Ep 02: Preparing New Engineers for Success

Matt Brickey: All right. Good morning. This is Matt Bricky and welcome to the Cloud Currents podcast. I'm here with Abdel Sghiouar and Abdel, welcome.

Abdel Sghiouar: Thank you. Thanks for having me. Thanks.

(0:21) Introduction to Abdel Sghiouar

Matt Brickey: There's a couple of categories that I want to get through today. So I would jump right into the questions if you don't mind. Mostly I wanted to talk about you to get started, right?

Your journey has been very diverse and I want to get into some of the motivations of why you're at Google now and some of the things that you've done. But I noticed that you know, you got started in the technology heart kind of hardware business, ERP business. What was your path into this industry?

Abdel Sghiouar: That's a very good question. So I am trained as a software engineer, right? I have a master's in software engineering back home in Morocco. That's what I, that's what I did my studies. And I guess two main things when I was living in Morocco at the time, like 99 percent of the tech industry was basically consulting, right?

I was like, just companies doing consulting for the French market, obviously, because Moroccans, we speak French, so we can work with French companies. And, even in the consulting space, because I had a lot of friends doing that. They were not doing actually like product development. They were just doing a lot of maintenance.

So maintaining apps, maintaining like Linux systems, stuff like that. System administration and stuff like that. And I guess that that just didn't really interest me to start with. I wanted to work on products and there was virtually no company doing that back in the days. So I ended up working in a data center just by coincidence.

I have a friend who is slightly older than me. Who worked in this data center and he pulled me in to join. So I joined as a data center engineer. That was kind of my official title, but maybe more like a modern-day DevOps. So basically a jack of all trade doing everything needed because the company was small.

We were four people when I started there. So there were the boss, the assistants, my friend, and me, and we had to do everything. And what's particular about this company is that it was the first career-neutral data center in the country. So before this company opened, if you wanted to host your server somewhere, you had to go to one of the main three ISPs we have in Morocco, right?

What we had at the time. And then you had to use their data centers. And obviously if you go to ISPA and you use their data center to host your stuff, they're not going to allow you to bring internet from ISPB. Right? Because they want to sell you their internet. So, career neutral means we were not tied to any of the ISPs.

We had all of them and we could connect you to whatever you want. So we, we, we were very appealing in the sense that pretty much companies that just didn't want to be stuck with one ISP could come to us. And we had some like extra, like free smart hands. If you're in the data center space, you know what smart hand means that that was free.

We didn't charge for the service. Other companies would charge you for like per hour, right? If you open a ticket and say, restart my server, that's like a hundred dollars, right? So, so, so yeah, so that was the background. I started working there. So I started with hardware and that's how technically I got into Google.

Matt Brickey: Did you, do you think that the jack-of-all-trades experiences in your early life helped you where you are now?

Abdel Sghiouar: Yes, yes, pretty much. Yes. That's actually one of the things I always tell people when they ask me about advice is like, you have to kind of have that T shape we talk about quite a lot, which is like, you know, a little bit about everything and then, you know, one thing very well.

My one thing very well was Linux and I obviously, obviously knew everything about distributed systems, databases, security, all that stuff, right? So, that fundamental knowledge continues to serve me today to basically pick up whatever I need to work on and work on

Matt Brickey: it. Yeah. Do you find that with the younger developer, sorry, the younger developers that you mentor is that are they kind of single threaded or do, are they coming to industry with that jack-of-all-trades mentality?

Abdel Sghiouar: I, I guess that this is not really a new problem because even, I remember when I was a student in my class, we had people that were like, I want to be a Java developer. Mm-Hmm. , you know, and they're still doing their masters, and I'm like, like, you don't know what you're talking about, because you pretty much like haven't experienced anything else.

Right? Right. Because obviously we were studying c and t net and Python and we were doing a lot of things, but people were like, so it's not like an, it's not like a new mentality with the new generation, as we say. But I think, yes, in the sense that more and more it's becoming the norm that people come into the market with a pre-existing idea of what they want to do.

And this is particularly true with front-end stuff. So people come with, like, I want to do React, or I want to do, you know, Angular, or I want to do Flutter, or I want to do, like, all these JavaScript frameworks, people have, like, a predisposed idea of what they want to do. So, so it's not a new problem, but it's more permanent, I would say.

So how did you end up at Google? So I went, so basically they, they reached out to me. I had a LinkedIn profile, which was pretty up-to-date. It was like data center engineer. And I guess, you know, recruiters in Google, like recruiters in a company that have this like recruiter version of LinkedIn that they can use to like look for keywords and regions and stuff.

And I guess that's how they found me. They just found my profile on LinkedIn. I got an email from a recruiter in Dublin and he was like, Hey, we're reaching out from Google. I didn't believe it in the beginning. It took me like three days to reply. Then I replied, and yeah, I got in. I did the interviews, was like, all in all, I think it was seven interviews in total.

Okay. So two, a couple of them online, and then they flew me to Belgium, because that's where I started working, and then I did my, my on-site interviews, and then I got a job.

Matt Brickey: Great, well, what's made you stay? Cause you've been there, I think that's the longest 10-year job, right? That you've had.

Abdel Sghiouar: Yeah. 10 years. Yeah. Great. I just closed 10 years on the 19th of August. Oh, congratulations. Thank you. I mean, it's because I haven't been doing the same thing for 10 years. I tried a lot of things. On average, every two years I am doing something new every two to three years. On average, I am working in a new team that may be the same job title, but just doing something completely different.

I think. It's the fact that, I mean, it's a company, it's a very old machine, very well-oiled machine, it just works because, like, this is compared to my other experience, which was a startup. Startups are like a hit-and-miss, right? It depends where you work. Sometimes it's not like you know, Oh, we're not going to give you your paycheck this month because blah, blah, whatever.

And then you get involved with a lot of things that you don't necessarily want to be involved with. When you work in a big company because a lot of things are already figured out for you, you're kind of shielded from all this, like, noise around you. So you can get to focus on the thing that you need to focus on, right?

So, so it's that, and also the mobility slash moving around inside the company is relatively easy. I moved like three times, changed teams like three or four times. So, and yeah, all in all, it's just, it's a cool company. Everybody around you is smart and very cool and know what they're doing. So it's pretty cool.

Matt Brickey: Is that something that was intended to allow you to move around like that? Or is that just something that the culture has organically gotten to by bringing some of the best in the industry? So the team that naturally, I think that, that your personality will lead you to want to do that.

Otherwise, you'd be out the door after, you know, three or four years.

Abdel Sghiouar: Yeah. I mean, I guess it's because it's a big company, it attracts people with pretty much all kinds of type of personalities. Right. But I guess if you are like me and you always get, get like, like how to say gets bored quickly. With things and you want to change and do some new stuff, it's a good company to be in, right?

Because there is always so much to do all the time. But yes, it's, it's internally, culturally, it's like internal mobility is pretty much encouraged. Technically, managers cannot even block you. So if you want to, to, to like move to a different team, you can just apply, do the interviews. And then if you are, if you accept the offer from the new team, you can just inform your manager and move.

The only thing they could do is technically or potentially delay your move. If they need you to finish what you are doing, but they cannot just say, no, I'm not, I'm not going to allow you to move. Right. Which I know is a little bit frustrating for managers because they don't really like that, but it's just the company, how it works.

Right. But no, I, I have met people before that have been doing the same thing and they still doing it for like a very long time. I mean, I guess it depends on what you are in for. Do you want to just like do the job and consider it as like a nine to five and just do other stuff on the side, or do you want to like.

Make, make, make the most out of your career. The way I would, the way I see it is that, and this is pretty much something that happens all the time. Like, inside of Google or inside pretty much all the big tech companies, it's a pretty much-accelerated career thing, right? You're learning so much in a very short period of time.

And it happens very often that people actually get a job in a big tech company as an individual contributor. They spend like five, six, seven, 10 years, whatever. And then they move out. And when you move out, they go to become a CTO, right? Right. Because in these like six years you've been there, you have seen so much and you have gained so much experience that's outside of the big, big company.

Your, your, your, your, your experience is valued more, I would say. So, so yeah, no, it's, it's all in all a very cool company to be in.

Matt Brickey: Yeah. I hear a lot of parallels to, you know, I have a lot of larger consulting organizations in my background too. And I hear. A lot of the same parallels is that, you know, if you got into an area that you didn't like, you can move into a different area and that, you know, the changes of the clients, the changes of the people that you work with is what made everybody stay.

That's what made it fun. So that, that's a pretty cool culture that they've built there.

Abdel Sghiouar: And I guess, I guess, because I talked earlier about consulting, I think consulting really depends where do you work, right? And what kind of projects you work on. I guess the thing that really did not encourage me when I was in Morocco was working on maintenance because no developer wants to do maintenance, right?

No one wants to maintain somebody else's code. So, so, so because I had friends who were just doing that and it was crazy. I remember this team that two of my friends were in. They were four of them and they were maintaining all the applications of the hospitals of the city of Grenoble in France. So there's like 12 hospitals and they have like 250 apps written in like.

Something like 20 programming languages, everything from Java to JavaScript to the net. And they had to maintain all of them. Right. So, so I was like, that doesn't sound very fun. Yeah.

Matt Brickey: That sounds like “run away”.

Abdel Sghiouar: Exactly. And the problem also with this is that as you like, see, since you're working consulting, you know, this contract expires, right?

So the company, the contracts between the client and the company will expire after a certain amount of time. So there is even no guarantee that you will just be working on the same thing three years from now.

Matt Brickey: how did you get into hosting your own podcasts?

Abdel Sghiouar: Oh, so I have two podcasts. One is active and the other one is a little bit dormant. So the dormant one was a COVID project. Like everybody started off projects in COVID. Sure. Called the cloud careers podcast. I did a few episodes and then I put it aside.

The one I'm doing right now, which is very active is the Kubernetes podcast. That's a podcast that Google pays us to do. So that's like a company thing. We had the host before and that host ran the podcast for four years and then the host had to move out of the company and they were looking for new hosts.

So they Proposed to me and one of my colleagues, Kathleen, Kathleen Fields who lives in Seattle and we said, yes, and we've been doing it for a year now.

Matt Brickey: How does that really fit with you know, your team and your work? Like, are you using it almost in a professor or professorial way? Or, you know, give me a, give me a little color on how do you use it for your job?

Abdel Sghiouar: Yeah. So when the podcast was started, it was called, well, it's called the Kubernetes podcast. And the whole idea of that podcast was it's a weekly or bi-weekly show where the host would just interview people who work on the Kubernetes project or the cloud native ecosystem in general, right? So they don't only do Kubernetes, they do a bunch of other things, right?

And so the show had like a lot of content around open-source cloud native projects, right? And sometimes they would talk about, like, cultural things and community stuff and whatever. When we came in, we wanted to bring our own personal, like, like, flair on it. So we started doing a couple of few things.

One of them is event coverage. So we actually go to cube cons and, you know, cloud-native conferences. We did, we did, like, two cube cons. We did one cloud-native security con. We did WASM con, which was very recently, like, in September, I think, or, yeah, in September. And the episode is coming out this week.

And we just go around and interview people, right? And when you listen to these episodes, you have that like that, that feeling of being on the ground because you have the background noise of the conference, you know, the conference hall is really very, very noisy. So that gets picked up in the microphone and then the people just answer a set of questions.

And then we stitch these questions together and then it's mostly about questions about what are people in that conference for, like, what are you here for? What do you expect to get out of this? Blah, blah. And it's, it's very interesting for me because. Most of those have been done by my co-host. I did one, which was KubeCon Amsterdam this year.

But you have done like three, I think, or four. And what's interesting for it, in it, is that everybody comes to conferences with different expectations. And then everybody comes in and learns something new or gets surprised or whatever. So it's, it's very, very interesting to have that kind of content where people talk about what have they experienced in that conference to hopefully encourage somebody who would listen to the episode to go there the next time, right?

Right. And then we do, of course, we're bringing in our own network. So we have people that we know that have been doing cool stuff. I did a very cool episode that received a lot of good feedback and we're planning to do it. So I, I have in my talks, in my day-to-day job I do like talks, right?

Like as a developer advocate, I do content and stuff. And whenever I go on stage to talk about a technology I am like the old grumpy person who doesn't like new stuff. So I like to do this thing called the history of. Essentially I'm, let's say I'm going on stage to talk about containers and when they matter.

Well, before I start into that, I will do 10 minutes explanation of the history of this stuff. Like how, where did we came from to today having this technology available to us so we can use it. And so we interviewed Phil Estates. So Phil works at AWS right now as a principal engineer, but he spent like 20 years at IBM.

And IBM, he was working on Linux. And then from Linux, he went into the container space and then started what is known today as ContainerD, which is a container runtime, right? It's a very popular one. And he's one of the maintainers. So he moved to AWS to continue working on ContainerD. So basically AWS pays him to work on open source, right?

And so we had this story, which we had this interview for like 45 minutes, which... When I listened to it later and a lot of people that listened to it felt like it's two people sitting by a fire Like a side fire chat, right? Just two people talking like like one person have been there for longer than the other person and telling them how things were before this whole container space Started and that was so interesting.

So I agreed with my co-host that we will do more of those. So we're looking to do something around WebAssembly that people like, we will bring somebody to talk about WebAssembly in browsers because now it's becoming a hot topic, everybody talks about WASM and WebAssembly modules. But I want to have somebody to try to explain where this thing came from.

We'll do maybe something around the GVM, we'll do more of those, right? So that's, that's, that's very interesting for me. And that's to answer your question, in a, in a professorial way, that's, that's probably the kind of content we do where we're teaching people stuff, right? But we do also other content, like we talk to open source maintainers, we talk to people from the community.

So it's a little bit of everything. It's a very various show with a lot of content that, that, that hopefully everybody can like take one thing out of it. You don't have to listen to all the episodes, but the episodes you would listen to. Would be interesting and because the show have been around for a very long time and have very big followers We don't really have issues finding people usually when we reach out to people they want to talk to us, right?

(15:03) How Cloud Experts Can Build their Own Brand

Matt Brickey: Yeah, I think it's a great way to reach people now, you know, given the amount of remote work. We always had remote work prior but you know after the pandemic it's still very viable It's a great way to reach a wider audience and speaking of that Reaching a wider audience, you know, you've got, you've been a public speaker and a content creator for quite a while.

What advice would you give others that are trying to kind of break into building their own personal brand?

Abdel Sghiouar: Yeah, that's a, that's a very good question. I mean, we have to admit, we have to acknowledge one thing. Public speaking is not for everyone, right? Right. Like, there are people who are good at it.

There are people who are not good at it. So like, so, so I want to break away this, this like because like one, one, one of the things that I'm noticing right now in the tech industry is that people who are doing content creation like me, That kind of building a cult of personality around himself, which is bad.

We want to be able to tell people what are you in for and what you should expect from this, right? So the first thing is you don't have to do it. It's not for everyone. Like it's just a thing that you do it if you want to. If you want to, probably the easiest way to start is start with small events. So like meetups would be a great, great starting point.

There are a lot of meetups everywhere. There are a lot, if you're in Java, there is this thing called Java user groups. If you are in Go, there are Go meetups all over the place. Start there. Go to a meetup, attend, see how it looks like. Maybe talk to the organizers, propose to them something. Maybe they will be interested in having you.

Start with a small audience. Go, go, do something small and then graduate from there. The second thing is, I mean, I think building a brand for yourself has to come with value. Like your value on the internet cannot be being the person who just complains about things all the time. Which there are a lot of people doing this on the internet.

Yes, exactly. So, so so look at, like, I mean, what you could do, there are multiple things you could do. You could, you could maybe build content that would teach people stuff, right? I know a bunch of people who are doing that and they're very good at it. You can, contents could be articles, could be videos, could be whatever, right?

Again, don't, don't aim to start a YouTube channel and get it to a hundred thousand dollars within a year, a hundred thousand followers within a year, maybe start small, make something small, doesn't have to be two hours video, it can be like five minutes or ten minutes streaming is becoming very popular on a lot of, I know a lot of programmers who stream while they are coding and there is value in doing that because then people see your coding process and maybe they will learn one thing or two from you.

But just don't, don't like, There is quite, as you said, there's a lot of those people who just complain all the time on the internet. And then there is also this like new trend of like Instagram tech influencers taking pictures of their desk setup and Showing their mechanical keyboards and stuff like that.

That's that that's there's no value in doing this stuff, right?

Matt Brickey: So that's actually last for a few minutes and then it's gone, right? Exactly. It'll replace it.

Abdel Sghiouar: Exactly. It's funny because funny is We're talking about this because literally yesterday on Twitter I was home, I'm working from home and it was cold.

So I just put in like a Moroccan dress, which is like a very popular Moroccan dress, we call it Jalaba, it's like a full. with like a kind of like a hat. It looks like a Jedi dress basically. So I put it on and they took picture with my desk and they just tweeted it. And I was like, so am I becoming a tech influencer?

Is this working? And literally that's picture got like, I think this is the highest number of likes because it was sarcastic, right? It's not supposed to be serious. So yeah, I mean, your brand on the internet have to have value. That's the TLDR to this. And, and, and people have to come in to look at what you're doing and learn something from you hopefully.

And not just, Like, you know, you're not just taking pictures randomly and posting them.

(18:32) Skills Gaps in the Cloud Engineering Space

Matt Brickey: Essentially. Good. That's a very good answer. Very good answer. Cause I think a lot of folks do focus on the me aspects of it. And really, you're just really trying to deliver content so people can learn something from you.

You're, you're a vehicle for them to learn. They're not there for you necessarily. Like you said, it's not Instagram. That's a, that's a different story. So with that said, right, that was a really, really good answer. If you could go back you know, when you first started your career, because you've went from You know, hands-on doer.

And now, now you're doing a lot of advocacy for other developers. What core skills would you tell yourself to focus on earlier on?

Abdel Sghiouar: Very good question. I think right now one of the main things I would have loved to learn early in my career would be writing like learning how to write. I, I want to write more.

I am working on a few projects where I want to be, where I'll be able to like blog more, have maybe a newsletter for like the container space and stuff like that. And, I know a few people that are very good writers, like I met people across my career that are very good writers, and each time I talk to them they either came from a writing background, you know, there's a lot of people that come into the tech industry where, like, tech is not their background to start with.

They were doing something else, like physics and math and things like that, right? And, and in, in, in doing this, that kind of technical fields you have to write research papers and you have to write dissertations and stuff and you learn how to write. Or I ever met, like, one of our directors in my team right now came from journalism.

So his background is journalism, right? And then learned tech along the way. So, so that's probably one of the main skills I would have loved to develop very early in my career. Because it doesn't, it doesn't come like, like public speaking. It doesn't come naturally to people, right? You have to learn it.

Matt Brickey: Yeah, and it's a craft you have to keep practicing.

Abdel Sghiouar: Exactly, exactly. So, I met, so whenever I meet people who are really good writers, I try to like pick up their brain, like can you recommend a couple of books that I can use, and I'm going through the process of reading some of them now. And it's, it's, it's a lot of, like, it doesn't look like, but it's a lot of technical stuff that you have to know how to do.

It's like formulating words and stuff, right? Basically writing is difficult than talking. Talking is easy. Yeah, it's cheap too. Yeah, exactly. Talking is easy and cheap. And actually when you are doing public speaking, that's the funny part. When you do public speaking, it's, if you, if it comes naturally to you, it's even easier because when you are speaking, you're just translating ideas from your head directly through your mouth.

But. If I would be asked to take one of my talks, 45 minutes, and turn it into an article, it would take me ages. Because, because then, then, yeah, because then, then, then, because like when you are speaking in front of an audience, you can do a lot of things that you cannot do in writing. For example, you can repeat stuff.

You can say stuff over and over again, right? If it's spread over 45 minutes, no one, no one cares. But if you put it in an article, it becomes annoying to read. So, so,

Matt Brickey: so yeah, I think now, especially, you know, given our attention spans have shrunk over periods of time, we, we went from a reading society to a watching society.

I think you're spot on there making a succinct point in written word is much harder than making a succinct five-minute video.

Abdel Sghiouar: Yes. Yes. You can, you can say. We can say a lot of BS in five minutes video and not drive home a single point, like you can watch a five-minute video and go out of it without learning anything, but reading five minutes, an article, you will learn one thing or two, if it's well-written article, of course.

Right. Right. And so that, that's, that, that's, that's something I am like kind of focused on because I, I have a project in head at some point, essentially to write a book or two, and that's something I would have to learn, right? Because even in technical writing, you need to know how to write properly.

Matt Brickey: That's a good goal to have. I'd like to switch gears now. And I want to talk about, you know, your, kind of your, your core competency, which is training, you're mentoring and training new engineers, right? At least it's, at least it's your focus for right now. So give me an idea when you're talking with companies that are adopting new technologies, cloud containers, right?

AI, whatever, what skills gaps are you seeing that are the most prevalent in these organizations?

Abdel Sghiouar: Yeah, that's a, that's a very, very good question. There is a book I would recommend people to read after, after I answer the after I answer the question, I think talking about like this attention span, I think that's one of the problems I'm seeing is that people coming to the tech industry are more and more specialized and, and a lot of people are lacking some really, really basic skills.

I am literally talk, I'm literally thinking right now about for the remaining of this year to work on a talk that I will propose to some conferences next year, which would be called how the internet works. And that will be for four or five minutes explaining how DNS works and how web servers work and how HTTP works.

Because I guess a lot of times, like this, this comes through interacting with developers where somebody comes to you and they go like, Oh, I have this error. And then you start debugging and you start digging down through the error and they realize which is DNS issue, right? And, and in my head, because I have those like basic skills and I know how to kind of pinpoint where the issue is, and I have the possibility to debug systems.

In my head it's always like, but like, how come you've done five years of master's in computer science and you don't know how DNS works or you don't know how a web server or a firewall works or stuff like that, right? So, so the gap that I am seeing is, is so stupid that it's actually a gap in fundamental knowledge, right?

And I think it has to do with multiple things. So people are increasingly specialized or especially engineers are increasingly specialized. And I think probably even the way companies hire, they just hire people that knows the thing they need them to know. Yeah. They don't have like a, let's say a forward-looking way of, of looking at career development.

Like, let's hire somebody who is really good at technology and then being good at technology, they will pick up the thing we need them to pick up now. And if three years from now, we need them to pick up something new, they will be able to do that. Right. People will just go like, Oh, I went, I wanted a spring developer, like a Java spring developer, not a Java developer, not a GVM, not a programmer.

I want a specific person who has a specific knowledge of one specific framework. Right. So, so, so that's, that's, that's, that's kind of one of the main gaps I'm seeing today that people want to hire somebody who knows things now and wants to hire somebody who would be productive on day one. Because obviously hiring a Jake of all trade.

If you can't find them the difference would be that they need like a training period. They need time to like, like, you know, upscale and learn the specific things that company does and, and stuff like that. And yeah, so that, that's, that's basically the main thing. So, so whenever I'm talking to people, developers, companies, stuff like that, I always say the same thing.

You have to focus on the fundamentals. You have to have no, a little bit of all this basic stuff that makes up the internet as we know it. And then from there if you are a Java Spring developer, And tomorrow your boss goes like, Oh, we have a problem. This server go debug it. Well, at least you know how to SSH and then you know how to look for logs and, and you know how to look at processes and memory and stuff like that.

Right. Which today I think like a lot of developers don't know how to do that. Do you think that that's

Matt Brickey: a, is that an effect of the internet technologies being more of a utility now? You know, for companies like the one you work for, right. Modernize the internet quite a bit. A bunch of other big names did as well.

But do you think that that's what the universities are just skipping over, is that that is seen as, you know, that's already decided, it's, it works, right, there's no need to understand that, or is that just more fundamental on how the universities are teaching students to focus more on a framework than a fundamental?

Abdel Sghiouar: I guess it's a combination of both, I think. So I've seen, I've seen more and more universities like creating more and more specialized curriculums. And that's like a direct answer to, to what the market requires today. So you go on in a random university, tech university you don't have, well, you still have the standard, like software engineering master's degree, right?

Where you are teaching everything, right? Or even more generalized, a computer science master's degree. But then you will see like a data and ML master's degree. So like something very specialized, they will just teach you the data stuff and the machine learning stuff. So you learn the frameworks, TensorFlow and Airflow and all this stuff, right?

Or, I mean, I, I think you, you, you have a long career, so you probably understand this, like, a very long time ago when I was in university and I was studying, my networking classes were Cisco classes, right? Right. So, so we were not studying networking, we were studying Cisco. Cisco, yes. Right? Because Cisco was so clever, they came to my university and they were like, here are a bunch of switches, here is like curriculum, teach people networking, teach people this, right?

I think there is a lot of companies doing that they are influencing directly what people are studying so they come out to the markets with with already the knowledge that this company need them to have right so and then also the other thing that I think is like what became very popular over the last couple of years is all these boot camps where it's not even like a six five years master's in university it's like a six month program Come in, teach something, learn something for six months, and then you are a full stack developer, whatever full stack means, right?

So, of course, we're not gonna expect these people to know how the internet works. But like, my, my I was discussing this once with a recruiter I think At Facebook or something? I don't remember. It was like at an airport, just hanging out at the lounge, and then chatting with this person. And we were talking about this specific problem, and I was like, he was asking me, he was like, so how, how do you propose we can solve this?

How do you see a path forward? I was like, very simple. When you're doing interviews, ask people a very simple question. Can you explain to me what happens, in details, [00:28:00] between you typing and you typing google. com on your browser? And through the entire seven layers of the OZY model. Just explain this to me, like, step by step.

All the way until the request gets to the server and returns to you and it's rendered on your webpage. Because this is essentially going through everything. Networking, security DNS, internet. Going through all of this basic stuff. And if you're not able to do that, well, I think you have a gap in knowledge, basically.

Matt Brickey: Interesting. Yeah. And I'd be curious how many times you get the answer. What is the, what is the OC model? What is the OC model? That's an immediate, well, we're done here. I guess.

Abdel Sghiouar: Exactly.

(28:35) How Engineers Can Get to the Next Step in Their Career

Matt Brickey: Speaking of that, that's a really good segue into the next question, right? Is that, you know, young engineers, everybody's hungry.

Everybody wants to be in charge as fast as they possibly can get titles, whatever that, whatever motivates them. How do you recommend engineers get effective feedback to aid in their career growth? Right. Like what, how do you teach your engineers to like get to the next step?

Abdel Sghiouar: That's, that's a very good question.

I guess like, I mean, growing your career really depends on a lot of things. And one of them depends on probably the size of your organization and what are you doing in this organization? Right. I guess like my, my, my answer would be the more you grow your career, the more senior you become, the more your value toward the company you work for, the organization you work for is not about how well you know the technology.

Is about how well, you know, the technology works or it doesn't work for particular use cases, right? Your value as a senior engineer is not in you knowing how Java works. Is in you knowing this particular framework with this particular version on this particular kind of hardware. Is it going to perform yes or no?

Or this particular architecture with this particular set of open source software. Is it going to work? Is it going to scale? Etcetera, etcetera, right? It's that, that, that experience that, that, that. Is outside the core thing you were doing as a developer, which is writing code, right? Or maybe the other thing is your ability to look at somebody's old code as part of peer reviewing, which I hope by now everybody's doing.

But like you, you have somebody writing code, send it to you for peer review and you look at it and just by scanning it, you can quickly say like, this is not going to work. This is a problem, right? Or this is going to have a memory leak or whatever, right? So that's, that's what company effectively pays you for.

They're not paying you to write code. Writing code is, becomes secondary. What becomes primary is your experience in that particular project, in those particular set of tools that you are using, and your ability to actually translate that toward junior people coming to your team. And I guess, generally speaking, if you get to your career, you get to this point where you see you, yourself doing this, which is helping upskilling other people, you are effectively in the next step in your, in your career, right?

You're effectively senior. You have a senior level, whatever that means, right? Then where do you want to go from there? Really depends on people. I know that a lot of people like the, a lot of people like to be, or they are people person, so they go to management, which is completely a valid path to follow if that's the kind of things you want to do.

Management, engineering management is particularly difficult because there is, like, there is a difficult, there is a big difference between people management and engineering management, right? Yes. Managing people is helping them unblock themselves so they can move forward with the kind of work they need to do.

Being an engineering manager, you are also responsible for the technology, and you are responsible for your team's choice to adopt x versus y and, and, and how, how that will impact them and how they can be productive, et cetera, et cetera. So if that's something that you see yourself. Kind of more leaning toward just go for it Otherwise remaining as an individual contributor until pretty much the end of your career is also a very valid option And that's one option I'm going for my career for sure then I can find.

Matt Brickey: That's a good that's a great answer I find that some people think that that is not the right career path because they're not in management, right?

I encourage a lot of people to think about that as well as that. We don't need that many managers, right? We need a lot of people that are, they're brilliant to do other things than manage people or manage teams or anything like that. So I think that I agree with you there, strongly agree with you there to recommend to others that individual contributions are what makes this world work.

Abdel Sghiouar: Yes. The actual people to do the work.

(32:27) How Companies Can Attract and Retain Skilled Employees

Matt Brickey: How have you seen companies that you work with, maybe even your own company? You know, we're in, we're in a pretty tight skill market, right? A pretty tight job market for certain skills anyway. How have you seen them attract and retain the people?

Abdel Sghiouar: That's a very good question.

I think, I think that the best, the best answer to this would be make people work on interesting things to them. Because that's the thing with, with, with, with people in tech. I don't, I don't, I don't think that the problem I have is, is a me problem. I think it's a general problem in tech. Is it gets boring after a while, right?

It does. It gets boring just working on the same thing over and over again. And By the way, if you are the kind of person who just still manages Cisco switches 20 years, like, later, you're still doing it today, hats, hats off to you, right? Like, amazing. Good, good on you, right? So there was an article that I have read a few, a few, a few months ago, I think, or a year ago.

I think it was something called Shadow Developers. And this article was all about, well, all these, like, conferences and public speaking and podcasts and stuff like that we're doing. We're probably doing it to the tip of the iceberg of the developers, the people who are exposed to the internet, who are reading stuff.

But there is still, there is probably still, like, a database administrator working for that town hall in the middle of nowhere in America or in Sweden or whatever. And these people are not out there looking on YouTube, learning stuff on YouTube. They're just doing their job on their side, right? That's what the article called shadow developers, right?

And these are, these people will do these jobs for like 20 years, 30 years, all the way until they retire. They will just do the same thing, keep doing it, all the time. They are very good at it. And, and, and the people who are managing them, or the people who have these people in their team, they don't want them to leave, they don't want them to, like, they will do whatever they can to make them happy.

And just, like, please stay here, because if you leave, we're literally gonna, like, You know, we're going to die, we're going to bankrupt, the company goes to hell, right? But to answer your question, like, to, to maintain, like, you just need to make sure people work on things they find interesting, right? And also, and this is probably for developers specifically, I mean, the worst thing that you can ask developers to do is to put things in their path that will slow down, slow down how, how, how fast they progress, right?

As a developer, I want to write a feature, get the feature out, see it out in the wild within ideally 72 hours of me writing the feature and testing it, right? If your release cycle is three weeks, that's way too long. If your release cycle is two weeks, that's too long. If it takes 24 hours to run all your unit tests, you're doing something wrong, right?

No one wants to sit down and just wait for, no one wants to sit down and wait for 10 minutes to run a test. Everybody wants to be able to test things quickly, get them out and move to the next thing. So, so, so like there is this, like this engineering productivity is a topic that is coming back to the, to the, to the, to the discussion space very recently.

Like people are starting to talk more about it. And how do you make people productive and how do you make them like do things iterate very quickly and move to the next thing? And it's probably even, it's, it's probably, making sure your developers are productive probably will have a net positive on your business, right?

Because you, no one, I mean, developers are very expensive to hire. They're very expensive to, to, to keep to maintain. They're very, very expensive to, to source. So, so the more you can make sure people are happy and they can do their job quickly and effectively, the better it is for everyone.

(35:38) AI Adoption: Opportunities and Challenges for Developers

Matt Brickey: You mentioned something there about roadblocking developers, and I agree.

So that's a good way to pivot into kind of our emerging technology section of this. Do you think that AI, and I'm just going to use AI very generally, there's no specific use case, but do you think AI can help us there get rid of some of those roadblocks, menial tasks, speed us up, like you said, on tasks peer review, anything like that?

Where's the opportunity there for the developer community?

Abdel Sghiouar: That, that's a very good question. I think that the answer is yes and no. Yes, in the sense that if you are able to take an AI tool and add it to your workflow in such a way that it positively impacts how productive you are by maybe speeding up how fast you can find answers or helping you quickly spot problems in your code or in your system or in your architecture, that's a good thing.

However, the problem I have with this is it depends on how the tool works and it depends how it's trained. Because if the tool is trained on public information, like Stack Overflow and public documentation, I think. What will end up happening is that this AI tool will just recommend the same thing to everyone and if the thing recommended by this AI tool is bad, then everybody will end up just writing bad code, right?

So, like Stack Overflow, like Wikipedia, it's not the source of all the truths. Not everything on Stack Overflow is correct all the time, right? Right. And it's it's funny that, that we were I was trying to book a flight yesterday on the Scandinavian Airlines website and, I was trying to log in and I was getting this like internal edge error, which was a funny error message, like internal edge error.

So I just took a screenshot and I was like, what is an internal edge error? And then somebody replied to me saying that a few weeks back, this person was trying to book on the same website. And then it didn't work. So he opened the debugger in the browser and he found some JavaScript code and there was a big comment that says, do not use this in production, right?

So that's, that's essentially, I call that YOLO, YOLO, YOLO developer, which is like, hold my beer and wait. Watch this. Exactly. So, so probably people, people just blindly copy past code from Stack Overflow into their internal production systems. And to go back to the AI thing. If this AI is trained on public information, nothing can guarantee that the kind of answers the AI gives you are good, right?

Because the public stuff is public, written by everybody and anybody, and there is no assumption of quality there, right? If your AI tool is trained on your own code, on your own practices, on your own best practices, your company best practices, then that probably could have a positive impact on how developers work.

Matt Brickey: Do you see where organizations will focus more on proprietary models than public models? Because, to your point, The model will learn from a model that learned from a model that learned from a model. And eventually you've whitewashed out a lot of creativity, a lot of the maybe 5 percent variances, those kinds of things.

Give me your thoughts around that.

Abdel Sghiouar: I think that the private models is the way to go anyway, not just for these reasons we're talking about, which is, you know, not, not, not having like public models trained on bad data, but also for confidentiality reasons, right? Like no one wants to expose their data to the internet.

Data privacy is becoming a problem. There have been tons of leaks. There probably have been 20 leaks since we started talking now. So, so, so, so, I think private models is pretty much the way to go. And that's, I think companies that are going to make AI tools that makes it easy for companies to train all their own data in a secured way, safe, non-leaked way, will probably have like a, like a, a very, they will be able to do good business going forward.

Matt Brickey: The chat GPTs of the world, I am not quite sure what's the future for that product, to be honest with you. Yeah, it could just be just a quicker inquiry, right? Get a little more robust answer than a bunch of links or something like, I don't know. Yeah, it'd be exciting to see how that comes to fruition. On that, so I, you know, kind of an alternative viewpoint is that we, you know, we still see a lot of organizations that haven't even adopted cloud.

So how are they going to adopt AI, right? So that's one kind of thing. But what have you seen you know, the challenges that still exist to cloud adoption and how can we help companies overcome that?

Abdel Sghiouar: That's a very good question. I just wanted to give you an example about like the previous topic quickly, because this example is like, it's, it's a very funny example that I just saw this week.

There is a company out there. I'm not going to mention names. What they have done is their developer advocacy team took an example of a project that a customer built on their platform. And they just created an open-source version of it using AI. So, so you have a, you are a company, you are a hosting provider.

Somebody comes and uses your platform to host an app that they are using it to make money. And then you took the idea and you just made a copy of it and you open-sourced it, right? It's like, when I saw this, I was like, what, what kind of level of like... Like not understanding how NDAs work and how, how disclosure is supposed to work.

And, and like, I mean, if, if it was me who was the customer of this platform, I would sue them because that's not, that's not supposed to be like, that's not this, that's how it's not supposed to work. Right. So, so on the topic of privacy that we said this before, like privacy is going to be the number one problem that everybody will have to care about going forward.

I agree.

Matt Brickey: One of the models will be. Affected, right? Bias, hacking, whatever, whatever you want to say it. And then the whole thing will be thrown up in a, in a twist. And, and everybody be like, well, how do I, how do I get my own? I don't want to, I don't want to use any of that.

Abdel Sghiouar: Exactly. Yeah. But to answer your question about the cloud stuff.

I think that companies are coming around now to understanding what the value they can have from cloud. I think that the problem is when, when we talk about cloud, we have to talk about it in the sense of what are we trying to solve, what are we trying to optimize for, right? I know that there are a bunch of, again, loud voices on the internet writing stuff that have gone, Oh no, we have gone away from cloud back to hosting our own stuff and it just worked for us.

We saved however couple of million dollars they managed to save, right? And that to me is not an indication that, that, I mean, of course, with this kind of content, when it's created and put out there on the internet, it's quite divisive because everybody who is pro cloud goes, no, no, no, no, you don't know what you're missing because Now you're spending more engineering time managing your own hardware and software instead of like letting somebody else doing it for you, blah, blah, the common arguments for why you want to go to cloud.

And then those who are like pro or anti cloud, they go like, ah, this is the right way to go. You own the hardware, you own the software, you're like, you don't own anything. Or the other, the other argument is typically vendor lock in, right? The other argument is like, oh yeah, I'm vendor locked in. Now we're not vendor locked in anymore.

And my typical answer to that question is, okay, you're not vendor locked in anymore. Tell me how long it will take you to migrate your application from a programming language to another. Just, just like, let's, let's stop using Java and let's rewrite everything in Go. How long will that take you? Three months?

You are VanderLocked in by default, period. Right? So, so it's, it's, it's a useless argument for me. But so, so the main question you have to answer is, initially, what am I trying to optimize for? Am I trying to optimize for resources? I want to keep running the same system. with the same number of people costing me the same amount of money, but I want to scale it up.

And I know that with this team, we are not going to be able to do it because we'll just add more servers, add more complexity. So let's just run servers from somebody else, from somebody who knows how to run, run those servers very well. It doesn't have to be a big cloud provider. It doesn't have to be AWS and Google and Azure.

It can be whatever, right? So this is, this is probably number one. Like what, what are we trying to optimize for? The second thing is what is the, what is, I think probably is the first question is. Ask yourself, what is your core business? Because if your core business is flipping burgers on a grill and selling them to people, there is no business for you running servers, right?

That's not what you're supposed to be doing. Correct. For you, IT is probably even a cost center. That's how you consider it. So why are we like renting server rooms and, and, and, and, and, like, you know, paying for data centers and cooling and electricity and all that stuff, right? So, I think the third one would be, And this is, it's funny to me that we have to make these statements publicly, that people don't think about it.

It's, it's, it's a funny thing to me. You do not have to do exactly the same thing that everybody else is doing, right? You don't have to follow Uber. You don't have to follow Google. You don't have to follow Facebook. You are probably not at the size of any of these companies. So if, when you go to the blog channel, the tech blog channel of one of these big companies, where they say, Oh, we migrated from X to Y because X, Y, Z.

You're probably not have, you don't probably have the same number of requests per second those companies have. So you don't have this set of problems, right? There is a scale in running a distributed system, at which you need to take creative ways to solve them because of the scale, because of the size.

And not everybody have that size. So, so, so you have to ask yourself the question of like, am I the Uber of the world? If you are, then good. Maybe then you should follow what Uber is doing. If you are not, then... Maybe you are good enough with two servers, right? So fundamentally, it's a thinking problem that people don't do.

Like, people don't sit down and just use their brain to think about what we are trying to solve. And just, everybody just jumps on the internet because this person read that article from this. And then that person told them, this is how we should go. And therefore, this is how we should go. And this is where good engineering managers come, by the way, speaking of as managers, right?

(45:21) The Evolution of Cloud with Kubernetes and Containers

Matt Brickey: That's a fascinating answer. Cause you're right. There's a lot of lack of looking in the mirror and admitting what you are versus what you think you are. And then making decisions based on that. That's a great answer. So let's get one near and dear, close to your heart. So how have containers and Kubernetes evolved the infrastructure that we're talking about today?

Abdel Sghiouar: Oh quite a lot, I would say I mean, everybody, I mean, everybody, it's a, it's a kind of bad statement to make it this way. So everybody's running containers, but like pretty much pretty, for me, the sign that containers are here to stay was the fact that the big enterprise programming languages, your, your, your Java and your dot net.

starting to optimize their run times for containers, right? Yep. When Docker came out initially, it did not play very well with the JVM. So you could use Docker to run a Java app, but it just doesn't work very well, doesn't know how to handle memory, doesn't know how to do a lot of things. The moment Java or the Java people came around and said, Oh, we're going to optimize the JVM for Docker or for containers.

That's a good sign that this is technology that's going to be sticking around, right? Kubernetes-wise, I think that Kubernetes came at a moment when there was no other alternative to doing orchestration at a big scale, right? And it just worked for a lot of people, because before... Like, there was tools, there was Mesos, I think I remember Mesos, Apache Mesos, there was OpenStack, there was, like, of course, VMware, which is paid, and all these things pretty much did more or less the same thing, just orchestration, right?

Just, like, a bunch of workloads running on a bunch of distributed computers, make sure they're all running all the time, when they die, restart them, and that's it. That's essentially, fundamentally, what Kubernetes does, or what it was designed to do. It just, like, came at a time when there was no other alternative that can do it as quickly as...

Good and as easily, easily as in you could quickly understand how it works by just reading documentation. You can go from reading documentation to having an up and running Kubernetes cluster within 24 to 48 hours. Right? And will it have the same, will something else that have started around the same time had the same success?

Probably not. I mean, Docker swarm technically was announced the same week Kubernetes was announced. Docker Swarm didn't, didn't catch up, right? Apache Mesos was there, a lot of people were using it and it just died out. OpenStack still exists, or OpenStack or the other one. Open, whatever, OpenShift?

OpenShift, OpenStack. OpenStack was there and OpenShift came later. But that's more for enterprise, that's more for companies who already work with Red Hat and they have Red Hat as their provider, they just want to continue working with the same people they are used to work with. So, so yeah, I think, I think it fundamentally changed the way we do stuff in the sense that now we don't need to write bash scripts and you don't need to write like chef or, or puppet recipes anymore.

And you don't need to do all this basic like stuff that we used to do to manage hardware. You can just let Kubernetes do that on your behalf. What's the future like? I have no idea. I think Kubernetes is going to stay for, in my opinion. There is more and more people coming into it. There are more and more people using it.

And I think what's going to happen is potentially the platform or the tool itself. will become an implementation detail at some point. There is a lot of, there's this whole new thing of platform engineering where they consider Kubernetes is just a component in the entire stack because you still need a lot of things to install on top of it to make it actually useful for developers, right?

And we're going to, we're heading toward the space where the, where we'll be talking less about Kubernetes and we'll be talking more about platforms and whether these platforms are built on Kubernetes or not. It's, it's not, it's not important. That's an implementation detail. What's important is like the UI, the kind of user experience or the UX that developers get exposed to.

There will be still people that have to care about that Kubernetes cluster under the hood and have the knowledge for it, but it will If I, if I can give an example, which I think is a very bad example, it will become the mainframe of the cloud native space, which is it's there. It's running. No one cares about it.

Matt Brickey: Right. So, so yeah, that's, that's, I think how the future looks like, are there any up and comers or open source projects outside of Kubernetes that you're keeping an eye on?

Abdel Sghiouar: Well, the second one is Istio service mesh. There is quite a lot of things happening in that space. Yeah. I think that they are important and they are going to be very important in the future for many reasons.

One of them is this privacy we've been talking about. Like making sure that people can implement zero trust and security in a super easy way, especially in distributed systems like microservices-based systems. So that's Istio. Specifically Istio because not, not just because it was open sourced by Google.

Because by default, by far, it's the most mature project in the service mesh space, right? It just recently graduated to the CNCF, so it became like a graduated project in CNCF. And very recently, Cilium did, and I know that Cilium have also a mesh product but less popular, I think. And also because in the space of Service Mesh, there is a big shift happening with how the implementation used to be and how people are trying to solve the drawbacks in the old implementation and just like come up with new architectures.

So that's, that's another one. I guess the third one is this whole platform engineering space with projects like Backstage from Spotify. Which is like what we talked about, like how can we give developers. Access to that, like, nice UX, so they can take everything they need. So they can use everything the platform can allow them to use in an easy, simple way.

There is a funny conversation I've been having. So I, very quickly, I didn't interview, which we didn't publish yet, but I didn't interview with There is in Norway, there is something called NAV, and NAV is the Norwegian Wealth Fund Wealth Fund, right? So they, they, they are the people who provide pension, healthcare, and stuff to the entire country of Norway.

And these people are very, very, very forward thinking because they are running in cloud and they're running in Kubernetes and they're doing a bunch of things. And they have a project that they have open source called Nice. It's pronounced nice, but it's written in N A I S. So what it is, essentially, it's a framework that you run on top of Kubernetes.

And it allows you to deploy stuff to Kubernetes in an easy way. So you don't have to deal with the underlying raw Kubernetes objects. You can use their own definition, their own spec to run your app. And you can say, this is my app. It needs a database. It needs to backup. A publisher-subscriber queue type thing, and it just like kind of ties up all these like underlying components and just make them available to you as a developer quickly.

And during the conversation, we've been talking about one of my questions was like, so how do you, let's say, for example, I have a developer because they internally work with their development team and they have like 800 development team inside now. So the people I talked to are the people providing those services through nice.

So the question was like, well, how do you know if like. Suddenly you have 15 teams that needs a new feature. How do you go about implementing this new feature in the platform? Right? If 15 teams now go, Oh, we need Grafana. That's just a very stupid, stupid example. So how, what's like, what's the, what's your, how do you prioritize implementing features?

That's what is my question, right? Or is it based on people asking for it? Is it based on? And the honest answer I got from one of the people I was interviewing was, well, we know, we talk to people, et cetera, extra, we have a product manager, but to be honest with you, 99 percent of developers just want the same thing.

They just want a place to run their app and they don't care about anything else. Right? So this goes back to this whole thing that I, I, I say jokingly all the time that every single developer, every single team and every single customer think that their use case is unique, is unique. Which it isn't, everybody just needs the same set of things.

So, so, so yeah, so that's, that's essentially what I'm keeping an eye on. And yeah, Kubernetes is still backstage, platform engineering in general. So

Matt Brickey: one of these days we might find out just like they did when they did a study on how many plots for movies there are. There's actually only six plots, repeat, repeat, repeat, right.

So we'll find that, that the same thing in the future. And speaking of the future, so I wanted to, I wanted to wrap up with a few, get me, put you on the spot a little bit, get a crystal ball out and ask you some what's going to happen in the future. So the next five years, what skills do you think are going to be the most in demand?

AI stuff. Yeah. That's like, that's like the whole thing on the, on the market today. So how would you prepare if you were talking to a bunch of college freshmen right now, that how would you prepare them to be ready for that AI push?

Abdel Sghiouar: I guess go and read a bunch of books to understand where AI came from.

It's not a new technology. It has existed for literally 25 years. Very simple example, very stupid example, spam filters in, in mail. That's AI. That's machine learning, right? So go pick any book written by any famous actor, author, and just read it to understand the history of this technology. This is, this is, this is a recommendation I would give to anybody coming to tech.

Understand the history of stuff. If you understand the history of stuff, you will understand why we're using certain things and why we're not using certain things anymore.

Matt Brickey: So it's more evolutionary than revolutionary, right?

Abdel Sghiouar: It's, it's evolutionary and extremely cyclical. Yes. Things come back every 10 to 15 years.

That's the tech industry for you, right?

(54:08) Outro - Challenges to Training the Next Generation of Engineers

Matt Brickey: Yeah, we, yeah, we burn out of ideas pretty quick, but then we, we will rekindle them when the time's right. That's, that's a fair statement right there. Exactly. So what concerns you the most about training the next generation of engineers?

Abdel Sghiouar: Right, so go, going back to the college freshmen, what's concerning you when you, when you think about that, people not understanding security.

Or people not understand insecurity threats. It's a, it's a huge problem. It's gonna be become a big a number one problem. I was watching a 60 minutes, I watch 60 minutes all the time. I was watching a 60 minutes episode that they put out about the five oldest national security agencies in the world.

So the FBI, the CIS CI Ss, whatever it's CI. Yeah. The British one. The New Zealand one. The Australian ones, like five people who are the head of all these like agencies. And they, they together this month or last month, they were together for the first time in person in the U. S. having a meeting.

And the interview with 60 Minutes was like, number one problem is privacy. And it's privacy beyond, it's privacy and security beyond, oh, my password or my credit card got stolen. It's actually because our life is becoming so intertwined with the internet. Actors or bad actors now are trying to do things that impact people in real life.

I think you're from the U. S. So you remember the Continental Pipeline saga that happened? Yep. Company got hacked, no more gas was, no more oil was delivered to the East Coast. It caused mayhem, two hours waiting lines in pumps because people were scared that there will be no more gas. So that's an example Or just imagine how much your life thing depends on the credit card.

And what if the credit card stops working? What are you going to do? How do you go in and out of parkings? And how do you pay for stuff and buy things and eat and do stuff, right? So that's, that's meant, that's going to be like the number one problem going forward. And I think we're training more people on actually not understanding how to write safe code.

And I don't think there is enough good tools in the market that can tell you whether something is easily hackable or not. I know it's a big problem. If somebody, if somebody is looking for an idea to start a startup that can make billions, just use AI for this.

Matt Brickey: I would agree with you on that. Yes, sure.

Prevent them from, prevent the opposite use case of everyone using the same thing. It gets hacked once and it's just a widespread problem. Exactly. Agree with you. So let's flip that coin a little bit. What gives you hope about the new generation of engineers that you're working with? It's,

Abdel Sghiouar: I think that people come into the industry, they are coming with a pretty, pretty open mind, more and more pretty open mind. So like, like things not, not being done a certain way. Right. I guess if I have to talk about something that I personally faced, I would say that the whole problem of diversity in the tech industry is becoming less and less and problem because the industry is becoming more accommodating to people from all backgrounds.

Right. Correct. And if I want to give a very precise example, I would say the open source space. I mean, the open-source space. 10, 15 years ago was extremely toxic. It was a male-dominated space and it was really bad for new contributors.

And if you just compare that to the cloud native space, for example, where the projects are more accommodating, they're nicer. KubeCon, there is always a Contributor Summit, which is like a full day where all the contributors to all these open source projects get together and they discuss like, like working and...

Burnouts and stuff like that. And I think to this, on this line of open source, also we're not doing, we're more and more doing open source differently than we used to do it before. Open source 15 years ago meant a bunch of people in their garage. with like probably a dark, dark room, just hacking code in their spare time.

And now there are people actually, yeah, in a hoodie. You gotta wear a hoodie. Exactly. I gotta wear a hoodie always. And now, now I think companies are actually paying people to actually work on maintaining open source projects. So that's, that's, that's a very good positive change that I'm seeing. It's coming with its own drawbacks, but it's a positive change in general.

Matt Brickey: that's a good answer. Good, very, very well-thought-out answer. Do you think in the future, say five to 10 years, that we'll be able to close the skill gap that we're, we're dealing with now, or do you think it'll open wider?

Abdel Sghiouar: Oh, I don't think so. I think we are growing, we're growing more and more specialized and we are pushing people to be more specialized.

I don't think we are, I don't think there is any interest slash any like efforts, effective effort to try to teach people the fundamentals. And I'm not sure if that's a good thing or a bad thing. I, the only, the only reason I would say that's a bad thing is again for the security stuff we're talking about.

All the safety, all the privacy stuff. I am hopeful that in five or 10 years, somebody will build a tool that can analyze code and tell you this is going to cause a security problem. And, and, and, and hopefully we'll be able to do that through, through all the open source stuff that everybody relies on.

I do, I do a lot of talks in the software supply chain security space. And that's like a big kind of ball of warm. Where basically nowadays, no one is really writing code anymore. Everybody's just pulling a bunch of dependencies from the internet and then writing whatever their business need them to write, which is the business logic, right?

No one is building libraries. No one is doing that stuff. And that's, that scares me because if one bug is introduced in, or one backdoor is introduced into one popular library in, say, Node. js, for lack of better examples, then everybody doesn't npm install and suddenly everybody has the same backdoor, right?

So hopefully, I'm, I'm hopeful in a few years from now we will be able to use AI in a kind of effective way to solve these kind of problems. Good. So

Matt Brickey: yeah, we'll put our hope around AI, but you don't think the people will actually fill in the gap. I don't either.

Abdel Sghiouar: I don't think so. No, I'm not very hopeful for that.

Matt Brickey: Right. So just one more, one more question on that, on that, and then we'll wrap up with a final question. So what roles or job titles do you expect to see to grow in the future? So is again, back to that college freshmen, what should they be looking at, you know, four or five years from now when they graduate and enter the workforce?

Abdel Sghiouar: Oh, I mean, if, if I have one advice for stuff to look at would be the security space. I mean, if you want to make money, go into the security space. It's a very in demand. Job right now and there isn't that many people doing it or there isn't that many people who are really good at it Doing it well.

Yeah. Oh, yeah doing it well So go into security study cyber security study Do participate in some bug bounty programs. There are like tons of them on the internet learn how to hack your way around Linux and how to debug Linux and how to, like, look at processes and be, be good at the command line because that's most of what security is about.

And I guarantee you, you will have very, very well-paid jobs once you graduate from university.

Matt Brickey: Yeah, I would agree. I think it would be a very prevalent field. Final question. And I thank you for your time today. Your answers have been fantastic. So if our audience forgot everything that you said, except one thing, what would you want them to leave with?

Abdel Sghiouar: I mean, there is more hope in the way this industry is going forward. We're becoming more accommodating, we're becoming more, more, more, like more open to people from all backgrounds. And also I think if I want to leave people with anything. I don't think that, I care a lot about politics. I follow politics.

I don't think that politics and tech mixes up. I think tech is a beast that progresses at its own pace, regardless of politics. So, don't get discouraged by looking at what's happening around the world and thinking that that's going to somehow negatively impact people in the tech industry. That's not going to happen.

Tech is going to be there regardless of the politics and regardless of who is president, who is doing what, whatever, we're still going to need the banks and we're still going to need credit cards and we're still going to need developers. That's a great final answer. I

Matt Brickey: love that answer. Abdel, thank you for your time. And I thank our audience for our time.

Abdel Sghiouar: Thank you for your time.