Subscribe
Share
Share
Embed
The Value Creation Mindset explores the decisions successful leaders make to create real, lasting value for their customers, teams, and businesses.
Hosted by A.J. Singh, each episode features candid conversations with founders, CEOs, and builders who have been in the trenches, unpacking hard-earned lessons on leadership, technology, strategy, and execution. No hype. No shortcuts. Just clear thinking, first principles, and practical insight from people who have actually built things that work.
00:00:07:24 - 00:00:38:10
Unknown
Hi. I made you sing. Welcome to the Value Creation Mindset, the podcast that explores the principles behind the decisions successful business leaders make to create value for their customers and themselves. Today, I'm joined by Al Katz, the CEO of Priority Software and a serial product builder and entrepreneur who has built commercial products, platforms and companies over the years.
00:00:38:12 - 00:01:05:18
Unknown
Yeah, welcome. Hi. Nice to be here. And it's a real pleasure. And thanks. You know, as, Stephen Covey says, you know, begin with the end in mind. And I like to kind of explore the beginning and the end of everyone's journey. We both earned our gray hair. Seen quite a bit. But tell me something.
00:01:05:18 - 00:01:16:07
Unknown
You know. Being in software, working with computers, writing code.
00:01:16:09 - 00:01:39:25
Unknown
There's a certain electricity to it that you learn when you're very young. You you can in. What's unique about it is you can have an idea of a thought. You see a problem, you can imagine a solution. And you you can build something with your own hands, sometimes in minutes, sometimes an hour or sometimes a day, sometimes in years.
00:01:39:27 - 00:02:07:18
Unknown
But you can see it from start to finish, and you can see how it works. You can see the result. You can feel that the value being created, it feels like electricity. And every, every engineer and developer that I've ever met knows what I'm talking about. I when did you first feel that? Well, we're talking a long time ago.
00:02:07:20 - 00:02:34:12
Unknown
So, I try to remember, I think, the moment someone told me, hey, I have I have this problem. They think you can solve it. And I said, yeah. Why not? Let's give it a try. And then I said, you know, behind the computer and said, okay, let's figure out what how. At the end of the day, I'm trying to see how I can bring value.
00:02:34:13 - 00:03:05:04
Unknown
And I said, first of all, first rule always say yes. And now the rule. Failure is not an option. So go like right off my guideline. Those are the two bounding boxes there I. Exactly. So, so the moment I got this challenge and then, you know, you have to, like, you have this enthusiasm that you have two that catches up with you, and you have to always reflect it to your customer or the people you're working with.
00:03:05:06 - 00:03:24:07
Unknown
Yeah. And bring him to be part of it. And the moment a part of it, you have the so it turns it in there. It's you're like it's contagious. You know, everybody's into it. And then once you solve the problem and you have this programing tools, you have, you know, the ability to solve a problem, you have the satisfaction.
00:03:24:07 - 00:03:46:13
Unknown
And you see everybody's looking up to you and telling you like, hey, he's a genius. How did he do that? How did you do that? Look, you solved the problem and you know, it's so this is like the energy, the opportunity. It's it's all around you. And and once you have, like, you know, you have this programing skills you build up and sure, I mean, you go around meeting customers, meeting, you know, companies.
00:03:46:13 - 00:04:05:28
Unknown
And it's like, we feel like, hey, I can solve everything, everybody's problem. I can solve this with, you know, with the software back then, I mean, there's no AI. So we had to, you know, basically program everything, but always yeah, I was like, I really delivered on the job training. You know, it's like, can you solve this? Sure.
00:04:05:28 - 00:04:33:05
Unknown
No problem. Then when we were like, oh, it's interesting where you I always say yes because my first job out of college, I was working as a manufacturing engineer at Motorola. I have a master's in mechanical engineering, of all things, and I got a reputation for, sitting in the staff meeting, and, someone would talk about a particular problem, and I'd.
00:04:33:07 - 00:04:55:00
Unknown
I'd always raise my hand. Hey, I can fix them. I can do that. And, you know, we all pay our dues. And I got to be a little of a grind, because that bike is the answer is, okay, great. You own that. I need it done by next week. And I was like, oh damn it. Now I'm working 60 70 hours a week.
00:04:55:00 - 00:05:22:13
Unknown
But again, when you're in your when you're in your 20s, you're paying your dues. And I don't know, I, I started out as just, manufacturing engineer in charge of a production line and I learned a lot of manufacturing. I mean, when when you're building real physical products, that if you get something wrong, they're going to cause a fire.
00:05:22:15 - 00:05:56:23
Unknown
Which which was a concern back in the day. With software, it's a little bit different. It shouldn't be because you're still building products, not just software. And I think sometimes engineers as well. More developers than than engineers or coders. Anybody who's like, everyone feels this electricity, you want to solve it, right? You always say yes because your mind's always spinning and you see an opportunity to do something more efficiently or save other people time or save you time.
00:05:56:24 - 00:06:21:00
Unknown
That's the that's the energy. That's the passion that I think some of us have. And you're also right. Other people see that, right? It's a little infectious, but you've got to temper it a little bit so that you have a little more predictability. And over the years we learn either the, you know, either the easy way or the hard way.
00:06:21:00 - 00:06:43:00
Unknown
And for me, it was definitely the hard way, kind of what you have to do to to harness that and to scale it, because there's only so much we can do. I'll, I'll ourselves. Right. I mean, sure, we have. I know that can help a single person do amazing things as long as they know what they're doing.
00:06:43:00 - 00:07:10:15
Unknown
I mean, that's still a requirement, but talk to me about how you what issues did you run into when you built on that success? And you kept saying yes, and you took on larger and larger challenges. How did you pick the people who you surrounded yourself with? How much control did you have over that? Or you just given a team?
00:07:10:17 - 00:07:38:11
Unknown
How do you struggle with that? Well, everything is a challenge, especially people. Oh my goodness. Yes. So I think it was like everything trial and error. I was like, you know, first of all, as you said, there's like this way of, you know, you're you having this project and you're getting more projects and said, okay, I can't be any more a one man show.
00:07:38:11 - 00:08:00:18
Unknown
I have to hire programmers to do the job. And then, you have to find people that are really self-sufficient and, they can think out of the box. So I had a lot of, programmers and, but it's a trial and error, you know, someone can make a writing session and the interview, but he knows everything.
00:08:00:18 - 00:08:25:15
Unknown
He knows, you know, from the, you know, academic point of view. He knows everything. No, people know everything. But then when it comes to be practical and solve a real problem, he gets stuck. It's like if he freezes up and you get. So I had a I had a lot of people come to work on me, and I changed a lot until I found this group of people that think the way I think, you know, the end of the day, you need to complete one another.
00:08:25:15 - 00:08:45:18
Unknown
It's like a marriage in some. So some sort of way, because then of a date, people grow them a lot, and most of the problems, you know, are being solved. I know in the coffee break in and I know when you eat supper. So, so this is the people you have to be comfortable with, the people you hire.
00:08:45:18 - 00:09:10:10
Unknown
You have to be, you know, you have to be in the same state of mind. And then, you know, it's like, things that, you know, in incorporates, you have like this product team and, and, you know, and documentation and R&D and everything it takes, I don't know, hundreds of meetings to, to generate something here. You know, you can think about it if, if we're in the same vibe with someone in the, you know, if I'm in the coffee break and they said, hey, look, let's solve this, this level.
00:09:10:13 - 00:09:32:09
Unknown
And we met after a few hours, okay? Already decided that let's combine the things together and that's it. A product was born. And so, yeah, it's, this is the way I used to work. I mean, this is what the. But back in the days when I still did projects. Yeah, yeah. Before projects. Very different than products. Completely different now I.
00:09:32:12 - 00:09:55:12
Unknown
You're absolutely right. You know, it's like the the root of an idea, a concept, a solution can come to you in moments. You or a couple of, a couple of your close colleagues, maybe you are. Maybe you are kicking it around. You've been working on, something. So you have something in your head. You can see it.
00:09:55:15 - 00:10:24:27
Unknown
You know, that's what needs to be done trying to express that to other team members. Get everybody on the same, on the same page, moving in the same direction. A little bit of a challenge. How which different sort of methodologies did you play with or toy with? How did you try to add structure to what you were doing to help you and your team be more effective?
00:10:24:29 - 00:10:47:01
Unknown
Well, it really depends in which phase because it was, as I said, total different when we're doing projects and then doing products and then doing platforms. So it's like, so I mean, it's in, the commercial space. So products, I mean, things that are going to generate revenue products and platforms and that and that and that and that in that particular mindset.
00:10:47:01 - 00:11:06:24
Unknown
Because that's yeah, that's that's my sole focus. So I think, when we went to the products page and again, it was a learning curve. We had no guidance on how to do things right. It was more of gut feelings. So we created our our own ways of, well, you know, if had the link, the effort.
00:11:06:27 - 00:11:25:27
Unknown
So it started. I was like trying to figure out, who's doing what and how to collaborate, starting looking out at what, you know, solutions are out there of, you know, managing code and source. And back in the days, there weren't too many options. I mean, usually had like, I don't know, Microsoft went with we used visual tools.
00:11:26:00 - 00:11:55:02
Unknown
Yeah. Was that, how was it called Tim? Something. Oh, yeah. Tim Foundation something. Yeah. Game Foundation server. TFS. That's what they grafted. And that was a successor to Visual Source Safe. Exactly. Yeah. No. So so I, I grew up there. Sure. But I think the way I work until today is most of the time we're not like, you know, sitting, planning.
00:11:55:09 - 00:12:21:11
Unknown
And then after a few weeks or months, coming out with the, with the solution. And then people posted the first time, we are very agile. So we even back then, well, I quit, you know, something and it you know into the waters. Give the customers, try to experience it, get the feedback and keep up evolving and evolving. So at the end of the day, I think it was a lot of, I mean, we didn't know what was waterfall, what was agile.
00:12:21:11 - 00:12:42:22
Unknown
We didn't know all this, you know, methodology without the logic of, of how to how to create a product. It was more gut feeling. It was plum. Look, I totally get that because, again, if you're if you're that close to the wire, you can feel like, you know, when you're creating value, when you're not, you can feel it.
00:12:42:25 - 00:13:13:00
Unknown
You know, when you're hooking up your, tires getting traction or just spinning your wheels. And sometimes you've got to try one thing and try another thing. And over the years, a lot of these concepts come and go. They have different names, but they're the same things just kind of repackaged. So, you know, on on one side you've you've got just extremely agile, which is a bit of a euphemism.
00:13:13:02 - 00:13:35:26
Unknown
I was talking to an engineer, a few years ago. He was working on firmware for an IoT product, and, we were handling all of the, all the back and platform stuff. So I was talking to to this guy who was based in the UK. It was based in the UK. At that point. It was a 1 or 2 man show, which is all good.
00:13:35:29 - 00:14:07:29
Unknown
But as we were talking to him, I asked him, you know, what methodology are you using? I know this might be you, but what's your mindset? What do you you know, how are you approaching this? And he said, oh, I'm extremely agile. And I said, so you mean you're just making it up as you go along, you know, because even if it's just a one person show, still a person's responsibility to bring some kind of structure to what you were, to what you're doing in some kind of discipline to what you're doing.
00:14:08:03 - 00:14:33:27
Unknown
But I, I, and and then on the other extreme, you've got the, you know, classic waterfall spends six months in design. By the time the design is done, the market has moved on. That's not what they need anymore. So like with most things, especially today, I mean, today with the changes in AI, I mean, I've done the last two weeks things that would take years.
00:14:33:27 - 00:14:57:11
Unknown
I'm telling you, years I, I created like four different products that I like. I had something that was written a while ago, this system that connects outlook, and the file system to all of Europe and, and I had to and I wanted to write it and I had like three front end, you know, outlook client, Gmail client and, and administrative.
00:14:57:13 - 00:15:38:04
Unknown
And I said, generate it to multi-tenant, convert it from Microsoft dot. Net C-sharp to Python back end. And I mean, I and I pushed it to like, I know it's, you know, compile it to a container and, and push it to the cloud. I did things that would take months or years just to understand this concept. And, I mean, it's one of the thing and, and I realize other stuff, you know, that like, in this whole, realm of AI, the MSP concept I saw, I understand it this week that people got it worm and that there's, there's some a big, problem over there.
00:15:38:06 - 00:16:12:21
Unknown
It's like, I'm calling it the double dipping problem because you have, like, would you explain what MCP is and then get into this? Yeah. So MCP is, it's by anthropic. It's, they, started a year ago. It's called model context protocol. It's kind of giving you an API layer for AI. So the moment you have this, ChatGPT here or later Gemini that wants to call your service.
00:16:12:24 - 00:16:44:15
Unknown
So the way you do it, besides giving it an API layer, a backend layer, you're surrounding it with this MCP layer that gives the ability for an Lem for a large language module to talk to your API. So call to talk to your tools. And the thing I figured out, for example, that, at the end of the day, this MCP is is an is an A, it's an all of them that it's typical and it's a and the other side you have your let's say reasoning module.
00:16:44:21 - 00:17:07:04
Unknown
So what I'm calling a double handling because at the end of the day I created like this knowledge base that you can query. So I'm already building your the answer to you. And I'm giving it back to this MCP client ChatGPT or whatever. That again retake takes you through your answer and re tries to build it. How he thinks fit is fitted for your the users query.
00:17:07:08 - 00:17:31:27
Unknown
So that's what I call double handling. So you're losing I mean it's it's costly and you're losing a lot of time because you're building the answer back on your own. You're not you're not just losing time. You are like, this is an engineering. What you call a tolerance stack up. None of these things are perfectly accurate. Neither are human beings.
00:17:32:00 - 00:18:12:09
Unknown
But the the the. If your accuracy on one side is 90%, the accuracy on the other side is 90%. You put them together, your accuracy is less than that. So or the the overall the overall plus minus doubles everything stacks up. So that's a very interesting finding. I mean, how I mean, this is part of the whole a genetic model fundamentally to kind of build a core service, make it, findable, make it easy for other agents to access, data and services that you are providing.
00:18:12:09 - 00:18:41:07
Unknown
And at this point, I mean, lives have been commoditized. They're everywhere. You can scale them up, you can scale them down. They are running on tiny little IoT edge devices, doing really amazing things. And they're running, in NASA data centers and in the pursuit of AGI. So, on that, on that, on that and on that front in this kind of.
00:18:41:10 - 00:19:12:15
Unknown
Swimming in this mix and driven by this, kind of, this, this electricity, which I, I can feel it coming through all of this tech here. I can see how it said you are, about this, what are, say, the 2 or 3 biggest lessons you've learned recently about how, how best to think about LMS and these various AI models and this new infrastructure.
00:19:12:18 - 00:19:35:15
Unknown
What's the right way to think about it? What's the analogies that you use to understand it better? Walk. Walk can help, get people past that initial fear curve in into actual actually utilizing it effectively. It's not just a promise, but the reality. I mean, you know, so, so if you could talk about that, that would be great.
00:19:35:17 - 00:20:02:27
Unknown
So first of all, there's no secret sauce. I don't believe there's no single answer. A single one. I thought you were the secret. Does not yet. I'm. So the thing is what I believe in. At least in there for enterprise. The biggest problem with, I think with AI is there is that there probabilistic? It's not, it's as I said, it could be 90%, but it's not 100%.
00:20:02:27 - 00:20:24:03
Unknown
And what I'm trying to do is, to get gain trust and see it. Let people see that it's giving them always the precise answer. And that's 90% of the time, because 10% of the time feels good. If it's not a good answer, it depends. If it's on tabular data, I can take, decision but is based on something that is finance decision.
00:20:24:05 - 00:20:49:06
Unknown
It could go sideways pretty fast. So yeah, what I'm trying to do is, first of all, work. As I said from the beginning, work with people, find the problem and from there build the solution. Then I'm I'm leaning to try to do everything deterministic. What I'm trying to do is, you know, everybody's running building agents and I'm saying, no, this is not the way.
00:20:49:06 - 00:21:10:16
Unknown
Because again, you're going to get the agent to 90%. But what happens with the 10%? So now what I'm trying to do is try and convert everything to be more deterministic. I'm using AI for for building the structure like I just. And this is another thing I did this week. I created the system that interviews you. I say I have, I have a goal.
00:21:10:16 - 00:21:34:14
Unknown
I want to, I don't know, do an onboarding for a customer, onboarding for suppliers, onboarding for a new employee. So I created a system that interviews view and and even building a prompt to complicated. So you're not building a problem. It's I'm actually interviewing you and I'm asking you question it. And by the answer I'm kind of, find it, figure out what to be the next, what with the next question.
00:21:34:14 - 00:21:53:08
Unknown
And then I'm building this plan. And this plan is the deterministic, plan of, of, scripts or code that needs to be run. And this plan is built from it's actually four type of plan. I've been in this the business problem, what you want to do, how it's going to achieve it. And this is a very okay.
00:21:53:08 - 00:22:14:09
Unknown
This is very detailed. And this is just for a few questions. And then I'm building this validation plan what it needs to validate then data gathering plan, what kind of data needs to gather from the from the systems. And then an execution plan. But everything on the plans is deterministic. No AI's building the plan is with AI running.
00:22:14:09 - 00:22:34:22
Unknown
The plan is executing code. And and I did what we called human in the loop. So what is human in the loop? Each time I hit a problem, I'm asking you. You're the special. You're the expertise the person is in charge is. I'm asking you, how do I solve the problem? And you just write in free text, oh, you need to do this or you need to do that.
00:22:34:29 - 00:22:59:16
Unknown
And this will regenerate the plan. So again AI is there for generating the plans. It's there for handling the problems. And it's and it's therefore learning from multiple use cases that you generate. It kind of fixes how to generate this main problem. That is the they interview the interviewer a problem. So I think at the end of the day you get again AI that is with you.
00:22:59:16 - 00:23:11:25
Unknown
But for the for the repetitive work, it's deterministic. I think that is a phenomenal insight. I,
00:23:11:28 - 00:23:36:27
Unknown
I wish I can do. I was talking to one of my best one or best friends who works at who works who works at Google. Also mechanical engineer, but he stayed a mechanical engineer. I kind of left them. I been in this office space and he's following much the same idea. Meaning use AI to generate the starting point code base.
00:23:36:27 - 00:24:08:17
Unknown
And now your interview style mechanism is really fascinating to me because you are concentrating each of these decisions into questions, prompting so much so, so the the, the use the probabilistic model to build a deterministic engine validated by the actual real life human being who knows rules and policies and, and and so there's hope for us yet we're not we're not obsolete.
00:24:08:20 - 00:24:35:05
Unknown
Quite not quite yet. And I'm sure here is that oh, that pattern makes total sense to me as a way of extracting huge value out of AI. Yeah. The biggest problem is that everybody is going the wrong way. Everybody is building agents and sub agents and letting the agents build the guy. Next time it's look, think about it, each time it's regenerating everything, it's trying.
00:24:35:08 - 00:25:01:04
Unknown
It could work 90% of the time. But what happens is and 10% of the time and I'm saying, well, it's kind of all this needs is more GPUs and more tokens. It's an excellent business model for you for the IT. Yeah. For the AI companies. Yeah. Yeah, yeah. No I mean I when when software went into the SAS space everybody wants a subscription.
00:25:01:04 - 00:25:35:16
Unknown
Everyone wants recurring revenue that drives increased multiples when when you exit and make sense and effectively the software space copied the, the the financial model of the cable companies and cell phone providers, the most loved companies on the planet, right. But I at at the same at the same time, though.
00:25:35:19 - 00:25:51:13
Unknown
I like it as long as there isn't the long term lock in or the lock in is because you're delivering intrinsic value to your customers. The the the.
00:25:51:16 - 00:26:27:12
Unknown
It this is if you're an engineer, if you're a developer out there, and many of them are scared, I many of them are small. Hey, this is going to rip with this. This is going to replace me. And I got this feedback. But we've been at this now for 26 years where when we built version one of our tech platform, which again, we've been in the code generation, the model driven automation space for this entire time.
00:26:27:12 - 00:26:56:08
Unknown
And we continue to be. But when I would go out and do demos, 20, 25 years ago, years ago, where we could we we could generate huge amount of code in a fully deterministic way, the completely trustworthy way, which that hasn't changed right now. We are doing this now. Read about 80 to 85% on there. But people would freak out saying this is going to replace me.
00:26:56:11 - 00:27:21:28
Unknown
And, and and my comment was, oh, is that. No, no it's not. Listen, it's your job to create as much value as you can for your customers, for your company, for yourself or your family. It's all flow. So if you can I mean, you know, it's it's like, oh, it's a little bit of analogy, but I think it still fits.
00:27:22:00 - 00:27:45:09
Unknown
Before the world of power tools, if your job was to put screws and things you you had to do it by hand. I comes along now all of a sudden you have a power screwdriver. Now you can drive screws faster and more reliably than ever before. You might be 4 or 5 six times as productive. Now it's not going to tell you where to put the screws.
00:27:45:11 - 00:28:15:07
Unknown
It's not going to tell you where not to put the screws. You can ask it, but you're asking for for can trouble there. But seeing it as a force multiplier that that increases the value that you as an individual can bring to bring to your business or a force multiplier that, if applied correctly, can help you and your and your company and your colleagues punch above their weight class.
00:28:15:09 - 00:28:47:09
Unknown
That's more job security, not less. So this kind of I mean, it's no different, is it that then the fear of robots and manufacturing, right. Oh no, I'm going to be put out of work. The things you're going to change. But there's something called human progress here. When is it good to be less efficient? So when I hear, Galaxy brains like Andrew Yang speak out against automation like it's an evil thing.
00:28:47:12 - 00:29:20:00
Unknown
So what? That tells me is, all right, you've got to get from one point to another there are two possible paths. One which is more efficient, less costly. Right. And, and, and and if it gets you to the same point, why would you not choose that? Right. And and there are some people who call them Luddites. I'm not really sure what, but they're dead set against that improvement in efficiency.
00:29:20:03 - 00:29:52:25
Unknown
And I know I have like, this really negative reaction to any of that stuff. Yeah, I think today, first of all, the the job, the way we work with the, the way we consume data is changing and changing rapidly. First of all, everybody's afraid of change. So you have to learn new skills if you want to be. I mean, to take advantage of this huge tools that are out there to make your job much more efficient.
00:29:52:27 - 00:30:11:17
Unknown
You need new skills. Think about it first, you have to be an expert on the I mean, in the space you're working. And then you had to learn. And there are all kinds of software tools to use to get to where you want to get to by, you know, learning multiple screens and learning. I don't know how to program them.
00:30:11:17 - 00:30:38:17
Unknown
And you have to be like a sophisticated user, maybe some coding expertise or whatever. Now everything is becoming conversational, so you need to know how to ask the right questions to get, you know, think about it's conversational and and, and some of the stuff is becoming autonomous. So and with the human in the loop. So you can be more focused on, let's say think about it, you could do the same job and get ten times more.
00:30:38:20 - 00:30:54:08
Unknown
So at the end of the day, companies are going to be much more. If it's all about scale, you know, and it's going to bring scale. I think at the end of day, people should, you know, think how they can use it to scale up. At the end of the day, you could do the same job. Why not?
00:30:54:08 - 00:31:15:03
Unknown
You can serve more customers in less times. You can be much more precise. You can be much more. Let's say the, the quality of the stuff. You're sending out a role focused on the business and less about the heavy lifting. So think about you're putting the heavy lifting to the AI, and you're now everything is more scalable and you're just, doing a better job.
00:31:15:06 - 00:31:20:04
Unknown
It's it's the.
00:31:20:06 - 00:31:23:04
Unknown
You're right.
00:31:23:06 - 00:31:38:07
Unknown
One of the challenges there is how do you if you're starting out, if you're a young person who wants to get into software engineering?
00:31:38:09 - 00:32:09:29
Unknown
The entry level jobs are fewer and far between. Because one thing that is definitely needed is I agree with you. You need a human in the loop. LMS are are inherently non-deterministic. It is like Forrest Gump. They're like a box of chocolates. You never know what you're going to get. You need. But but the the pilot of these AI systems needs to be pretty experienced.
00:32:10:01 - 00:32:58:18
Unknown
Getting them that experience is, well, is difficult. I mean, growing people, growing engineers into architects, developers into engineers. Right? Getting people to think at a higher level of of abstraction, the whole people side of it is, is, is, as you were entering earlier, is way harder than the tech side. And what have you found me like? What has been some guiding principles that that you've latched on to, to take people under your wing and help them grow to, to see this, to take advantage of it, to think, think and work more like you.
00:32:58:21 - 00:33:30:14
Unknown
Well, I'm still struggling over there with people. I mean, I had, the thing is, I'm trying to think, first of all, you know, a lot of people that came for programing, it's a shift in paradigm. I mean, it's like you, first of all, they need to understand the way I works because they're they're built on this concept of if then else everything the way people think, at least in people that have a pure is an experience.
00:33:30:14 - 00:33:55:00
Unknown
Everything is built on state machine. And now you have to understand that there's a different way of doing so. You have to understand how to trust, where to trust and where to entrust where you can, you know, let the let de I take control and want to take control back. So what I'm trying always to show them is, first of all, always with examples, you know, throw them in deep water.
00:33:55:03 - 00:34:16:14
Unknown
So them I mean, how are you going to solve it and show them, you know, how to how to think in this new way and, and and give them real life examples. I really believe in real life examples. You know, and and try to go deep waters as fast as possible. This is the only way to learn. I mean, as much as I can educate them, it's theoretical.
00:34:16:16 - 00:34:41:00
Unknown
The only way is to throw them hard problems. Some of them drown, but some of them, you know, overcome and become better. But as I said, there's no secret sauce. There's no, golden rule. It's it's basically like an I. Yeah. You know, sometimes you win, sometimes you lose. But remember what I said? Failure is not an option.
00:34:41:03 - 00:35:15:01
Unknown
If one door closes open the other one if you get thrown out of the door, climb back from the window. You don't strike me as someone who gives up easy. No, I you can't. I mean, you know, it's it's it's like I my wife sometimes tells me that I can be exhausting, like I you're you're right. I think I can be because again, I, we latch on to something and we pursue it and we try a number of different things.
00:35:15:01 - 00:35:36:19
Unknown
And the pursuit might take a very long time. Exactly. But I give up. I say, I think this is the secret. If you want the golden rule, never give up. The thing is, people give up very fast. They don't have the patience. Sometimes it takes weeks, you know, I can think of a problem. And I had I hit this wall and I said, okay, I tried this and it didn't work.
00:35:36:19 - 00:35:56:22
Unknown
And then you have this sort of a, you know, this, like, it comes pop pops into your mind, and we go in the shower or whatever in the supermarket, and then you have this breakthrough. But the thing is, people give up pretty fast and especially the day when everything is injected to them with everything has been summarized, everything has been chewed up for them.
00:35:57:00 - 00:36:14:27
Unknown
People. They don't have the patience. I think maybe, the patience and keep on trying. This is the only way that, I mean, this is the secret sauce. Maybe it. I tell you, when you said that, it really got to me because,
00:36:15:00 - 00:36:38:22
Unknown
My dad, my dad got me into software development. He got me into programing when I was a kid. All I wanted was an Atari 2600. My friends at home, I just wanted to play video games. I kept on asking that, no, no, no no no. And one day, wasn't even my birthday. He comes home with a Commodore Vic 20, and he teaches himself how to program.
00:36:38:22 - 00:37:05:19
Unknown
And he's also a mechanical engineer and he, you know, and he gets me a subscription to Computer Gazette magazine, which has printouts of all this basic code. So I'm sitting there typing it all out. I had no way to store it. So if I turn the computer off, I lose it all. But reason I bring it up is the first thing that he did when he got it again is mechanical engineer.
00:37:05:21 - 00:37:32:02
Unknown
He he wrote a program to automate a set of engineering computations that his colleagues were doing by hand. What would take his colleagues two weeks to do. He could do in 20 minutes with the code that he wrote. Now, the only big challenge. Well, there were two challenges. One, he ran out of the five K of Ram, so he had to get the eight K expansion pack and two.
00:37:32:03 - 00:38:02:06
Unknown
He scared the bejesus out of his colleagues at the office. They did not understand. They didn't. It was it was showing them something that was far too scary. But that's how I got my start. So yeah, it's another scary thing. Okay, fine. But we know what the principles are, right? We have to study it. You have to have enough confidence in our own ability to separate the the what matters and what matters and what and what doesn't.
00:38:02:09 - 00:38:26:14
Unknown
That's funny. When you brought up the shower, because I swear I solve all my most difficult challenges in the bathroom, right? Beating my head against the wall for a week. I mean, this actually happened. I was out at a client trying to solve a very challenging problem. And with the team, we were in a war room working, you know, 60 hour days for a whole week.
00:38:26:16 - 00:38:59:29
Unknown
Couldn't quite get there, came home, took a shower Saturday morning, and the key thing hit me. Wrote the code over Sunday, tested it on Monday. It worked. And this is the this is the thing that saved my clients business. They had customers ready to walk. We saved. We bought them another 20 years with this effort. So, is this I have some but but again you say don't give up.
00:39:00:01 - 00:39:31:14
Unknown
Never give up. That's again I, I go back. These are words that my father told me and yes sir, I get a little emotional here because I in, in I he's he's he he's still with us. He's 86. If you ask him if he's, retired. He says no, I'm unemployed. That's, I had the same way, but about ten years ago, 11 years ago, he had a stroke and I dropped everything I sold out there.
00:39:31:14 - 00:40:03:00
Unknown
I was with him, and I, He's he's recovered. He's had a whole he's had a whole series of setbacks. Stroke, seizure. He broke his ribs, he got cancer. He overcame. He broke his hip. You get past all of it. But right after he had a stroke, he was. He couldn't walk in a wheelchair. After about a week in rehab, they suggested that we go to a support group for stroke survivors.
00:40:03:08 - 00:40:36:21
Unknown
So I took him there. We we we were sitting around a table with people who had survived strokes in many different ways. Most of them were way worse off than my father because my father to eat. He didn't have any cognitive damage. It was he was incredibly lucky. But as all of these people were sitting around and they were saying something, he came to my dad's, came to my dad's term, it was my dad's turn.
00:40:36:22 - 00:41:05:23
Unknown
And the first thing that he said to everybody is never give up. And so, you know, I we are whether we want to be or not, we are, we we are we are more like our parents than, than we ever thought we would be. But I'm I'm certainly grateful for both both my parents would, especially my father, for kind of instilling in me a couple of these things.
00:41:05:25 - 00:41:24:21
Unknown
I mean, we learn early on, right? So he always said that never give up. And then every conversation I had with him in college, when we ended the call, he would tell me to stay rational. These are timeless things, right? I mean, you know, I just I mean, I can totally relate. I had I did the very similar path.
00:41:24:21 - 00:41:46:23
Unknown
I mean, my father purchased me a single spectrum, you know, vacuum. And then I went on these and the Commodore 64 and, and, you know, this, I got yachts, you know, playing with games. But I need to, you know, you had to learn all these commands. And at the end of the day, I think I didn't play much games, but it was really interesting what to do with it.
00:41:46:23 - 00:42:10:27
Unknown
And then, I kind of left off the computers. And the next time I met with computers, when I was back in the military, so the thing is, I was this, the space that has that inventory and people were, you know, gathering data in papers and stock counting, stuff like that. And I said, hey, there's is a software computer program over here that does that.
00:42:10:29 - 00:42:33:23
Unknown
Nobody uses it. Let's work with it. Why do we have to work so hard, you know, why do I decide to be college friends and said, why? Let's use the computer. No, we can trust it. So I took by myself to study the programing language that they wrote the software. I just bought the book book of it, and I, I tried to, you know, improve the software.
00:42:33:23 - 00:42:49:09
Unknown
And then of the day I had to rewrite it. And this is how I got into computers. I just did that. I was lazy, I didn't want to go down. And the stock counting software, I, I said I could be solved with software. And then, you know, it became this big thing. Everybody came to see how it was.
00:42:49:11 - 00:43:08:03
Unknown
I this place was computerized and this is how I got into computers, really by mistake. It was like, you know, I tried to solve a problem for me, you know, I didn't want to go do the hard or the hard jobs. It looks the that the best engineers are a little lazy. You want them? Yeah.
00:43:08:05 - 00:43:32:01
Unknown
You don't want somebody, you know, it's it's like. Plus it gets boring doing the same thing over and over and over or over again, either either manually or otherwise. I mean, the only promise I've, I've ever made turning my people. And I will say, after 26 years, I've never broken this. The only promise I make is you will not be bored.
00:43:32:03 - 00:43:51:16
Unknown
Same here, I said. I'm always looking at cutting edge. This is the thing I used to tell people that I'm. You know, when I interview employees, I will tell one thing for sure here. You're not going to be bored because you always kind of get it in cutting edge. At least on the software industry. You don't have anyone to ask because it's cutting edge.
00:43:51:17 - 00:44:11:19
Unknown
You have to invent the wheel yourself. It's not. There's no best practices. And that's what I think. The people that worked for me really liked because, you know, it wasn't boring that it wasn't this repetitive job. It wasn't like the small thing they didn't see, you know, everybody that worked for me, they had to deal with, with databases, with security, with scale.
00:44:11:23 - 00:44:37:11
Unknown
Was talking to customers, feeling the, you know, the pressure of something that this is a company, something doesn't work. You know, a lot of phone calls. They feel they felt the pressure. It's something that most of the people that work in software don't feel. They have this army of support teams and product and cure and people that, you know, they don't they don't see the product, they don't see, they don't feel the pain and they don't feel the success.
00:44:37:14 - 00:45:05:27
Unknown
And I think, you absolutely right. I mean, it it can turn into kind of an open loop. I'm always advising clients, listen, expose your engineers to your customers, let them directly feel that the, the, customer's pains and the challenges and their desires. Because, like, I haven't met any any engineer who doesn't want to do good, who doesn't want to do a good job, who doesn't want to help the customer.
00:45:06:04 - 00:45:36:18
Unknown
But oftentimes they are kept insulated in the back room somewhere and surrounded by product management and support. And away from customer service. And so all of these insulating layers get in the way. So what I like to do is, well, I do this myself. I mean, one of our clients, we, we, we have built, the, the entire platform for a IoT solution.
00:45:36:18 - 00:46:00:15
Unknown
This is, very cool. Leak detection and prevention system. It's an electronic validate. It gets installed on the water main, and it detects leaks. It automatically turns the water off you. You can manage it. It it's talking to a back end platform app and all this stuff. So we built all of that stuff. But who, who, who touches it physically.
00:46:00:18 - 00:46:30:06
Unknown
Plumbers pumping out of customers. So, many of there are there are many plumbers who have my cell phone. And when they run into issues, they call me directly. And that's not a waste of my time. I love that every call I learn something. We feed that back into what we are doing so we can create more value so we can smooth things out and, I do the same thing with my people.
00:46:30:09 - 00:46:55:13
Unknown
It's it's not a a these are hard things, right? But too often the idea is, oh, they're a developer. You got a, you know, keep them in a closet and feed them under the door. And there are and there are and there are some engineers in the space who don't like to talk to, to customers who you kind of have to deal with like that.
00:46:55:15 - 00:47:19:00
Unknown
We don't hire those people. I mean, they have the place and everyone is born with certain talents and predilections, and that's all good. But if, if, if your point is to create value, how can you do that without making sure that that feedback loop is there not just for you as a leader, but for everyone in your in your team?
00:47:19:03 - 00:47:22:20
Unknown
I mean,
00:47:22:23 - 00:47:48:19
Unknown
Yeah, I guess you get what you're saying then of the day. That's that's before. I mean, how do I hire people? What type of people do I want? And this is exactly the people that they're in. They know they're interested in solving problems and pairing people and meeting the field, being in the field and meeting the understanding. Then they can celebrate the wins, because it's not only the the fall downs that they know they're the problem and they need to solve it.
00:47:48:19 - 00:48:10:21
Unknown
And would they have this connection with you, part of the product. They feel it's part of the they think about it when they go to sleep or how to solve problems, and they feel the pain with them when it doesn't work. But they can. But when it works and there's like, you know, this, you know, the when the customer like going live and stuff like that, they're partying with everybody.
00:48:10:24 - 00:48:49:11
Unknown
It's part of them. It's part of their success. It's a, this is the people. Yeah, yeah. No, I, I totally with you the, the one of our, one of our core values is take ownership. If you have a challenge, you have a product, you are engaged with the customer take ownership of whatever that is. And I've gotten feedback that I've gotten feedback from, other CEOs who we've been working with that they tell me, AJ, why do your why do your people seem to care more than my people?
00:48:49:14 - 00:49:13:00
Unknown
I'm like, I don't know, but I will take that as as positive feedback. Thank you. Compliment. Yeah. I'll, I'll take that. And remember, there's another thing that this a challenge with people. Remember I had this saying and excuse my French because it's not probably a good English grammar, but it's saying if it won't be simple, it simply won't be.
00:49:13:02 - 00:49:33:01
Unknown
And the problem with engineers that they want to make it simple for their side, not for the customer. So why do I why should I care if you ask, I don't know, move to ten screens and do ten configurations to make it work. I mean, from my point of view, it's good enough at the challenge. Another challenge them to always try to tell them, look, it has to be simple.
00:49:33:01 - 00:49:49:08
Unknown
At the end of the day, you have a truck driver, sales rep, people, someone in the field that now has a lot of, I know an elevator is stuck in someone's second job, and you have the the commission needs to figure out what to do. It has to be simple. It has to be something that is appealing.
00:49:49:10 - 00:50:13:28
Unknown
And this is another challenge with engineers because they want to solve the problem, but they're not thinking about the human factor. And this is another thing I tried to educate the AI people. Oh no, I, I'm, I totally with you there. I mean, engineers can sometimes fall into a, kind of a checkbox mentality. All right. Have have I checked the box?
00:50:13:28 - 00:50:46:12
Unknown
Yep. I've got my box checked. I'm good. Are you. When are you done? Right. You're building a future product, an app, whatever it might be. When are you done? How do you define that? What's the line? Right? And how do you know it's really going to kind of do that? Hey, one thing that drives me nuts and I would love to get your feedback on this, is everybody's calling themselves a full stack engineer now.
00:50:46:15 - 00:51:16:08
Unknown
And I'm sorry that 100% there's, so can can you do everything? Can you build end code and back and go? Sure. I'm sure, I'm sure you can. But you're going to be ten times worse at one of those and ten times better at something else. So we're not I mean, we have natural talents. You're not going to find all the skills and the insights and the instincts you need in one brain.
00:51:16:08 - 00:51:54:12
Unknown
It does not exist. So like for me, front end and back end engineer completely different UX designer versus UX engineer, different brain, different brain pathways. You need different people in in in the mix. Give me your perspective on this. What are the different types of players or get a natural intrinsic roles that the user needed in the space to kind of bring together and launch successful products?
00:51:54:14 - 00:52:22:01
Unknown
It depends on what time frame you're talking about, because today I'll give you a totally different story. Okay, if I was talking a few, I mean, today, today, I'm ashamed to say it, but today you can be a one man show with AI that bring give you the most beautiful user interfaces and the most, let's say secured and, and, and logged and tested backends.
00:52:22:04 - 00:52:46:11
Unknown
So this whole place of p of front and back and I could everything is becoming more spec driven. You have to I mean if you did full stack, if you, you know, I mean the basics of what is the backend and what is the farm and what are the tools that are commonly used, what are the pitfalls? It's enough today to be think, well, we called a spec driven architecture that the AI fills the gaps.
00:52:46:11 - 00:53:16:09
Unknown
So we have to be an expert. At the end of the day, I'm trying to think I this huge brains. It knows everything but doesn't have specialties. I mean, it doesn't have, specialization in and, you know, in, specific niche. And over there, this is where you need to fill the gaps. If today, if you know, in if you've done full stack and you know the concepts, if you can do a fully full on a fully blown product, it seems like it I don't know, it seems like it.
00:53:16:12 - 00:53:40:22
Unknown
But if you're talking one year ago, of course, you had. At the end of the day you had front end and you have 100 engineers. You have backend engineers, you have, someone that fills in the gap. But, I don't know. Today it's it's changed. And I told you I did something that, I created a product with, three different types of front end.
00:53:40:24 - 00:53:59:05
Unknown
And of course, I made I said, okay, this must been, you know, HTML and JavaScript and CSS and this. You've got to know you have to understand that, you know, what you're doing. But at the end of the day, I mean, it generated everything, and I know how to read what it did, and it did a pretty good job.
00:53:59:07 - 00:54:19:10
Unknown
And you know what? I told you that we did this, backend test that integrates a robotic warehouse to priority or system. Yeah. And I and I did it. I don't know, it took me an hour to generate the code and, with the documentation, and it's in production for, like, I don't know, almost a year now with zero problems.
00:54:19:13 - 00:54:33:22
Unknown
And, so, so I don't know, everything is changing. I mean, as I said, it's a totally different dancer between a year ago and today. Yeah. No, I things are definitely changing. The.
00:54:33:25 - 00:55:10:26
Unknown
So, so, couple of questions if the before we wrap up here. So what's your favorite code Ellen. What's your go to every week? It changes I start it off with, I mean, it started with, Vsco Dan, Celine Klein and what it's called extension. And then I went to windsurf and, and of course, you know, at the end of the day, on topic coding board, I mean, for coding are the best.
00:55:10:26 - 00:55:46:12
Unknown
I mean, the Gemini is good. They're all kind of other good ones. But Gemini are still at the top. I mean, they're they're leading the leaderboards in the encoding. So cloud code awesome tool. But now came Andy gravity of Adi gravity from Google. So I'm trying to find out the make sure. So I'm kind of using anti-gravity because it has this huge I mean, this great concept of, this interactive, agents that you can interact with the planet, it's not just generating, running, and then you just re running off, you're giving more instructions.
00:55:46:15 - 00:56:12:24
Unknown
You're actually interaction with the place. It's it's great for the planning and cloud code doing the coding. I mean, using both worlds mean this is my go to stack if you ask me today and and and I think, there's a lot of, heat coming now from programing language, you know, mostly Python, but now, you know, if you want to go, go for scale and, performance, maybe rust.
00:56:12:24 - 00:56:36:25
Unknown
Rust is the taking, is becoming, you know, the best thing and the best, the next shining tool in the boxes are how you call it. There's always a rotating mix, but, but every week it does. I mean, at the end of the day, you have to be agnostic because. Yeah, I think things you couldn't do last week could be done this week with, with this tool.
00:56:36:25 - 00:56:58:09
Unknown
So I'm all the time learning. I'm following a lot of YouTube. You know, people that are, you know, giving you the every week what's the latest and greatest and I'm testing it. So it's all the time you think what? You know, it works, but keep on testing what came out, because maybe it has something that can solve a problem that could be solved before.
00:56:58:09 - 00:57:16:28
Unknown
A week ago. And it's. The pace is like this fast. You know, my c always sell. I know what you're going to tell me in this, you know, this meeting. Remember what I told you last week? Forget about it. This is the problem is already solved this week. There's something new. This new tool. This is awesome. It solved this problem that we had last week.
00:57:16:28 - 00:57:42:09
Unknown
So this is like, every week. Well, you know, it's you. You seem to be having a lot of fun. You seem to have a lot of fun with this thing. I, I, you know, thank you for your time. I really appreciate your perspective. I appreciate you, I look forward to to meeting you one day in in person.
00:57:42:11 - 00:57:48:01
Unknown
Last, last question for you. You know.
00:57:48:03 - 00:58:12:18
Unknown
I believe our legacy will be measured by how well and how far we lift all those we touch. That's what I think about when I get up in the morning, and I'm using whatever talents and skills that I might have and that the the company has to lift up as many people as I can. That's that's what I believe will be our legacy.
00:58:12:20 - 00:58:28:01
Unknown
What do you want your legacy to be? I don't know, maybe just, people say, we'll say, you know, he did something ten, 20, 30 years ago and still working.
00:58:28:04 - 00:58:52:00
Unknown
But it's not you. You you build something that lasts. I. Well, yeah, but but I guess it it's it's unrealistic today, you know, at the pace of things that are changing. I mean people, I mean legacy people, old people like us, I have I, I, I am not a legacy person. I'm not going to I'm not going to go.
00:58:52:02 - 00:59:19:05
Unknown
I mean, all right, I mean, our age, I guess the around the, I don't know, 4050s, I've got to remember, remember people we lived in the 80s and I think, it's a totally different generation now. So I don't know what to expect from this generation, but it's coming. So change that much.
00:59:19:07 - 00:59:43:25
Unknown
Look, the people. What I was outdoors, I was, you know, I was having fun. I was reading French, but the words and even cell phones back then, we're talking to each other right? Today, my kids, I mean, they're sitting all day behind a computer. I mean, everything has been virtual, so we're talking about legacy. What do I mean, what is the word for people going to remember from you?
00:59:43:25 - 01:00:08:13
Unknown
Maybe your avatar, maybe your, I don't know, Facebook profile? What are they going to remember? I think well I, I you know, this this podcast is is going to be viewed on social media the same time I think social media should be burned to the ground. I'm not letting our daughter get anywhere close to it until she's 16.
01:00:08:15 - 01:00:45:13
Unknown
And like that, my fundamental premise is we are all the same and we are all connected whether we want to be or not. And if technology can, can help us connect. It certainly helped you and I connect. That's in that positive. But I also know that these these damn phones and social media apps and I mean, if you find yourself scrolling through anything, shut it off.
01:00:45:16 - 01:01:16:23
Unknown
Don't don't consume a feed. You know, that's like you're lining up at a trough and being fed whatever you might be given. I think this is a pendulum. It will swing back around. I, I do think, oh, you know, it's like for our kids, you I'd like to think this is a great time to be alive. There's more opportunities now for our kids than there were for us.
01:01:16:25 - 01:01:44:25
Unknown
We certainly had a lot less right and less technology, less time we up. You wrote our bikes, played outside, hung out with friends. And I remember just a few years ago, I was walking along the Delaware River in Philadelphia, and I saw all of these young people, maybe about 30 of them, just out of, college or seniors in high school or something about that.
01:01:44:25 - 01:02:01:05
Unknown
And, and they were all sitting on this, and, and, this bench area right next to the river on, on a beautiful summer day. These are, you know, boys, girls, mix.
01:02:01:07 - 01:02:26:01
Unknown
They weren't hanging out with each other. Every last one of them was stuck in their phone. And I remember and my daughter at the time was still in a stroller and was walking her along and, and I saw that I'm like, something is deeply wrong. They should be hitting on each other and hanging out and laughing and, doing something else, not just stuck to their devices.
01:02:26:03 - 01:02:53:15
Unknown
So the things will swing in the other direction. I, I think I'd like to I'd like to be hopeful, but, you know, it's it's, it's not a rom, as Donald Rumsfeld says. Right. You know, is this an unknown, unknown or known? Unknown? I'm not. So, yeah. So regarding, I mean, what kind of mark are you going to leave?
01:02:53:17 - 01:03:27:01
Unknown
Right. I mean, I don't know. Again, it's the thing I mean, I think everything we used to know is changing. I don't know, for better or for worse. So time will tell, I guess. Well, I mean, look, if you're always thinking about how you can create more, more value for others, our our, our customers, our employees, our children, it's to me that makes me feel like at least I'm on the right path.
01:03:27:03 - 01:03:54:02
Unknown
They're moving in a positive direction. If what I'm doing isn't creating any value for anyone. But there are lots of ways doing that, right. You can you can spend your entire life not not not focused on that. But, you know, the, two questions I always ask, right? Are are you in a good place and are you on the right path?
01:03:54:04 - 01:04:17:18
Unknown
And if you've if you can answer yes to just one of those questions, you should be able to sleep well at night. Because if you're in a good place but on the wrong path, you've got time to change. If you're on a bad place but you're on the right, right path, you know it's going to get better if you're in a bad place and on the wrong path.
01:04:17:21 - 01:04:40:07
Unknown
It can only get better. Well, not necessarily. I mean, that's the hard place you want to avoid, but it's. Let me ask you then, do you feel you're in a good place and on the right path? Well, you know, it depends if you're looking, you know, at the things you accomplish and you know, it's what the society's saying, that it's good.
01:04:40:09 - 01:05:02:11
Unknown
You can say, yes, you know, I, I got married, I have three children. I had my own businesses. I, I, I mean, there were acquired, so I got money. I mean, looks like, you know, everything is. But to tell you the truth, there's always this. You know, of this urge to to do more. It's like it's not.
01:05:02:12 - 01:05:19:00
Unknown
You're feeling good for yourself. It's not enough. I haven't I haven't made a any, you know, major Mark, I don't know who said if how people are going to remember me at all if they're going to remember me. And I said, you change people. I mean, people live with change. I helped a lot of people, you know, find jobs.
01:05:19:03 - 01:05:44:01
Unknown
Sure. Get out of, you know, of a bad place to go. But are they going to remember, you know, they're going to leave a mark. I mean, maybe even had the little effect on the industry, but then have the day, I don't know, I'm feeling. I mean, I'm happy I'm doing something that. It's interesting. I think I'm more in the place that I can, you know, be self educated and, and do the things I like to do.
01:05:44:01 - 01:06:06:20
Unknown
Another thing I must do. But if I think that, you know, if this is like, the right place, I don't know, maybe in another 20 words, I can. Oh, yeah. This was the right place. Or maybe I should have been somewhere else if I have it for now, I'm happy. Yeah. No, I, it there's the future is always is always uncertain.
01:06:06:21 - 01:06:35:08
Unknown
Anyone who says that they can look into it and tell you definitively how it's going to be is a liar, but I think the the what's clear to me is, dude, you still got a lot of music in you. So if you do something you like, you know, there's this. I think Tony Robbins said it back then. If you're doing something you like, you won't work a day in your life.
01:06:35:11 - 01:06:50:25
Unknown
So yeah, if you're doing if you like your job and then of the day most of the time is in the job, then, you know, you have what to get up to in the morning. I mean, there's the challenge to solve. So at the end of the day, you work. How do you call it? Work hard, play hard.
01:06:50:28 - 01:07:09:19
Unknown
So you work hard, and then you get the results and you can go, you know, have spend time with your family, go to the studio. Long vacations and do it and have fun. So as long as you're doing what you like and you have and you get it earned from that, so why not? Amen. All right. I'll.
01:07:09:20 - 01:07:20:12
Unknown
Hey, thank you for your time. Thanks for being on our, on our podcast. It's, really been a pleasure.
01:07:20:14 - 01:07:46:14
Unknown
Great to be here. And, I didn't tell you how I solved the same CPU problem. Well, tell me. So it's really easy. At the end of the day, as I said, there's, this, lamp that is calling me. And instead of building that answer, I build another API that only returns the raw material, the raw data with the prompt.
01:07:46:16 - 01:08:08:02
Unknown
It's a dynamic pump on how the L11, the client side should build that source. So actually the result is actions. So it's like I'm telling him the raw materials with instructions because I've said this is what I call the double dipping problem with MCP. Right. So I so I created so it's not just take the your API layer and wrap it with the MCP layer.
01:08:08:04 - 01:08:44:05
Unknown
Because then you're going to get the again you're going to get the performance ID the token the cost. And that elem can go and do something you didn't want to do. The, the I mean the MCP quiet this way. I mean I'm saving it works after two seconds. I'm not waiting appeal down. So it's been generated and so again this is just another thing that, you know, things technology is coming out, but people really don't understand where are the pitfalls and how to making, you know, behave the way you want you then how it could make it much more, performance efficient and, and this and then when the problem comes rises up, say
01:08:44:05 - 01:08:49:11
Unknown
up. It's there I this is how it works. I mean.
01:08:49:14 - 01:09:16:25
Unknown
It's I, it's no, no, I, I that's very smart I mean it it you are fundamentally reducing the uncertainty in the overall solution by taking this this is the challenge I said bringing taking what's good in AI for summarization, for context, understanding, for building code. But going back to the deterministic world, not letting an agent decide what to do and what to do and what to do.
01:09:16:27 - 01:09:44:15
Unknown
And this is that I think the the next thing about, enterprise, how you can leverage, AI with deterministic flows. And this is kind of off the track, the people just building multiple agents. And now we have skills and you have the cloud code SDK and they're evolving and evolving. But as long as there's let's say an L11, the middle could be an edge agent, could be just a, you know, just an alarm.
01:09:44:18 - 01:10:15:13
Unknown
You're it can always make a mistake. And this is what I've tried to shift back to this deterministic way of, working with a, with AI that that is a that is excellent pragmatic advice to our audience. Always combine AI with a strong deterministic element. It's not it's not going to do everything for you, but constraining it with real knowledge in a deterministic way is how you can get value out of it.
01:10:15:16 - 01:10:41:17
Unknown
I think the the hearing, you describe it the way that it's being done now, this agents everywhere talking to each other reminds me a lot of the metaverse. We all know how that turned out I think. Yeah. Was it 30 or $70 billion lit on fire or something like that? Yeah. But now looking at look at what's going on with open, I think about the billions of, dollars have been spent.
01:10:41:17 - 01:11:03:12
Unknown
I mean, at the end of the day, you see, now that you have to have multiple factors to make things work. So you have to you do I mean, it's not enough that you don't have enough. You need the you need the audience. You need the customers. You need the infrastructure. You need the the money. The good that can come not only from, you know, from selling, your LLN modules.
01:11:03:15 - 01:11:37:26
Unknown
And today, from my understanding, the only one that has all the ingredients is Google. That, like Microfiltration, they've got they have a vertically integrated stack. They do. And and I mean, you know, look in NI in any space and in any domain if you want, if you want to create is if you want to create value, you've got three core raw materials, you've got people, you've got technology and you've got capital.
01:11:37:28 - 01:12:10:07
Unknown
Everyone has access to some of that. They can make choices on who, what, what kind, etc. but to create value, you need effective leadership and a shared vision across the organization to focus those the those raw materials on a specific point so you can create value. And if you ask the question which player out there has the right people, the right technology, sufficient capital.
01:12:10:09 - 01:12:40:06
Unknown
But also a clear and shared vision and effective leadership, you need all of that in order to actually delivery it, know that actually delivery results. I, I who would you pick is closest to having that right mix in the Azure right now. As I said Google. Google I mean yeah Microsoft. But they don't have their own MPs, at least not frontier ones.
01:12:40:08 - 01:12:57:25
Unknown
You have AWS, they do have their own little MPs, but they're still lean. They're leaning on, Nvidia. But I think they just came out with their own. You, so they anthropic. I know they did. They, they, they have a tie up there as well. So so Google I mean they have their own they have the TPUs.
01:12:57:25 - 01:13:22:19
Unknown
So they have their own, infrastructure. They have the, the data. I mean, all the gmail.com. They're everything they have. They have the best fancier modules like Gemini and Nano Banana and all of them. So and they have the, the revenue that's coming out from all the other products. And, they have the roadmap of, being, you know, it's putting all they're talking about data centers in space.
01:13:22:19 - 01:13:46:09
Unknown
Now, note that to be much more efficient. So, the, the economics of that love to see. And they said the moment the cost of the one year, 1200 dollars, I think for a kilo to $200 a kilo to send, stuff to space. This is. Yeah, for the land. This is the, this is the turning point that they're going to.
01:13:46:12 - 01:14:06:06
Unknown
It's going to they think they're going to achieve it in the next few years. They're talking about 2035. That is going to be available. But I think it's going to be less. Well let's see. Will that be before or after we have self-driving cars? I mean we already have so very cars. They're just taking some time decisions that are not supposed to be taken.
01:14:06:13 - 01:14:38:05
Unknown
Yeah. No they drive they drive themselves right into other vehicles. Yeah. No. Exactly. But we do too. I mean, is going to make more accidents. We are they I you know what this is the question of theory versus reality. And, well, reality these days, with everyone stuck to their devices, it's pretty easy to manipulate. But the laws of physics don't care.
01:14:38:07 - 01:15:01:13
Unknown
And things like viruses don't care what our beliefs are is whatever that is, is, is well, and again, as as business people, as engineers, as people were trying to really create value. We have to be at the base our decisions on reality, reality, but not not the promise. That's fine. We all get that right, that that's what we're chasing.
01:15:01:15 - 01:15:34:23
Unknown
But the tempered with some hard, hard reality. And I think, it's the none of that is the sexy stuff, right. Okay. So are we doing this? Great. What are the economics of it? How profitable is this? Is this going to be. What's the risk? What's the risk we might be taking on kind of doing this sort of like, when when we've got, we have clients who might have, legacy product and they want to do want to modernize it.
01:15:34:27 - 01:16:04:00
Unknown
I mean, they may want to rewrite it. There's always some kind of urge. So they light up a, next gen project, which goes on for years and consumes millions of dollars. And while you're promising customers, we're going to handle that, and next gen, we're going to handle that and next in that, in that next gen reality is that next gen almost them comes, is that that the create the most value generally not.
01:16:04:02 - 01:16:33:21
Unknown
But you can you can chase something. You can throw lots of money on it. You can promise the world to it. You end up breaking your your promises. Your customers are going to walk. So sure, it's it's I mean, you know, so that's what I like about your, you know, the, the the pragmatic approach you are taking, keeping a human in, in the loop viewing.
01:16:33:24 - 01:17:01:26
Unknown
You know, it's like if you if you combined a very experienced, seasoned architect and engineer such as your well, I guess either of us with, with with with the latest tools, that's how we can create value. It's a combination. It's more of a, cyborg than a robot, I guess. And at the end of the day, today, you could create value fast.
01:17:01:28 - 01:17:20:25
Unknown
This is. You could never do that before. I mean, I just met another customer yesterday, and I showed him, like, in five minutes, and I showed him he can do it. You know, there's, like, this great, automation and AI tools that are even free, like an A10 for it. About it. Yeah. Yeah. Of course. So I took Ahmed.
01:17:20:29 - 01:17:42:01
Unknown
I connected Claude to, this MCP server of any ten, and now I would. I didn't even know how to work with an I. Then I'll show them. Give me one of your documents and let's and said he built me like the sales portal for, based on this document in my Europe system. And in five minutes, he was like.
01:17:42:03 - 01:18:10:09
Unknown
I mean, I was amazed. I didn't know the results going to come out. He built this flow in, and it did for me. And then I ran it and I saw the end result and I saw the customer going like this. Wow. The things I could do is this, I mean, the, the turning point of thing that before I spent a month and a lot of a lot of people and money and companies just trying to achieve this, and now you could just suffer like in 5 or 10 minutes.
01:18:10:09 - 01:18:45:25
Unknown
And again, he did everything in conversational. It wasn't even you didn't even write a single line of code or drag it dropped anything. I mean, this is what we're facing now, the the time to market a value. I think today is like a hundred times faster, of course, and cheaper and less people that need to be involved. So I think today, if you have the expertise, if you if you know what you what to ask and you know how to achieve it, things you can think about, you can bring value tomorrow morning to your customers.
01:18:45:27 - 01:19:08:02
Unknown
And this is I think this is the biggest change in, in the industry and the way we think and, the results that we can bring to the table. And then if and if, if he could bring out enough. I know five years ago we could either content products to the market that can have an impact today could bring 100 products or another 100 solutions.
01:19:08:02 - 01:19:29:21
Unknown
And I think this is the game changer, how fast and how efficient and how good think. But again, you have to be the expert. There is no cutting it. I mean there's no shortcuts to bring good products. You have to be an expert. You have to understand, you know, basically everything, because if you don't, it's going to fail.
01:19:29:21 - 01:19:47:08
Unknown
It's going to fail. When you scale up, it's going to fail, and you won't even understand how to fix it, because you have to be. And this is what I say if you are an expert. I think people like us that had all the years of learning for this learning curve and dealt with problems, hit the brick walls and solve the problems.
01:19:47:11 - 01:20:14:15
Unknown
So this is the place where now we can, I mean, this is the place for us today to bring the real value. If you take people that, just finished school, have a few years of experience, they still don't have the customer expert, they don't have the business value or the business perspective. And combining all these together with today with things we can do, I this is the real value I think we can bring to the table.
01:20:14:18 - 01:20:41:15
Unknown
Wow. Awesome perspective. Thank you so much. I look forward to our next conversation. Will you come back? For sure. I hopefully we have enough, material to, you know, to make, to cut and paste what we need and then make this. I think we will, I mean, that they I build this, they make this an awesome podcast.
01:20:41:18 - 01:21:08:16
Unknown
That's all we have to say. Exactly. Make make this awesome. I mean, I've just used the new tools in, of, notebook, you know, the, infographics and the, slide decks. I mean, you just bring something so rock and you get this awesome infographics out of it. And, I mean, I mean, I mean, if we give this podcast a day, you know, the collab stuff like that, I think it could, I mean, do magic with it.
01:21:08:19 - 01:21:33:19
Unknown
Well, look, if you can make me look better. Hey, I'm I'm there. I mean, this is easy. This. I, I've been told that I have a face for radio, so we'll we'll see how this all goes, but I, I now we need Alex to Alex do his magic. Make us look at as my wife say, when you take a picture, make me look 20 young, 30 years younger.
01:21:33:21 - 01:21:42:18
Unknown
There you go. So let's see. Alex, are you still listening?
01:21:42:20 - 01:21:44:19
Unknown
Okay. I think we're ready, my friend.