Speaker 1: 0:00

This is the Art of Network Engineering podcast.

Speaker 2: 0:11

In this podcast, we explore tools, technologies and talented people. We introduce new information that will expand your skill sets and toolbox and share the stories of fellow network engineers.

Speaker 1: 0:23

Welcome to the Art of Network Engineering podcast. My name is Andy Laptev and this episode I am joined by the one, the only, Kevin Nanz, an adjacent node. How are you, Kevin?

Speaker 3: 0:35

I'm doing fabulous, other than my internet having issues, I'm doing great. I'm still recovering from I'm eating right now. I'm sorry.

Speaker 1: 0:42

I was going to say what are you eating? Eating, but you wait until the show started. Start chewing.

Speaker 3: 0:44

I have a jar of candy next to me from halloween and I can't have it in my vicinity without without reaching for something what's your favorite halloween candy? I like a fruity gummy kind of guy, so like gummy bears um any kind of, uh like sweet tarts, that kind of stuff.

Speaker 1: 1:03

It's sweet, yeah, sweet stuff it's like my wife and kids. I'm like a reese's peanut butter cup guy all the way chocolate and peanut butter. It's like the, the marriage. That's just the perfect. Yeah, made in heaven, right? So, besides your halloween, uh, extravaganza over there of candy. How are things in kevin land? How are things?

Speaker 3: 1:19

they're good. They're good, um. Um, just been busy with work, nothing too crazy, still recovering from the election night. Election was recent. I didn't sleep at all. I'm not going to get political at all, but I just love the drama of the night. We're like what state's turning this and the votes I live off that. That's drama. I love it Better than any kind of fiction out there. It's fun for me. It is a lot of drama and I'm happy to hear you're busy at work. Um, how is that? By the way, how is it to be busy at work? You jealous, andy, don't you? You'll be a little bit, a little bit.

Speaker 1: 1:53

You'll be busy soon, don't worry about it. We're getting there. Well, kev, thanks for hopping on, buddy. It's always great to see you and um yeah, we're looking good. The internet looks solid right now. Um and uh, tonight's special guest is nicholas calcutta. Doc, do I have to call you dr nick? Hey, everybody.

Speaker 2: 2:09

Oh no, not yet. Not yet. No, couple years, 2026, hopefully, if I, you know, can do my homework on time.

Speaker 1: 2:16

So, nick, thanks for coming on. Tell the people, uh, who you are. What do you do, where do you work, what is your doctorate in? Let's, let's get a high level who this guy is before we dig into our topic.

Speaker 2: 2:25

Oh man, you got to ask all the hard questions here, right? So you know the Calcutta, right. So we'll start off. I've been networking IT space really for what? Over 12 years now, plus Officially 12 years plus, but you know longer than that. I got my bachelor's degree at Florida State College of Jacksonville in networking technologies it's much longer, but they added telecommunications in there, because they used to teach telecommunications too as well. Masters at Western Governors University we had a couple guys in Discord also did the same thing there Got a master's in IT management and currently going to USF for a doctorate in business administration.

Speaker 2: 3:05

And yeah, usf man, it's great. After this podcast I'm driving right down, got a hotel in St Pete, get there probably about 1, 2 am, then back up at 5 am, get ready for class at 7.30. So yeah, it's the sacrifices we got to get it done.

Speaker 1: 3:22

But yeah, so that's education-wise, I think you're the most educated network engineer that I know so far, besides Russ White.

Speaker 3: 3:30

There's Russ.

Speaker 2: 3:30

White and Nick Calcutta. That's education, but certs. I actually don't have a Cisco cert, so you know, took the CCNA once a long time ago.

Speaker 1: 3:39

But you've also been a network admin. Right, You've done the job. It's not just all education, oh yeah.

Speaker 2: 3:45

Done the job for many, many years in multiple different spaces, was that post like bachelor's, did you like?

Speaker 1: 3:51

hey, I got my networking bachelor's degree, let me go get a job.

Speaker 2: 3:54

So I actually got networking job, official networking job, like right around my associates, honestly. So I was working as a desktop support technician for the local government and the guy that was a network admin was leaving. I'm like, well, nick, you're going to school for it and you kind of know what he knows, so you want to go ahead and do it. I'm like, heck, yeah, why the hell wouldn't I?

Speaker 3: 4:15

Let's just do. This Sounds so much like government.

Speaker 2: 4:18

Oh yeah, hey, it was great, I got paid. I got a nice little raise, a five thousand dollar bump, right you know. And uh, my first project was getting rid of two core 4507s and then rebuilding the whole bomb. We had a bomb from presidio and I said, no, that looks like crap, I'm not doing that. We did the whole bomb myself and um did some 30, 50 chassis. The collapse core model is beautiful and half the price, right.

Speaker 1: 4:47

So that was great there, so you belong. That's the qualification portion of our conversation is you have an education and graduate education in networking. You have over a decade of experience doing the job. You are one of us, so that's why.

Speaker 3: 5:03

I asked so that the people watching yeah, the people watching and listening Okay, he's one of us, so that's, that's why I asked, so that the people watching.

Speaker 2: 5:06

Yeah, the people watching, listening.

Speaker 1: 5:07

Okay, he's, he's one of us and generally friday yeah, generally our guests are mostly us, but I just want to let people know because they might be fooled. At least I was initially by like this guy's a doctor, he's a college professor, like has this guy ever worked on a network? But you've done the job, you've you, but you've done the job, you've done the thing. So with that, I guess, said you and I spoke, I don't know, it was a month or two ago and you were presenting something at a tech conference, right, and I know that you wanted like a peer review on the doc and stuff. So I think that's kind of where we want to hang out on. This episode is what you talked about.

Speaker 1: 5:47

Now, correct me if I'm wrong, but my understanding is this presentation was kind of like the history of networking. Like, hey, we started back in the day with ARPANET, here's what you had to know, here were the protocols. And then you kind of do a time walk of over the years. Here's how things progressed, here was the growth, here's what we had to know. And then you kind of do a forward looking portion of like and you know what, if you want to do this job today, tomorrow and moving forward, or you know if you're in school and you want to do this in the future. You know these are. These are the trends you've seen. I really like the analytical like we all have an opinion, right. I these are the trends you've seen. I really like the analytical like we all have an opinion, right. I got laid off last year and I have opinions on like, oh, these are the things you got to know because I didn't know them. But that's just my little echo chamber, right, but yours looked really analytical and historical. So I'd love to kind of just walk the audience through this information that you found.

Speaker 1: 6:39

Not that we want to spend 35 minutes on ARPANET, but where would you like to start? Because really, where I'm going with this is at the end of this conversation. What do you need to know? What competencies do you need to get a job in network engineering, to stay employed in network engineering, to climb through the ranks and be promoted in network engineering and to stay employed in the future, because there have been a couple of disruptive technologies in the past four or five years that you know it could be seen as huh, I better learn that, or else. So, um, I'll shut up there, but that's kind of to just to set the scene. So where would you like to start?

Speaker 2: 7:17

I mean we can start off like let's, let's look at a brief history of the past. So tcp was developed in 1974. The first email was sent out in 1971. You know I came on the edge of frame relay um mainly just ethernet ospf, um ergrp right.

Speaker 1: 7:33

So really when you get in the field, you mean when you were working in the field, yeah, yeah, and we had the ugliest ospf ring.

Speaker 2: 7:40

we had one man network, um for all of our sites with the one giant OSPF. That's why every time the question is well, what's these stubby areas? Stubby areas, no, no, no, let the CPU cook. I pay for all that CPU on that router. I'm getting all of it done, okay. So you know, at first I'll say when I started going to school, for networking a lot of it's command line, there was no helpful GUIs, right, there's no ClickOps Terminal. You save that terminal, that serial port connector to console cable with your life. Right, that blue cable was your life. If you lost it, you were hunting down for it.

Speaker 1: 8:20

And we interacted with the infrastructure via CLI, because that's all we had.

Speaker 2: 8:23

That's all we had that Telnet, ssh, hell, sometimes SSH, if you're lucky. Sometimes it's only Telnet because the switches or routers are so old.

Speaker 3: 8:31

Yeah.

Speaker 2: 8:31

Right Dial-up modems too. Still dial-up modems in the network Did either of you guys have a dial-up modem.

Speaker 3: 8:39

Yeah, I did. I'm going to show my age here because I did. Yeah, okay.

Speaker 1: 8:44

My first internet connection for college was a dial-up. I remember the screaming sound and all that.

Speaker 3: 8:48

You're a little older than me, but when I was a kid I would sneak, not that much now I know how old you are. I would sneak out in the middle of the night and go to the computer room. My parents were sleeping and had to put a blanket over the motives.

Speaker 1: 8:58

Oh, really so you couldn't.

Speaker 2: 9:00

I've never had that problem. We just turned on a night and just let it go. That thing was always dialed in. Was it on a gateway computer, Kev. I think so probably. Yeah, he looks like he had a cow computer right.

Speaker 3: 9:16

The commercials were effective, man. The commercials made it fun they were.

Speaker 2: 9:19

I mean, I had a gateway. In middle school my mother had a gateway. She used it for AOL. We used it to play command and conquer in aol right you know what amazed that so.

Speaker 1: 9:31

So I'm going to interrupt you for a second, then we'll jump back looking at your timeline here on the, on the and and for the, the folks listening and watching. We'll um this document's on nick's website so we'll put it in the show notes and you can go out and take a look at it. But I see Ethernet was created in 1975. I've learned in the past year or two that AIML was created well before that. It's a much older technology. We are now today in 2024. Help me out with math. I don't know how long 75 to 2024 is, but about 49 years. Kevin's counting, 49 years, kevin's counting. And Ethernet still can't handle the AI HPC workloads because it's not able to do a lossless fabric like InfiniBand. So what I find interesting? Just one little piece of this timeline that you're looking at Now again, the Ultra Ethernet Consortium are working on it.

Speaker 1: 10:19

They have a V1 spec coming out soon. There's actually a lot of real work happening and we're going to have Dr Jay Metz on in a few weeks to talk about what they're doing there. But just one little piece of this, like to your point, there's all these technologies you look at. Ethernet is 49 years old, ai is older than it. But all of a sudden AI ML gets hot and then we're like in the networking space like, oh crap, you know 49 year old ethernet we have to revamp and rework because it can't handle, it can't create a lossless fabric. But we had 49 years to do it, but AIML didn't pop right. So it's kind of interesting like those disruptive things that come up that force. You know innovation in the network. So I'm sorry to interrupt, but I just jumped out at you.

Speaker 2: 10:59

I mean I think honestly that that goes into more of the future networking, right. So as our compute stacks have gotten better over the years, right, we can do crazier things, right. I remember the best time I had well, one of the greatest things you know when a gigabit internet where my limitation was no longer how fast my internet was to download a game or something, it was how fast my hard drive could write right and steam got pegged out, not a network, I got pegged out on cpu cycles or hard drive cycles. I was like, okay, so now, now the constraints move somewhere else. That's, that's nice and we kind of.

Speaker 2: 11:32

It's kind of how networking evolved, right, once you're on-prem systems there, a lot of just man networks, your frame relay, mpls, stuff, so your chassis switches, your access layer switches, that didn't have your Wi-Fi components built into them, right, there's no controllers built in there. There's no ROM on there large enough to throw a Lama model on there or even a router to run Viptela or another. Bring your own OS right. So we had a lot of more. This device does one thing and it kind of SDd. I remember when sd networks came out, right, not sd-wan. So I was like you know, sd networks. That's interesting, you know. We had a little class of it in college and I'm like I get it. But enterprise wise, I've only seen sd software defined networking really implemented effectively in large-scale enterprises or products like VMware, nsx, stuff like that. That's where software-defined networking really popped up. But I think it laid backbones for other things.

Speaker 1: 12:34

So you were talking about the internet. So that was like the explosion in the 90s, right, was that the big 90s to 2000s, the dot-com bomb. I'm trying to keep us in time here, like where, where we are in the technical evolution. So like, right, the internet. You mentioned when you started frame relay. I'm a fellow. I'm a fellow frame relayer. They told me when I took my CCNA I'd never see it in production and my first job in production was migrating like 75, you know, bank branches off a frame relay to MPpls like oh look frame relays here.

Speaker 1: 13:06

I know what a delce is, yay look at the delce mapping, you know stuff they say, we we'd never see um I'm pretty sure there's a government network somewhere.

Speaker 2: 13:15

He's in frame relay still.

Speaker 3: 13:17

Oh, 100, 100 not mine, I had not one lines on t1 lines for a while.

Speaker 1: 13:22

So so do we want to jump to the 2000s? Because really I I don't want to get too down in the minutia of the little technical stuff at each decade. But I guess what I'm trying to do is have us paint a story of the evolution and networking so that we can wind up like today, oh yeah, and then in the future.

Speaker 2: 13:41

I mean, look at Wi-Fi so as Wi-Fi got better, or Ethernet got better too.

Speaker 1: 13:58

we got out of 10 megabits to 100 megabits to one gigabit full duplex and that kind of stuck there for a while.

Speaker 2: 14:01

Only recently, 10 gigabit Ethernet has become consumer grade right for price wise.

Speaker 1: 14:02

I can't imagine what my dial-up was what were they like 64K. I don't even remember 56K modem, right 56K. I have a gig fiber handoff symmetric gig at my house now and I started at 56 K.

Speaker 2: 14:09

And that's what made T one's great Cause. T ones were at 1.54 megabits per second, symmetrical, and they were like, oh, this is blazing fast. This is great, right, you still need to know your. You know fundamentals, right, how to route, where to route to that. That hasn't changed. Right, the basics, tcp, your osi layers, hasn't changed. You're always going to need critical analytical thing skills. That's never going to change. You know one thing I can say about all it those two things never going to change. Um, packet analysis. I think today I was actually john capianca wrote the post where knowing Wireshark skills is kind of a lost art doing packet analysis, and I agree with it. It's been a lost art. I used to have I still have like all the Wireshark Bible on my desk. It's like thick old book. You know for all the graphs, everything that it can do to help you.

Speaker 1: 15:02

I'm so embarrassed to say I've never ran a PCAP.

Speaker 2: 15:05

Well embarrassed or lucky know.

Speaker 1: 15:08

Yeah, now, my excuse is I worked in such large shops that we had people to pull them for us, like there was a tap and somebody would pull them and then they would, you know, throw them on the bridge and the super bright architects would look and be like you know we'd be. We'd be at hours smashing our head into the wall and the person that solved the problem was always the person that looked at the PCAP. I go, well, this is what's happening. So we I remember having Chris Greer on I think that's his name the packet pilot guy. I'm sorry if I'm bastardizing his name, but he's. He's like the wire shark guy, I think, or one of them, and I'm dying to get together with him and he's going to set me up and I'd like to do some of that, because I just can't believe I've gotten this far and not done captures. Have you guys both run like Wireshark captures and looked at PCAPs and stuff?

Speaker 3: 15:51

Oh yeah, it's magical right, and I worked at the ISP. Never that was a never, I think, but ever since I've been into a non-ISP environment environment it's been. You know, logs don't tell you everything, right, gui logs don't show you everything. Um, even cli logs don't show you anything. The only thing that tells you the truth every single time is a packet capture every single time right.

Speaker 1: 16:12

Pcap never lies.

Speaker 2: 16:13

Pcap didn't happen right, yeah, sand knowledge used to be a bigger problem. Back with fiber, channel sands were more prevalent. I remember my first time trying to work on a was a silk, silkworm or brocade, I can't remember, it was just. I'm like you know, let me see a line of this, let's see if I can get it to work, and that was.

Speaker 1: 16:30

That was atrocious that's a storage array network yeah, storage, storage right now.

Speaker 3: 16:35

Yeah, storage is that what that?

Speaker 2: 16:36

server stuff the term server yeah, and then you know the nexus line, you know, and that's did Channel as well, that was. That's always a fun tool.

Speaker 1: 16:46

And I see you have on your slide like some scripting and this was the 90s and 2010s and as much as I'd like to say, everywhere I work, nobody was scripting. There were a few cats that were doing some scripts. It wasn't widely deployed. But if you get on a call with somebody and work on something like what is that Like oh, and work on something like what is that like oh, it's just a script I run. When I do changes, I'm like whoa so oh yeah

Speaker 2: 17:05

there always seem to be one or two cats around like oh, yeah, I you know I I do some scripting some scripting and routing protocols, definitely, you know, on those things. Just some scripting, not too much. I mean, we're talking about 90s to 2010s, apis on network switches, not that crazy, you know, not as open as were you, I think, towards more of the 2010s, maybe some Yang models.

Speaker 1: 17:27

What was the scripting back then? Was it like Bash? I might even be saying dumb?

Speaker 2: 17:30

things. Yeah, bash, I'm pretty sure, just Bash.

Speaker 1: 17:32

Just Bash. Yeah, you know, Like today all the rage is Python, but back then it was Bash. Right, Bash commands.

Speaker 1: 17:38

Yeah maybe it depends on the platform you ran on too. Not every switch is programmable like that. So before we go on to this next slide, I want to say something that scott robon said to me at a meeting we were at is that this is all additive and you alluded to that earlier. Right, like these are things you know networking fundamentals, route switch, p caps, like everything you just said. It doesn't go away or get replaced, which is part of what annoys me about this career, like you know what I mean. It's always more and more and more for relatively the the same amount of money.

Speaker 1: 18:14

And I know I sound like like a whiny cry baby when I say that, but um it scott helped me reframe that because I said to him once, eight or nine months ago, that you know my skills had atrophied and that I didn't keep up with. And he's like well, those, those skills are still relevant. You just have to add on to them, to apply them to these different contexts in these new systems. I'm like, oh, because in my mind, route switch didn't matter anymore. Right, and then we'll get into it.

Speaker 1: 18:41

But like ClickOps, we again somebody else was talking about. You know, you have these seasoned network engineers that come in and they've only been in GUI dashboards and they have no idea how anything under the hood works. They just click things and it works. So, um, you know, having that fundamental knowledge, hopefully all these tools that we're making that are abstracting everything away and not making us have to know the protocols and what's happening in the CLI Like I don't know what the future looks like and I know we're going to get there, but if nobody knows how the stuff's working and they're all working in automated GUI dashboards, like there's probably going to be three people left that they call when all hell breaks loose.

Speaker 3: 19:16

That's the thing. Yeah, if that's all you know, it's what's that? Mike Tyson quote you know everyone's got a plan until you get punched in the face, sort of like that, where you know the GUI, you know what, you know, kind of how you press this button, press this button and it works. Well, that's fine until it doesn't work. And once it doesn't work anymore, then you're, you're worthless. You have to know the underlying, you have to know how. You know the basics.

Speaker 1: 19:35

And as a network engineer, you have to know it. But then again, at said, well, you call the vendor when you need the person that knows right now. I hate that. No, I know it's not a great way to run your network. Um, but that was their answer to you know. Well, click ops is bad and guis are like well, you just call the vendor and you get the expert there I mean it's.

Speaker 3: 19:54

It's one way to do it as a public. I wouldn't want a public servant who has to pay for all that stuff. We just don't have the money to have. You know the expensive service contracts.

Speaker 1: 20:04

Okay, that's a really good point.

Speaker 2: 20:06

I paid for a Cisco smart net on the core switch for a long time at public service. I never called TAC once, only because that just took too long. They're like okay, got these longs. I'm like that's too long, buddy, I'll figure it out. I always followed the burn the boats methodology. There's no one here to rescue you. You got to figure it out.

Speaker 3: 20:28

That's the best way to learn too, because the next time it happens it's going to break again. If it breaks one time, it's going to break again. I don't know when, Maybe 10 years from now, but it will break again. That's where you're really making the money. There you can fix it like that.

Speaker 2: 20:41

You're like oh, you're a hero you can fix it like that and you're like oh yeah, you're amazing and that really goes to do. You have the right management over your head to allow that Right. So you, you can fail a little bit or take a little bit longer to fix it, but you fix it. You learn how to fix it Right, there wasn't somebody saying well, it's down, let's go call tackle.

Speaker 3: 21:08

Go call somebody right, get them on the phone right, and that's. I mean that that's going to vary based on the environment and what's affected. Like I was in the isp, where if something broke and we had, you know, thousands of customers down, then there there was no time. That was. The standard operating procedure was to call tech right away, you know, keep working on the issue but get them engaged as soon as possible, whereas, like at the university, we had, like you know, 200 down. We could take a little more time. It gave us a little more leeway to try to troubleshoot it and figure it out.

Speaker 1: 21:29

Yeah, exactly, it's got to balance it. Go sack some hackies. Exactly.

Speaker 3: 21:35

Kids don't. They'll be up in an hour. They're just playing their Roblox or whatever. They're probably not playing Roblox, but you get the idea.

Speaker 2: 21:41

They're just games, hey hack playing roblox.

Speaker 1: 21:52

But you know, you know, you get the idea. There's games hacky sack's dangerous man. I broke my leg playing hacky sack. That's why, that's why I'm tax port, that's right. X games on that one. So so, let's, let's say the dirty word, the cloud knowledge. You know everything, yeah, I know. So now we're into, like, network engineering skills 2010s to now, right like now, we're coming up to present day. We went from arpanet and everything else we talked about in frame relay to like, all right, what do people need to know today? And I see some GUI familiarity, which we just talked about. Cloud knowledge um, that's a big one, right, I mean, everybody's in a public cloud and for me, when I got introduced to public cloud, man, it's weird, just weird. It's very different.

Speaker 1: 22:26

Like have you guys? Have you guys dabbled at all? Have you had to do any of that in any environments Daily? I'm in there daily and so when you got introduced to it, was it like, hey, nick, spend six weeks doing the vendors training before you have to right right now? Now, maybe because you're a doctor, it was easier for you. But was it a weird transition for you to go from like on-prem networking and how all that works to like, okay, now there's stuff in the cloud?

Speaker 1: 22:53

Because for me, my brain just had a really hard time. Like if you're in multi-cloud and we'll probably get there, but like if you have workloads all over the place and on-prem and then something breaks and you get a call, I'm like where is it? How was it built? Which place do I log into? What do I check when I'm in there? Where I was so familiar with on-prem, I got to get in my disk switches, look at my routing, see what's going on, check the logs. Like I was much more familiar with on-prem than like the cloud to me just seemed like this weird magical place where each CSP did it differently and I don't even know where our application is. That's having the problem.

Speaker 3: 23:27

Did you?

Speaker 1: 23:27

have any of those issues going to cloud, or was it easy for you?

Speaker 2: 23:31

Oh, it was a pain in the ass. I gotta say that, and I'm pretty sure my coworkers are on this. They'll tell you probably the same thing, right? They're probably on YouTube saying, yeah, yeah, it sucked. It was a horrible weekend. So I joined the company recently two and a half years and when I joined I was a network admin and they're just starting their cloud journey. Like hey, we're already keyways there. We're going to go ahead and move everything else there, right? Azure I kind of knew I had a class on it. I had $100 free credit, went ahead and tried it out Cool, let's do it. But when we migrated everything up to Azure and we were testing things out before that, I hated it, because all your nerd knobs are gone. It's a GUI dashboard but with no helpful features, and it just makes you realize a dev made this and no one talked to networking. It's like the reverse option, right?

Speaker 1: 24:24

Almost every automation tool I have tried to use and work on I have the same thought. A dev made this and they didn't talk to anybody in networking.

Speaker 2: 24:32

Yeah, I'm like, well, there's no QoS. I'm like, well, shit, that doesn't help out the problem there, shit that doesn't help out the problem there, right, I can't QoS something here. Also, moving to cloud latency, you find out and this is a Silicon Valley quote I love all the time right, it's where Guilfoyle and Dinesh, you guys, ever watch Silicon.

Speaker 3: 24:51

Valley. Oh yeah, Best show ever.

Speaker 2: 24:52

Oh yeah, oh, best show ever. I watch all the time. But they're yelling at each other and they're sitting there saying hey, they're yelling at each other and they're sitting there saying, hey, you know Dinesh, like, well, you know, if my code wasn't running your crappy servers, you know your network, it wouldn't be a problem. And he goes well, my network didn't apologize to your shitty code. It would have been better, right. And when you go to cloud you're going to find out if you have shitty code or not. Real fast. The networking shows there is no more latency, there's no more sub-sub-millisecond latency to hide bad calls the cloud, even the same sub in the cloud, it's going to expose it real fast. It's unforgiving, right. And the app guy's like well, it's taking a little bit longer. And you're like I don't see no problem here and I can't figure out why. And trying to pull a PCAP out of Azure is like asking for government to give you your tax refund December 31st. Huh, azure is like asking for government to give you your tax refund December 31st.

Speaker 2: 25:43

It was just like this is rough. So it kind of felt like at first managing a network with one hand behind your back, but the more you're into it. The more you know it, you're like, okay, it's not so scary. Honestly, tim and Alex and Chris helped out a lot in the Discord chat. When I was like, hey, I got this problem here. They showed us a little bit hey, here's the black hole on the gateway subnet. That probably did that. I'm like, okay, cool, thanks, man. It kind of had to rework how it worked, but fundamentally, once you understand what everything does, you're like, oh, this is just a normal network, just whoever made this was drunk drunk.

Speaker 3: 26:19

Oh, it's like all my networks got it well, there's a different.

Speaker 2: 26:22

There's a bourbon high class network drunk, it's like nightlight.

Speaker 1: 26:28

No, this is tequila networks I swear that's like a sasher. How, how long? I hate this question but I have to ask how long did it take you? It's like when somebody asks how long it's gonna take me to get my certification, like, stop it. Yeah, was it weeks, months, years for you to go? From what the hell is this insanity called weeks to like oh okay, weeks of being in and bashing your head into it, kind of thing?

Speaker 2: 26:55

it was like two solid weeks of okay, I got this. Let's just you know, do it. And honestly, I tell all this to my students too you got to get your feedback loops. It's not like, okay, this is how to do something. You got to get that feedback loop. Here's the goal. I'm making this server talk to this server. How do I do it? Okay now, how do I do that if I have these servers across vpn tunnels on two different azure descriptions? Okay, test that out, right. Okay, create those feedback loops where you're learning, testing, learning testing real fast and the iterative design. If you do that, it's going to retain a lot more. So we were moving fast, going fast and luckily I had a team with me that was in the same you know high speed boat, fast. So it was great. So we just kept going, doing and it retained. That helps. Retain is doing things. That's why you know lab every day. Right, it helps out for sure, right?

Speaker 1: 27:48

so what I'm going to keep coming back to is what do you need to be a successful network engineer now, in the future? You need route switch, which is everything we talked about for 30 minutes, and now we're cloud. Um, I don't know if it matters which provider you choose. If you want to pick on market share, you go AWS, right. I mean, if your company that you're working in is in a particular cloud, that's what you do, but if you're somebody that just wants to learn a cloud, AWS is very I mean Azure.

Speaker 2: 28:13

You get the free $100 credits, but I think AWS has a lot of free tier. Yeah, easy to use, but I mean most small businesses are going to be. Well, we have Microsoft 365, so we're probably going to have Azure right, so get Azure is good. Also, the naming I think in Azure is better because Azure calls them virtual machines versus AWS is EC2, right. Lots of compute instances, yeah.

Speaker 3: 28:35

Hold off a tongue, okay.

Speaker 1: 28:37

Automation, should we say the other dirty word?

Speaker 2: 28:39

I mean automation tools. They've gotten better. I mean Terraform, ansible, yep. I mean my automation tool back in the day was well, you know I'm trying this SSH open source tool that can configure my Cisco switch for me and it always broke. Right Now they've gotten so good, especially Python. I mean hell, help, look at pi, ats these automation tools or testing, it's just it really helps out.

Speaker 1: 29:07

I spent years crying about automation and the past two months I've just been learning python. I know it's been terrible, I'm, but I'm learning it like it's not. I guess it's like anything Like if you can learn I don't know, ospf, timers or, like you know, like the other stuff. I almost quit networking over subnetting because I just my brain exploded and I didn't understand it and I'm like what is this? I can't do this, I'm not smart enough, but you know, now I can subnet.

Speaker 1: 29:34

So like I think, if you can learn hard things and routing and switching are hard things in my opinion you know you can learn things like automation, but maybe for newer people it's not that big of a deal. Like, if you're a computer science major and you come out with like seven languages, great, I got to know Python, no big deal. So let me ask you this question Do you most of the network people I meet got into networking because they didn't like math and didn't like programming? Check, yeah, I, most of the network people I meet got into networking because they didn't like math and they didn't like programming Check Yep, yeah, I mean, that's why people get into networking, not all.

Speaker 1: 30:05

And again, dr Nick's going to tell us something different.

Speaker 3: 30:07

Dr, Nick's got a school list.

Speaker 1: 30:10

Most of the guys oh yeah, I took all this. Like Tim is an example that used to be on the show, like he was a computer science guy and didn't like coding, but then he got introduced to networking and he was like he had that moment we all have like, oh my God, this is amazing, this is what happens. Like, are you kidding me? There's this. You kind of fall in love with networking, and I don't know a lot of devs maybe they fall in love with coding. But the reason I think I was so resistant to automation for so long is because I thought networking was how to get away from that. Now here we are and if you look at any job description, you have to do some automation at the very least Ansible, right, and you should probably know some Python. So I'm just saying all that to say that somebody like me who cried for years that I didn't want to learn automation I couldn't learn automation, I failed out of programming computer science in college Like the stories we tell ourselves can really create the reality that we live in. And then it's like well, if I say I can't I wrote this article about storytelling recently and then I'll shut up, but I used to say I can't run because I have bad knees and because I told myself I couldn't run. I didn't run for 10 years because my knees were bad. Then I got laid off last December and I got depressed and I'm like, well, I can lay in bed, which isn't going to do anything, or I'll go out and run Cause I know I'll feel better and I'm going to run until I feel better. I ran 400 miles in eight months. I felt better. It got my mind and body where I needed it to be able to be productive in interviews.

Speaker 1: 31:33

But the difference between the guy who didn't run for 10 years and the guy who ran 400 miles in eight months was a story I told myself. I told myself I couldn't run. So I didn't run until one day I said I'm going to run and I ran. So it's the same with automation for me, I think for me. I don't know if it helps anybody Like you gotta be mindful of the stories you tell yourself, because if you tell yourself you can't learn Python, you're not going to learn Python. Once I told myself, if I want to get a job in networking again, I'm going to have to learn automation. I'm learning automation. It's not that bad, but I had to change how I thought about it.

Speaker 2: 32:05

There's a lot of resistance.

Speaker 1: 32:07

Well, yeah, it's why the Network Automation Forum has been created. I mean, there's a bunch of people trying to solve this problem. Why won't these networking people adopt automation? You know what? I mean? Every vendor Like what's wrong with these networking people? For me, it just I told myself I couldn't, it was beyond me, I wasn't smart enough. It wasn't true, and I don't know if any of the other networking people out there feel that way. But, um, okay, kev, you mentioned you guys don't do much automation, right? I?

Speaker 3: 32:32

mean I can't relate to what you're saying, like I never felt, like I wasn't smart enough, I'm not good enough, kev's got it all. I'm not conceited, I'm just. I'm just saying like it's just. I just never had the desire, you know, and I've just avoided it. And yeah, you know, it's you kind of have to, at least for me. I have to need it in my current job for me to want to do it I call bullshit.

Speaker 1: 32:54

So let's say you have 300 devices, you have to update an snmp community string on. How are you going to do that?

Speaker 3: 32:59

I get one of the admins to do it.

Speaker 2: 33:03

Delegation. I like it.

Speaker 1: 33:05

But that was something that I ran into in production and then my eyes were opened, Like, oh, I can't do this no-transcript, divvied it up.

Speaker 3: 33:30

There's like six network engineers and we all just took a block and did it manually and it wasn't. It wasn't that painful. It could it have been automated? 100 should have been automated, probably, but we survived, you know your tax dollars hard at work I mean it's going well.

Speaker 1: 33:49

Six network engineers making a hundred thousand dollars, so six hundred thousand dollars a year in salary. Spending a week updating an ip helper address working for you people.

Speaker 2: 33:59

It's a public servant. Okay, yeah, take the normal number minus 25 okay, exactly.

Speaker 3: 34:04

But, kev, I'm right there with you, I, I get it like it's gonna just I'm providing a counter yeah, yeah, no, no, and I'm I'm completely agreeing that it should have been on. It's something that someone on the team not me should have known how to do, but not my problem. But until the day comes where it's like a thousand that I have to update and it just can't do it manually, that's the day we're like, okay, I'm gonna look at python and get some kind of script going, but I need to be forced because same people in general don't don't change unless they have to like. We resist change as as general, as homeo sapiens, we just don't like change. Um, so that's really, it's really what's going to come down to. And luckily I'm in the public sector. That is like 30 years behind the industry, so I'm good, I'm going to retire, I'll be good, but you guys, you guys are in trouble.

Speaker 1: 34:51

I was, I was forced as well. So I agree with you a thousand percent. You know I had to be forced into it because I was going to resist as long as I could. I hope that you stay in the public sector and never explore in Python.

Speaker 2: 35:07

That Florida State pension ain't bad.

Speaker 3: 35:10

Yeah, frs Florida State Retirement System.

Speaker 2: 35:12

I left FRS seven years and six months, six months before my vesting.

Speaker 3: 35:17

You got to be vested man, go back in, I can hook you. I can hook you up with your job at usf if you want. I got some connections still but you know, you know, we'll talk they don't pay well, but you'll get your six months of uh of frf I.

Speaker 2: 35:34

I have the chief hr officer in my class, so automation to put a to put a bow on this.

Speaker 1: 35:42

What are you going to learn Python? So?

Speaker 2: 35:46

so this is so my Python, so my automation script. I ran for Cisco Meraki. I honestly never touched Python before. It took me about what was it? Six hours to write that, with a little bit of Googling, a little bit of stealing and, honestly, knowing HTML and JavaScript and do some Code Academy for like six months. I kind of knew the logic. So when I saw this code on Stack Overflow, I'm like all right, I can steal that. Ok, I know how to use that. Ok, api call, I know how to use that. Let me grab that real quick. And then I had it. And then a little bit of people on discord saying, well, how about add a qr code, add this? And I was like all right, within six hours I had a maraki script that rotate your your wi-fi password and email you qr code.

Speaker 2: 36:32

Yeah, I just honestly, it just takes the getting shit done mentality yeah that's all it takes is just do it, don't be afraid of it, just like your running story, forrest Gump. You know what I want to say.

Speaker 1: 36:44

You know what I want to say. If, if you can read a route map or write a route map and understand that logic, I don't find programming all that different, at least in in an easier thing. Like I remember first looking at route maps oh yeah, this is my cell for, like network people who hate coding Like if you can understand a route map, route maps are kind of little mini programs of logic that are just running to do things with routes and or tags or ACLs your policy-based routes, like all that stuff.

Speaker 2: 37:29

It's just basic logic.

Speaker 1: 37:31

Guy Chris Margetta used to work with the juniper. He's like dude, look at the, look at the Cisco CLI that you're in. You're telling me you can't program. Like look at this mess. Like this is all logic and commands you're putting in. But again in my head it was just different. Like Kevin had said, you know, until you're forced, we're heavy on time here, so let's try to move it along here. So you got a bunch of things in here Alignment with business processes, something I didn't do early enough that I should have. Right, like what does the business need? And you know, I always thought I was the network and leave me alone.

Speaker 2: 38:05

I don't care about the business, which is the dumbest thing to say, out loud.

Speaker 1: 38:07

But like the network, it's not my problem, you do. There's so much value there. Von Sharp has said it for years and I just didn't get it until, you know, a couple of years ago.

Speaker 3: 38:14

I don't. I think it's the old way of thinking where, like you're an IT person, you get in a cubicle Don't bother me, I'll be sitting here doing my thing and I'm just like dead to the world and you just can't do it anymore. Yeah, you got to be a business person.

Speaker 1: 38:28

Right, and again, I always have reasons, excuses, whatever. But like working in these gigantic Fortune 100 companies that are so siloed, there's just this mentality, like I am so separated I don't even touch the LAN because I'm a WAN guy, because the place is so big and there's tens of thousands of devices for me to then feel like I'm connected to the business and I'm I mean, we are but I just felt so disconnected from so different parts of the organization technically, let alone the business. But you'll go far if you can get aligned with the business and learn how they talk and see what's important to them and you know it's really going to be good for your career. Vxlan, it's everywhere, dear Lord. Evpn, vxlan, overlay all the things right. You got that in there. Yeah, kev, you messing with any of that?

Speaker 3: 39:13

Yeah, we're in a transition right now for our data center and we're starting to do VXLAN.

Speaker 1: 39:19

Cool, it's intense. Right, it's a lot. Yeah, it is, it's a whole lot. It's awesome what it does, but man, yeah, the idea of not having to span VLANs and all that.

Speaker 3: 39:29

It's super cool. It's just a lot.

Speaker 1: 39:33

I'm a wordsmith, I don't know. It's a lot.

Speaker 2: 39:39

Sounds like. Next time I'm in Tampa, drop off bourbon at your house.

Speaker 3: 39:41

Yes, exactly, I'm slowly, slowly draining it. I really have to work.

Speaker 1: 39:47

So we have. We're getting close to time. I guess we can go as long as we want, but in the interest of trying to stay somewhere around an hour here, there's two big slides that I think we should hit, which is well, we might've done this already. Impact of cloud on networking, but we can maybe touch on some of those points if you think they're relevant. And then the future. I want to touch the future before we're done here, and we've been going about 50 minutes, so let's say we have 10 minutes left. Real quick DevOps mindset. What does that mean? Does that mean that you're supposed to use automation tools and GitHub?

Speaker 2: 40:17

and.

Speaker 1: 40:18

CICD workflows in your networking. Is that what you're getting at?

Speaker 2: 40:21

You can. But DevOps mindset is it's not developer and ops and separate cubicles. You're embedded in those teams, not only what your customers need, your cellmates there in IT as well. Right, I might say cubicle, cellmates, but just cubicle, never come out there, but yeah.

Speaker 3: 40:39

No, I like that, I like cellmates.

Speaker 1: 40:41

So the networking and the developers working kind of together hand in hand, yeah, okay, that would be smart.

Speaker 3: 40:49

So it might be a segue then, I don't know so the old ways of doing it where you wore multiple hats, you were a network person, you had no systems, you had to know everything, and then we saw kind of a separation of silos where I'm a network person. So are you seeing that it's kind of coming back now?

Speaker 2: 41:03

Yes, a hundred percent. You're expecting to wear all the hats again.

Speaker 2: 41:06

And I think honestly, if AI it's just going to be more prevalent too. You know we say the AI word here, but it's going to be more prevalent and we'll, especially through the cloud environment. You kind of got to wear a little bit developer hat and work with them and you got to know systems and networking, because it also has to help you out and know the big picture. Yeah, that's the best thing you do to help your career out. Don't just go down one path. You need to know the whole picture, otherwise you're developing the silo right.

Speaker 1: 41:32

Be a generalist right Generalist, yeah, yeah, and that's another pendulum that seems to swing.

Speaker 1: 41:36

Like it was generalist and then it was specialist and I guess it's some great book about how being a generalist is like the best thing you can do. Um, you know right now, and I forget what the heck it's called, but it's. It's a great book about that. Um, and I try to do a little bit of that in the data center, because I was a WAN guy but we were always talking to the app and the server people. So I got a server at home and I learned ESXi and started spinning up VMs and trying to figure out the vSwitching and I wanted to learn more and more about just the ecosystem I was in, besides just networking, and there's so much to learn. So, really quick, I guess, impact of cloud on networking. I know we don't want to spend a ton of time there, but what are the highlights here that we haven't touched on already? I mean, I guess all I just said was cloud was hard and you said, ah, it's not that hard, too weak.

Speaker 2: 42:22

Well, I mean, how we got to the cloud changed dramatically over the years. Right, you know, it used to be just the internet. Now we have MPLS connections, Now you have what Megaport, Alkira and all the other kind of networking as a service providers, as well in the space. So how we get to networking in cloud has gotten very verbose and you have to pick the right ones for you. So that kind of changed it quite a bit and also it goes with your pendulum going from fully in cloud going back to data center.

Speaker 2: 42:54

That's the same long lines Like what Twitter went backwards after Elon Musk back to data center for that's the same long lines like what twitter went backwards after elon musk back to data center for a lot of things right, and it's a case-by-case scenario. You know, sometimes you go full cloud and realize why the hell did I move it up there cost me like triple why? Why did I do that right? A lot of these things are anecdotal. You don't know it's there until it's in the cloud and then you're like well, like I can run it cheaper back in office. Or you missed the control, or you built something so special you can't run it on generic Azure or AWS machines. You need something very specific.

Speaker 1: 43:22

Somebody say something about why people go to cloud and I thought it was interesting. I want to say it was Mike Bouchon, but I could be misquoting somebody. I thought it said something to the effect of like people know now that the cloud is more expensive, so what you're paying for is their operations right, you're paying for flexibility and scalability, faster speed, flexibility but you're also leveraging their operations, their operators, like their people, right, you don't have to. You're not doing hvac and and smart hands and cabling and generators and like if you're yeah like that's just you, I, I press a button, you, you do it.

Speaker 1: 43:57

You operate that network Right. Um, yeah, it was just a different framing I hadn't thought of before, um, but but it seems. You know there was that whole repatriation thing and I know cloud people get all upset, especially the cables of clouds guys. When you say that like it was one guy, I think it's more than one company that like repatriated. But you know they went to cloud, it was super expensive, they pulled back, but I guess hybrid and that's something they'll always say to right Hybrid's kind of like the way to go the future. That's where it is Right, cause I guess some apps make sense. I wanted to ask you earlier and I we can cut it out if you don't want to answer it, but when you said you guys moved a bunch of your stuff to cloud, you guys moved a bunch of your stuff to cloud.

Speaker 2: 44:35

I wanted to ask, like, did you refactor your apps, did you? Just, I want to say some of them had to get refactored. I remember we spent the app team did herculean effort refactoring like three, four sprints worth of stuff in one night because I'm under the impression if you don't refactor your apps for the cloud, it's not great it's.

Speaker 1: 44:50

You're not leveraging the technology properly and you're spending a ton of money just to run monoliths somewhere else.

Speaker 2: 44:56

We had an old megalith monolithic application and kind of had to refactor it and just didn't. There's a lot of things that weekend. That was a fun weekend, Definitely, yeah. Also, we're still refactoring a lot of our code for just taking advantage of all the new cloud stuff. I can tell you I work with some great app dev guys who are just forward thinking and on the cutting edge let's try this and they're doing some great stuff and as a network engineer, I'm doing every day hey, I need this turned on this private endpoint, this, that, this. So they're moving at a fast pace. We're moving at a fast pace, right behind them. That's why the DevOps mentality has to be there.

Speaker 1: 45:30

I think the speed that was the biggest thing I saw in production when COVID happened COVID, you just said COVID, covid I'm a little stuffed up, Sorry. When COVID happened, everybody went home and our VPNs could not. I mean, we were getting kicked off and couldn't join before all that happened because our VPN infrastructure wasn't robust enough and built out enough. And then everybody went home and you couldn't get on at all If you weren't on at like six in the morning, signed in, like you just weren't getting on until like after 12 when people started lunching.

Speaker 1: 45:59

But I remember I don't know if it took two nights, three like whatever, it was basically overnight or like over two nights. They built all that capacity out in a cloud and it was all back up and like it would have taken us I mean the circuits we would have needed and the equipment to order and rack and stack and like I don't know how long it would have taken us to build that capacity to send the whole company home and have VPN access, but it was done in a night or two in public cloud and that just that made an impression. I'm like, oh wow, that was. We couldn't have done that on prem fast. You know, that's pretty awesome.

Speaker 2: 46:36

So I got done that on prem yeah fast. You know that's pretty awesome, so I got lucky.

Speaker 1: 46:38

I work for the government, so we never left the office, all right. So let's, let's hit this last section here, the future of networking. Right, and I love the first bullet point here. Is it a dying field? Oh no. Is network engineering still a good career to pursue, to get into, to have a future in? I publicly said once that I probably wouldn't tell my to get into, to have a future in. I publicly said once that I probably wouldn't tell my kids to get into network engineering. But I think that's when I was out of work and trying to learn programming. I was very upset.

Speaker 3: 47:07

Do you take that back now? I'm curious, Do you? Do you regret saying that?

Speaker 1: 47:11

Um, I regret saying a lot of things. Um, tom Hollingsworth reframed it for me at nfd. You know he said it's, it's not, it's not dying, but it's shrinking. He's like, listen, there's still cobalt programmers, there's still mainframe operators, like there's always going to be a network. Um, if you look at all the software abstractions and tooling and cloud and nai, now right, like I mean, I've seen tools that can do the work of 10 network engineers with one at former company you know software companies I've been at and and that's that's a very lucrative space to be as a software company.

Speaker 1: 47:45

It's, it's, you know, kind of imagine if I could come in as this, as this, as a sales guy to where you work. I'm like, listen, I know you got six guys spending you know two weeks doing IP helper addresses, but I got this killer. You know AI enabled automation thing. I mean we could have done that in 30 minutes with one button click and you can get rid of five of those guys. Like it's, it's such an easy sell if it works and I've seen it work. And in wireless I'll say you can connect the dots yourself, but on what product that might be?

Speaker 1: 48:14

So I don't think it's dying, it's shrinking. I don't know if you know part of it's thankless and there's maintenance windows and on call and you're always getting beat up Like it's a really hard physical and emotional job sometimes. Like I'd much rather my kids. I don't know what I want my kids to do, but you always want your kids to have a better life than you think, right, like my dad was a cop and he wanted me to go to college to not be a cop. So like, all right, I'm a network engineer, I'm not getting shot at. That's good, um, but yeah, I don't know.

Speaker 1: 48:43

I don't have a good answer to your question. If, if my kids love technology and the networking bug bit them, go for it. But I, you know, I don't even know how many network engineers we're going to have. I mean, my kids are seven and 10. Like, in 10 years is it going to be? I think the BLS, the Bureau of Labor Statistics, whatever it's called? I mean, I think it's still growing. It's not going away necessarily, but I think the growth is way lower than it used to be. I mean, listen, we're not building the internet out anymore, right, like it's all there. So I don't know if it has anything to do with that too, like are we building a ton of networking, new networking infrastructure anymore, like I don't know. But is it a dying field?

Speaker 2: 49:24

Nick, I mean this is no, I say no, like I put on there for you but and talk to the group. But I don't think it's a dying field, I think. But, uh, I don't think it's a dying field. I. I think it's hard to attract people because cyber security is a sexy thing right now. Ai is new, sexy, so cyber and ai are the sexiest, shiniest thing. Right, and people don't understand. They go oh, my first job's gonna be cyber security and I'm like the hell, it isn't okay, let's talk about that, right, and I do that first day class. All my students say, oh, what do you want to be? I, I want to be in cyber. I said you're not Okay.

Speaker 3: 49:56

Next, Crush your dreams right away. I like it.

Speaker 1: 50:00

And why do you say they're not? Because my understanding is they're going to secure a network that they don't understand how it operates.

Speaker 2: 50:07

They don't know systems, they don't know network, they don't know app dev Right To get those cyber roles.

Speaker 1: 50:12

They need more than three hours in a networking class to have the foundational knowledge. Like a cybersecurity boot camp. Two weeks, $9,000. You make $200,000 in salary. It's nonsense.

Speaker 2: 50:28

How do you put together a network that you've never set up yourself? We all know how each other one of us probably can go behind each other and say, okay, I can see how they fucked up there or they did action here, right. How can you be a red teamer without knowing the normal vulnerabilities in business?

Speaker 1: 50:41

Cyber's not an entry-level job in general, right I? Don't think it is, maybe you can come out of college doing it. But the people I know in cyber cut their teeth in networking first.

Speaker 2: 50:50

Or they were hackers themselves. I mean, they always were tinkering or something. They're tinkerers. That's the old school 90s, 80s, computer thought. Right, and I don't think it's a dying field, I just think it's just out of the spotlight. Right with the plumbers. But cloud helps us become cloud engineers. Now we're cloud network engineers. We're you know they really just changed us the guy who fixed your internet. You know, that's the men and women who fixed your internet. That's who we are right.

Speaker 1: 51:18

I'm glad you said that, because the people that pivoted to cloud seem to be doing better in their career, across multiple metrics, than the people who stayed on prem. I don't know if that's just the few people I know who went from NetEng to cloud. I mean, they're making more. It's a sexier job. They're always going to have a job.

Speaker 1: 51:40

Well maybe not right, I don't know. Again, these are all just assumptions I make. Like, cloud is where everybody wants to be. Everybody goes. They'll spend To your point. A company will just spend whatever the hell they got. We got to go to cloud Sign the checks and it'll take a year or two to figure out. Oh hell, this is super expensive, yikes. And it'll probably take another year or two to repatriate some stuff on-prem. Do you see the cloud fervor?

Speaker 2: 52:10

pulling back at all? I don't see it pulling back. I see it different lights. It can pull back on something. It all depends, the famous words IT, it depends. It depends on how big you are, it depends on how small you are, it depends on how forward-thinking your boss is and also how non-buzzworthy or buzzworthy, whatever, your leadership team is.

Speaker 1: 52:33

But if you're a network engineer and you want to future-proof your career, cloud Should you pick cloud. Right, I mean simple.

Speaker 2: 52:39

Get your.

Speaker 1: 52:39

AWS, yeah, yeah, like just painting broad strokes here, like not that you have to. If you're worried, kevin's not worried Public sector, you're good, you'll be there forever. And I mean that with all love and respect. Like that's awesome I respect, like that's awesome, I love that kind of security in the job that's my research field.

Speaker 2: 53:01

You know they don't give enough public sector it guys enough credit. That's why that's my whole research papers okay, yeah, yeah. So yeah, I don't think it is. And quantum computing come up up the corner here, that that's a scary thing and we could talk probably about 45 minutes on quantum, on how scary encryption breaking that could be, but but also wasn't to our networks, right? You know who knows what that's going to look like in the future?

Speaker 1: 53:20

Do we want to do? We want to say the third dirty word? We said cloud, we said automation, ai, baby. So you know, as I'm trying to future-proof my skillset, I'm learning some programming, some basic. You, you know python and github and stuff like that.

Speaker 1: 53:38

Um, I'm reading an llm book that phil gervasi turned me on to. Um, because I I don't really understand the underlying. Like, if you're using an ai tool, like I'm the kind of person that like, well, how, like, what does that mean? What does that mean? What is it doing? Like, how is it? So? I'm because it might be the engineer in me, so I, I'm trying to learn the machine learning kind of stuff underneath, because I think by the time you see the AI tool, it's just too much Like, can you learn AI? I don't know. I think you have to learn ML. It's actually doing machine learning and they, I think they call it it ai and I don't know enough about it. But how would you study ai? I mean, you're dr nick here, like you know, is is it? Is it something we study yet or is it just?

Speaker 2: 54:22

oh, there is a lot of papers on ai. In fact, half my class are doing ai, half the classes before that. Most professors are ai people. It's, it's out there, right, because it it's not only the new sexy, it's the new calculator, right? So before you're in school taking calculus, you, you know, I put this number upside down, the code checks out, right. I think you know it's a tool, it's a great tool if you're willing to use it, but also not let yourself abuse it it, but also not let yourself abuse it.

Speaker 1: 55:17

I've seen it do some pretty cool stuff. I was I mean I was anti-ai for a while because I'm like it's gonna take a hard job. This is bad, and I know kev, he's rolling his eyes, but then I started using it.

Speaker 1: 55:25

I started using it. I'm like, oh, wow, like this is like. If you use it, like you're saying nick, if you use it the right way, like if you want to brainstorm as an example, like oh my god, it's amazing. If you want to like generate a list, or if you want to like rewrite something or like even for, even like for job interviews as an example, you can feed it a job description, feed it your resume and say, hey, help me prepare for this interview. And like the questions it spits out they may or may not ask me, but they're things I wasn't thinking of, they're different angles. It gets me kind of in that conversation with the interviewer before I go in, like just one use case of like wow, this is amazing. Not to mention the networking use cases.

Speaker 1: 56:00

I know I've said it before publicly, but I was messing around in SR Linux and they have a I forget what the branded thing is, but it's a chat GPT agent in there and you can talk in the CLI in natural language and it'll tell you anything you need to know, and it gave me all the commands I needed to do what I was trying to do. It was much more elegant than the question mark and other CLIs. I've worked in right Like trying to find the code and the syntax and the higher you know the folder I need to be in the what the hell I'm doing? So I love the. I love what I've seen so far. Are we going to become too reliant on it? I don't know. If we have AI tools that are going to do the work of 10 and one person can do it and now they're just in GUIs, like you were saying, kev, what's going to happen when all hell breaks loose and the building's burning down and the AI's on fire?

Speaker 3: 56:46

I'm personally really looking forward to AI taking over everything. I'm in the public sector again, so in in my world, right, I want the ai to come in number one. It's it's in the public sector. It's we don't fire anyone, right. There's no layoffs, um, unless you like, break a law or do something really legal, then well then, we'll fire you. But typically you know it's hard to get fired from the public sector, um, but and it's also hard to get add people to your team, right, add new lines. It's hard to to get new reoccurring funds for salaries but generally work lean.

Speaker 3: 57:25

So like, yeah, exactly, and you're lean and you're never getting out, but it's really easy, for some reason, to buy things. We have a certain amount of funds we have to spend before the end of the year. We have to buy stuff, or else we lose it Right, and if we lose it, we don't get it back next year. So we have to use these funds. So buying AI, paying tools, buying software, whatever to help the few people that we do have their you know the job to do it better Sounds like amazing to me. I'm not worried about losing my job too. I'm not worried about layoffs or anything like that. I just want my job to be easier.

Speaker 2: 57:56

Yeah, so and public sector. I can see that I mean again. And also you can't get fired from public sector.

Speaker 3: 58:02

I was I feel like that's a story there. I want to hear it, but I don't know if we have time for it right now.

Speaker 2: 58:08

No, that's probably about 30 minutes.

Speaker 3: 58:15

And you know what, meet me at a sports bar down there. Yeah, I mean, I'm actually going to USF tomorrow for lunch for my old co-workers, so maybe I'll see you there.

Speaker 1: 58:20

I'll tell you how good the systems are going and they're leveraging AI in the testing. I have a buddy who's an airline pilot and in the military and he's telling me that our kids will be flying in pilotless commercial jets.

Speaker 3: 58:34

That scares the crap out of me commercial jets.

Speaker 1: 58:35

that's good, so like 20, but it's how good we like. And again, a car isn't a plane, but we have driverless cars. I mean they occasionally run people over, but they're a thing, they're working right. There's people flying down the highway at 80 miles an hour without their hands on a road, with like an auto driving car in some senses.

Speaker 1: 58:52

And when, when, when you talk to a, an airline pilot, and they tell you how automated everything is already, like there's not much to do if you don't want to do things like to your point, kev, like their job has been very simplified and it safety has come as a result of that, because all these systems and checklists and so the airline industry there's, there's, there's like a lot to learn there. But to the point we said earlier, when the shit hits the fan and things get bad and the systems fail and bad things are happening in the plane, I really like having a human up there or that network expert who knows like, oh yeah, you know everything's down, but I know what to do, as opposed to like you know the tools are failing, what do I do? The plane's going down. As opposed to like you know the tools are failing, what do I do? The plane's going down Like I don't know if I'll ever get in a plane without a human up there.

Speaker 1: 59:44

Because that's a lot of trust to put in software in my estimation, but it's just an example of how far, like where this stuff could go. Like can we automate networks with AI and make our jobs easier? Absolutely, Maybe, Because there's roadmaps at Boeing to make pilotless commercial airliners.

Speaker 3: 59:59

Yeah, but I feel like a plane is easier to understand than some users. That's really what it is.

Speaker 2: 1:00:05

And applying AI, especially to a lean team like Kevin's talking about right. So if it can take away 20% of the minutia that he has to deal with, his team can actually innovate instead of struggling to tread water, and that's really where I see AI helping out. It's okay, take care of some of the boring stuff, or some of the stuff that takes away my time, and let me innovate and iterate.

Speaker 3: 1:00:28

Yeah, we have projects that we can't keep up with because we're doing housekeeping, because we have technical debt that we're trying to keep up with. Oh, technical debt that we're trying to keep up with when we have buildings that are literally being constructed, that we have to have new networks for, and we're just drowning in work. So it's the BS that we don't want to do. But to your point, andy, I feel like man. No matter how far technology advances, users are still going to be users. I can't say that politely, I'll just say they're going to be users. Like I get tickets for firewall rules, or like I want you know this application to go here. Well, what ports do you want? What sort of? What's the destination? They don't know any of that. Just make it work, make it magic, and ai, no matter how good ai is, they're not gonna be able to figure that out without a lot of interaction, a lot of you know, looking at things and digging into it.

Speaker 2: 1:01:13

So yeah, yeah, yeah, I mean a lot of looking at things and digging into it. Yeah, yeah, yeah, I mean a lot of good things. Ai can do with accessibility too, and stuff and helping everything out there. I mean it's going to be a great tool.

Speaker 1: 1:01:23

I want to circle back. So we're going to wrap up here. But I want to circle back to Kev's question earlier because I've had more time to think about it. But what I tell my kids to go in network engineering right, I mean, that's what this whole episode is about is, like you know, is this a good career field. What do you need to know? What do you need to do?

Speaker 1: 1:01:37

Like as much as I might take shots at it, like networking has been so good to me I was climbing ladders, getting physically and emotionally beaten up every day as a cable guy for 400% less money, and you know, and network engineering has given me such a great life, not just financially, but I really enjoy the work. I'm still fascinated by it. Like I'm talking to, you know, the ultra ethernet council guys soon and they're like creating a new transport stack and like I'm still like in love with this stuff. Like wait, wait. They're like redefining ethernet, like that is insane. And when you look at what they're doing, like oh my God, I'm amazed how smart people are. They can like right. Like I'm just trying to figure out what they created. They just sat around like we're going to make stuff. So I mean, I'm still in love with networking. It's a great career field.

Speaker 1: 1:02:33

Um depend, you know, like, like all of us, like when you come in and like kev, you said like easier, right, like you want it to be easier. It was easier in 2012 when I got my ccna because, ony, it was route switch and that was enough and I didn't need to know python and I didn't need to know cloud and I didn't need to know ai and I didn't need to work with developers and I didn't need a cicd pipeline and github and I didn't need to CICD pipeline and GitHub and I didn't need to have a GitHub repo in my LinkedIn just to look like a real, like it's just so crazy that every five to 10 years there's three more technology stacks we have to learn for the same salary that we had in 2012 when it was just around Switch, like you know. So that's kind of that old grumpy like urgh, leave me alone, I don't want to learn any new stuff. Like me alone, I don't want to learn any new stuff.

Speaker 1: 1:03:17

Like this is hard, but it's a great, you know it's, it's a great career. I can't really. I mean, what, what do you got? Like would you? What would you tell your kids like what do you think about the career?

Speaker 3: 1:03:23

me or nick both whoever? Kevin, you go first, you have kids um, yeah, so I have a 14 year old and a 12 year old and I would my number one, my daughter.

Speaker 3: 1:03:33

She's not technology, she's as much as I've tried to get her into tech. I want a female engineer in the family. She just has no interest in it. So I'm starting to give up. But my son is into video games and technology and I would not tell him to go into networking. I would tell him to go to programming, do some kind of programming, just because I feel like that field is much more open. I mean just because I feel like that field is much more open, like the opportunities are much more available there. With less knowledge, with less background that you need, I feel like the entry-level jobs there are just easier to get. But I'm curious, nick, I know you said you don't have kids and stuff. But say, if you were a 17, 18-year-old kid graduating high school, what job would you get? Or and what steps would you take to get there in the next, like three or four years after that?

Speaker 2: 1:04:22

Well, I tell 18 year olds this every semester, so I actually have 25 percent enrollment rate with women. So that's my highest enrollment rate this year.

Speaker 3: 1:04:29

So I need to get my daughter to talk to you.

Speaker 2: 1:04:31

So you're saying yeah, I mean I don't know why, but like it was great to see you know, all three of my classes, 25 percent out of 20, 24 students were female. My guys Great is, 25% out of 24 students we're female. I'm like guys great. Thanks for showing up. Let's have fun. Now let's talk about technology. Let's make you all nerds and by the end of it they might know some networking. Most of them know movies and TV shows they should watch for being nerds.

Speaker 2: 1:04:54

But honestly, programming is never a wrong thing. But what I tell people is no little programming, no little programming, no little networking, no little systems. You got to find out what makes your brain itch. You got to find the itch and scratch it. It could be programming, it could be networking, some parts of programming. You don't get a good feedback loop where you build it and you see it. Some people need to see it right or touch it. Tactile Networking has a tactile component to it. Unless you, unless you're a cloud network, right, then you never get to see the switch go blinky, bleaky, right. You know, I like be networking because I can see the switch. I got cables, I I knew how to work on it like you're working on an engine right and I talked my switches like their children.

Speaker 3: 1:05:33

all right, honey what are you doing today? You know, like, what's going on here, right, that's weird yeah, you know.

Speaker 2: 1:05:38

Yeah, talk to him sometimes. You know, the firewall went out the other day and I was like, come on, honey, what's wrong? What's going on over here? So honestly, I would tell them they got to figure out what makes them, what makes them happy, like I'd say, networking only because it's fun. It's always a Rubik's cube to solve. But so is cyber right and it can lead you into that path. Right App can lead you into cyber. A lot of things lead you into cyber, which I have fun in, cyber a lot. But.

Speaker 2: 1:06:02

But networking, my default mode, my factory reset button for me, is management and networking right. I've been managing for many, many years, more, more years in retail actually, no, I think pretty even between retail and um it. But networking has always been a fun thing for me because it's I make. I can make something work. I get a feedback instantly hey, this work, ping done, no, ping, Okay. Whatever Blame, dns, pcap. So that's, that's my take on it. You got to find what motivates you, and it may not be networking or may just be wifi networking. Are they teaching networking in your college? A hundred percent. We have a lot, of a lot of switches. We actually have a lab. You can take switches home. Right, you actually can sign out switches and routers take them home um.

Speaker 1: 1:06:49

We do all hands on moving networking from like computer science.

Speaker 2: 1:06:52

No, we are, I don't know. We we doubled down. We actually added python scripting for our networking courses. Cool, right, um, we have a hell of a cyber security lab, but you have to take networking first. You can't go the cyber courses without networking first is this um, is this a computer science track?

Speaker 1: 1:07:10

is this through neticad like, what do you have?

Speaker 2: 1:07:12

yeah, so we have we have neticad built in. So four courses are neticad. You know, cisco, one, two, three, four go through neticad, right, but there's like four different degrees that kind of come through networking. Even our biomed students take some networking. Wow, right, because their biomed equipment may be on the network. So we still have it. We just launched an ai degree and I think we just launched a bachelor's in cyber.

Speaker 2: 1:07:37

But we've been always kind of retooling some of the higher end Cisco, the networking degrees. I think at one point we had Fortinet. We do VMware as well for certain things. We used to have NSX lab, so it's all hands-on. We're a state college, we're not university, so we're not theory. And I always crap on universities like, hey, you're teaching theory to networking, throw a switch in their face. So up here I pick on UNF because we have University of North Florida. So I always make the joke not even a joke interview question You're next to somebody else and you can't talk to a server. What's the first thing you do? The UNF students are like well, I'm going to check the ACLs and the routing and the FSCJ the state college first. I'm like, well, I'm trying to ping the server first, right, check the cable, right. So the theory nothing replaces hands-on.

Speaker 1: 1:08:24

Nothing replaces it. You see a problem in the future with these things coming together. Older network engineers retiring the people with that old school institutional knowledge of how the protocols are working. Newer engineers not knowing, not having the experience with the non-abstractions right With the let's say they're the ClickOps people, right, they're dealing with all the software and they're less knowledgeable on what the older people knew.

Speaker 1: 1:08:50

And then you know, the needs of the network keep increasing. There's more and more devices, there's more and more people. So it's kind of this weird confluence of, like increasing demand on the network, less skilled, older engineers that have that really deep knowledge and the people that have come in to replace them again, this is a generalization and somebody's going to yell at me, but like they don't have that deep knowledge, They've come up in a time that it's all software to find this and ClickOps, that Like. Are we headed for disaster Because there's not going to be deep experts when bad things happen?

Speaker 2: 1:09:26

I'd love to say no, but it's always a risk. I think it's a risk in any industry, right? When it happens, it's all one sentence You're going to learn today, son.

Speaker 1: 1:09:34

It's what's going to?

Speaker 2: 1:09:35

happen. You got to learn today. You're going to grow up real fast. What do you think, Kevin?

Speaker 3: 1:09:41

I don't know if it's my bubble of people that I'm surrounded by, but this job attracts people who like to figure stuff out, and I think it's that question isn't giving enough credit for those people who are just learning the GUI. When it does eventually break, those same people, hopefully we'll then look at, you know, the white pages. We'll start looking like, digging into it and figuring out hey, this button that I usually press does this, and now it's not what the heck happened. And I think that mentality is the core of our job, that's the core of who we are as tech people, and that trait will won't go away as time goes on.

Speaker 2: 1:10:16

Reverse engineering, something to figure something out. Now, as long as you have again people that want to GSD, get stuff done right or networking is the perfect spot for them right. That's why full sun, fridays, you know, ask for permission. Don't ask for permission, ask for forgiveness.

Speaker 1: 1:10:36

Kev, I think you nailed it. We all have that curiosity, that deep-seated curiosity, that need to tinker. We want to know how things work. So yeah, even if the tooling that we've used is easier in some sense, you're right, the people that are attracted to networking want to know how things are working, and if it breaks, you're going to figure it out.

Speaker 3: 1:10:52

So yeah, that's a really good I mean, I'm kind of in my mid career. You know I'm 40. So I'm far away from retirement and far away from my beginnings. I'm kind of in the middle. And so I see people who are 20 years older than me who are leaving the industry, and I also see the kids coming in who are network admins. And the kids coming in are hungry. They have that hunger. They want to figure stuff out, they want to get their hands dirty, they want to get under the hood and look at the stuff. And I'm getting to the point where I'm getting too old for this crap. You know mentality where, like, I want to spend time with my kids, I want to do this kind of stuff. It's hard for me to to spend hours doing it at night, but they're still in it, they're still doing that and that's where I was 20 years ago. But that's the mentality that those people will be.

Speaker 1: 1:11:31

You know they'll be under the hood 20 years from now do you feel that those kids are smarter, like they're coming in computer science or they're programming, or like do they know more than we knew 20 years ago? That would be my guess that they're, that they're smarter in a sense, but your face is saying no, smarter is a hard word you know.

Speaker 3: 1:11:47

It's like yeah, that's the part I do have more knowledge, like a variety of knowledge Probably. I mean Nick would be the expert there. I would say probably, but smarter is a hard word to.

Speaker 1: 1:11:59

Yeah, do they have more skill like technical skills?

Speaker 2: 1:12:02

They know technology. But, honestly, you can't replace hunger and that's a learned trait from behavior and that's what I look for for interviewing. Are you hungry? Yep.

Speaker 1: 1:12:14

That's the biggest thing.

Speaker 2: 1:12:15

You got to be hungry. And when I find somebody who's hungry I don't care they don't have CCNA or nothing I'm like you know what? You're my guy, that's it. You got it out that you're hungry. Let's do this. I can mold you like clay if you're hungry you're gonna figure it out 100.

Speaker 1: 1:12:29

My wife has a saying I love. She says you can't teach hustle. She, no, you can't. She played basketball at a high level and she's coached people and like you're, you're not gonna. You're not gonna get somebody to like chase that ball down and jump on the floor, like it's either in you or it's not and it's hunger. It's the same thing if you're hungry and you're gonna go after it. Like kev said, if you got people that are hungry and want to learn, I mean that's the best. I think they're the best people to have around. They're gonna bust their ass and learn things. They. They want it right, they want it. Um, nick and kevin, thanks for being here. It's a really good conversation.

Speaker 1: 1:13:00

If you're in networking or you want to get into networking, just internalize that you're always going to have to be learning something new. You're never going to know it all and every five to 10 years there's going to be a new sexy thing that you're going to have to learn. These are things that I realized lately and I'm like, okay, there is no finish line. But if you're hungry, if you're curious, if you love technology that's supposed to be one of the great things about being here. So just don't burn out and don't expect it to get easy.

Speaker 1: 1:13:27

You can check out our link tree. It has links to pretty much all of our stuff and the reason I bring this up is we have a server called it's all about the journey discord server. Why I love it so much is there's about 3000 people in there and they're all helping each other. So there's a CCNP study group happening now, there's a CCNA study group happening now and there are just a bunch of others, an automation group in there. They they were working on some problems today. So if you need a community, if you want a place, it's free, come in, say hi, hang out. You can find that link at a link tree. Forward slash art of NetEng, as well as a bunch of other stuff in there the Cables of Cloud podcast We've got some merch that we haven't updated in forever, but all the links to our socials. But check out the Linktree, check out Mr Kevin Nance on TikTok at Adjacent Node. He's always creating amazing content. Nick, where can folks find you if they want to get in touch, learn more, see what you're doing?

Speaker 2: 1:14:20

Mostly my LinkedIn. Honestly, I do have Twitter, really more Twitter. I kind of follow people on. I don't really post much on there, I need to get better at it, but I kind of curated right, but 100%. My LinkedIn website too as well, which I just got the domain. Today I finally registered domain, so we'll see.

Speaker 1: 1:14:39

Awesome. Well, thanks so much for joining it was so, we'll see. Awesome. Well, thanks so much for joining us. Great conversation as always, and we'll see you next time on the Art of Network Engineering podcast. Hey everyone, this is Andy. If you like what you heard today, then please subscribe to our podcast and your favorite podcatcher. Click that bell icon to get notified of all of our future episodes. Also follow us on Twitter and Instagram. We are at Art of Net Eng. That's Art of N-E-T-E-N-G. You can also find us on the web at artofnetworkengineeringcom, where we post all of our show notes, blog articles and general networking nerdery. You can also see our pretty faces on our YouTube channel named the Art of Network Engineering. Thanks for listening.