that’s, that’s, that’s one of the parts of it. And I think the same thing is going to apply with like, if anyone can spin up a SaaS application, this is going to look like, there’s just going to be more crap, right?

Ritesh Shah, CTO at Intuji (00:23:46:04 – 00:24:13:18)
More, more crap. And it’s the same thing to do with YouTube content and everything these days its like it’s more crap. Only noise, only distraction. Initially I used to love like YouTube content. There was more knowledge, people speaking. Now all AI speaking. Now only the crap, only the craps. How I’m, its is not that right now I’m influenced by.

Ritesh Shah, CTO at Intuji (00:24:13:18 – 00:24:36:08)
It’s not that I’m not influenced. I get influenced with that. Right? But I now I think it’s oh, I’m wasting my time on this. All this crap is not value for me at the end of the day, it always comes back to PSV Thinking. If I’m where, am I, is it valuable or not for me? If it’s not valuable, I skip it?

Ritesh Shah, CTO at Intuji (00:24:36:09 – 00:25:01:02)
Right. Yeah. I was trying to give you an example of, like one thing that came to me is the YouTube content is these days I, they are scary, like very scary noise around AI coming, taking off your job, everything coming up, everything, but when I use it on a regular basis, I see it that it’s far away from there.

Ritesh Shah, CTO at Intuji (00:25:01:04 – 00:25:09:11)
They’re all free, Deep Seek was like, it has changed already.

Julian Wallis, CEO at Intuji (00:25:09:12 – 00:25:15:03)
Yeah, it was over the. Everyone’s already forgotten about it. It lasted two weeks and then.

Ritesh Shah, CTO at Intuji (00:25:15:03 – 00:25:34:05)
Deepseek is like, What now? Nothing. No ChatGPT is gone and Deepseek is there? Everyone is like. But now when I when I did try out Deepseek for 1 or 2 days, then, what is it? It’s not even worth it trying. So.

Julian Wallis, CEO at Intuji (00:25:34:07 – 00:25:48:09)
Yeah, it’s just the tech world has a very short memory, I guess is the way to put it. And very, one thing you got to keep in mind, this is what a lot of people don’t talk about, is so much of this, and we’re getting well off our original topic. It’s all right, a good discussion I think.

Julian Wallis, CEO at Intuji (00:25:48:09 – 00:26:06:10)
But it’s, so much of this is backed by investment and stuff like that. They’re like, people need to keep the investment prices up and know, like the stock prices up, the investment going. And so it’s all. As much as people probably don’t want to admit it, I’m not saying there’s hype, there’s not hype there. I’m a big advocate.

Julian Wallis, CEO at Intuji (00:26:06:10 – 00:26:25:21)
I know both of us are. But there’s a difference between misleading hype and hype. And like the reality of the situation is, there’s no point making it out to be something that’s not. It’s very good in certain situations where we can use it. But all of like open AI, anthropic, everyone have got raises they’re doing their right? They’re all trying to get find more funding.

Julian Wallis, CEO at Intuji (00:26:25:21 – 00:26:48:10)
Right. They’re all trying to get more investment. There’s a reason there’s PR for days. And then the pumping out, like sponsorships with influencers and just so much so, even Lovable, as you can see that, of LinkedIn they’re doing heaps of stuff with all these influencers. Paid posts and everything. Because they’re trying to ride the train. Obviously they’re trying to ride.

Julian Wallis, CEO at Intuji (00:26:48:10 – 00:27:08:06)
They’re getting the investment, make bank while it’s while it’s, raining, I guess. But the point of it is, is that. Let’s come back to reality for a bit and my, I’ve always maintained. I know you have as well, but like, it’s the people that use AI that are the ones that are valuable, not people that are going to be replaced by AI.

Julian Wallis, CEO at Intuji (00:27:08:08 – 00:27:31:01)
Right? It’s the people that use it incredibly well, or the people that are going to be in, not just to generate some emails or whatever, its not the point. But then we should have as a business, fundamentally, systems and a team that use AI brilliantly and systems that provide them the ability to use AI within guardrails that represent the company, I guess is a simple way to put it.

Ritesh Shah, CTO at Intuji (00:27:31:03 – 00:27:56:16)
Just a message to our customer, whoever is listening, all the listeners, basically AI, it’s just a tool. It’s not a brain. Let’s not take it as a brain. AI can do the research, bring it to your table, but you are the one to make the decision and then select what is right or what is wrong and then select basically, properly select is the right word I guess.

Ritesh Shah, CTO at Intuji (00:27:56:21 – 00:28:27:02)
And then with, also the developer and everything, developer, anyone today is anyone can write in their LinkedIn job profile. A developer I will accept that, but I won’t accept engineer. Engineer is a different thing than being a developer. You can write code. Writing code is more just, skill, like ten years, 15 years, 20 years back, it was a bigger skill.

Ritesh Shah, CTO at Intuji (00:28:27:02 – 00:28:46:00)
You, you need to learn a lot of things to write. But this has been solved. This issue has been solved by AI. I get that, I accept that, AI can write code. I can’t deny that. But what is the quality of the code, it’s the engineers decision.

Julian Wallis, CEO at Intuji (00:28:46:01 – 00:29:10:17)
Yeah, I think in general it’s not a replacement for critical thinking. And that could be content, code, database, whatever it is. Right. It could be anything. But it’s not a replacement for critical thinking. And when, you can tell when someone uses a as a replacement for critical thinking. And yeah, that to me is the biggest part of what we’ve discussed.

Julian Wallis, CEO at Intuji (00:29:10:19 – 00:29:21:03)
And that’s that’s the important part of when we’re architecting products or we’re building products, applications, whatever we want with with this technology. It’s the same principle applies.

Ritesh Shah, CTO at Intuji (00:29:21:09 – 00:29:47:15)
With, dealing with digital transformation, doing digital, always building some products here, always building some products. So, one of the thing is like, we always keep in mind that PSV Thinking. We are like profit, scale, value. There is nothing like that. Then there’s nothing, but with programing I see a lot, a lot of variables that can be missed out by AI as well.

Ritesh Shah, CTO at Intuji (00:29:47:15 – 00:30:24:18)
There has to be security, scalability, maintainability, reliability. Everything is there. It’s not just, documenting is like one of the key, key, key things. We’ve been a lot these days there’s been due to AI, I don’t know, I should see the, the graph around how GitHub’s data size has improved, increased because, AI is generating code, a lot of data, a lot of code is going to be pushed to GitHub for sure.

Ritesh Shah, CTO at Intuji (00:30:24:18 – 00:30:42:19)
The data size has to be increased for sure, but those codes lack the proper documentation, proper maintainability, proper architecture to it. Everyone is developer these days so there is nothing engineered.

Julian Wallis, CEO at Intuji (00:30:42:21 – 00:31:00:15)
It’s not to say. Obviously there are people who are using AI, you know, and they’ve done well, like they. But again, it’s where the critical thinking, the skills and everything, they’re using it as an assistant, as an augmentation rather than a replacement. That’s exactly what you’re saying is a very, very important part of the piece. And I think one thing that needs to be noted as well.

Julian Wallis, CEO at Intuji (00:31:00:15 – 00:31:29:00)
Let’s take, for example, the bigger something gets, the more complex it is, the more AI is going to struggle with it. Right? Because it might perfectly document and do the security for a small script or an app or small thing, whatever, like a very small scale thing. As you start adding very complex, you know, you have 15 integrations with third parties, microservices, all different databases, all very complex, different stakeholders, different experiences, whatever, whatever.

Julian Wallis, CEO at Intuji (00:31:29:02 – 00:31:58:04)
Yes. It’s just gonna obscure details. At the moment, the context windows are very small in general. You know, taking into account all of that, I’d actually, and it just popped into my head then. I’d love to know what like the context window is of your brain and what you can more the point of all the connections. Right. And I think, I’ve seen this before in like at, in like a stakeholder communication thing where, like when there’s, so many stakeholders, there’s so much more to go wrong.

Julian Wallis, CEO at Intuji (00:31:58:06 – 00:32:23:19)
Right. If you’ve got a 1 to 1 relationship? It’s so much easier. There’s some, it’s absurd the amount of, when the dependencies of all the different pieces, the amount of dependencies and possibilities goes massive. Like it’s huge. Its the same principle here, I think, is that again, for the routine tasks, to a degree, to the, the more straightforward things and yes, absolutely it’s easy.

Julian Wallis, CEO at Intuji (00:32:23:19 – 00:32:32:18)
You should be using it. But for the biggest stuff it’s like, it’s an assistant, it’s a tool. As you said, it’s not a replacement for critical thinking.

Ritesh Shah, CTO at Intuji (00:32:32:20 – 00:33:00:20)
Yeah. Take it as a library research tool, its great. It does a good job. It gives you ideas and all that, but you can’t. You basically, you understand your business properly. You understand your customer properly. You understand your legacy, your everything. You understand, your, AI won’t to understand all those aspects. And the solution will be very, very common.

Ritesh Shah, CTO at Intuji (00:33:00:22 – 00:33:12:14)
Everyone has the same solution. There’s one be a uniqueness. There won’t be an identity. Everyone will have the same thing and which, which will won’t be valuable anyways.

Julian Wallis, CEO at Intuji (00:33:12:16 – 00:33:36:16)
Yeah, correct. Because it’s, it all becomes the expectation. It becomes the baseline. Right. And I think it’s important to say that AI’s obviously developing quickly. We’re learning just as much as everyone else. But it’s just important to keep these things in mind that we might be having a very different conversation in five years time. But it’s like, at this point in time, this is what we want to, you know, want to be making sure that we’re moving forward with we want to make sure we keep these things in mind.

Julian Wallis, CEO at Intuji (00:33:36:16 – 00:33:41:01)
And if we do, yeah, I think everyone will get the best out of it.

Ritesh Shah, CTO at Intuji (00:33:41:03 – 00:33:51:06)
Thanks, guys for joining in. Thanks for listening to us. And this was a very good conversation. We enjoyed it. I hope you guys will enjoy it as well. Thanks, Julian for joining in.

Julian Wallis, CEO at Intuji (00:33:51:08 – 00:33:53:08)
Thanks. Appreciate it. Thank you.

Ritesh Shah, CTO at Intuji (00:33:53:10 – 00:34:22:16)
Thank you. That’s a wrap on this episode of The Disruptors. If you are serious about unlocking your EdgeFactor®, the thing that sets your business apart and helps it grow sustainably. Don’t forget to follow, share and leave a quick review. We will be back next week with more real-world insights and no fluff conversations on how to rethink your systems, modernise operations, and build value that lasts.

Ritesh Shah, CTO at Intuji (00:34:22:18 – 00:34:26:21)
Until then, stay safe, stay focused, and keep disrupting!