Viktor
00:00:00.000
let's put it this way. If a company, especially public company, is going to lay off people for whatever reasons there are, there can be many reasons, right? They're not going to say simply, "We're laying off people," because that's a bad signal to shareholders, right? "Oh, you're not doing great." So you need to have a reason why, how that is a good thing, And it cannot be perceived as if we are, "Nah, we're not doing so great." And AI is a scapegoat right now. There, there is always a scapegoat depending on in which period of time that is happening. Right now that's AI. I'm not saying that some companies are not literally laying off people because of AI. I'm just saying that it's a scapegoat for those who need it This is DevOps Paradox, episode number 373, "Are layoffs really because of AI?"
Darin
00:01:46.947
If a company lays off 4,000 people in one day, and they also announce at the same time record-breaking revenue and potentially profits, and they attribute it to AI, is that just a, a really smart press release, or… Is that really automation, or is it just a press release listing body count? This is what happened to Cisco in May of 2026. They announced 4,000 people being laid off, record revenue of 16 billion, I believe. Look at my numbers. S- 15.8 billion to be exact, 12% year-over-year net income of 3.4 billion, up 35%. The CEO framed it as a reallocation towards silicon, optics, security, and AI. But here was the real key part to this. The CFO told analysts it was not savings driven, which is one of the stories we typically hear when there's a big layoffs, you know, we're not trying to save money. But rather it was a rapid reallocation This story hits a little too close to home for me because that story sounds very familiar to what happened to me in September of 2025. I was rapidly reallocated, I
Darin
00:03:09.963
So I can't put mine against it being AI. I know what's happening in the company. I know where it was going. It made sense. As a business person, it made sense to me. So zero hard feelings. But a lot of times it seems like they're just playing the AI card
Viktor
00:03:25.783
let's put it this way. If a company, especially public company, is going to lay off people for whatever reasons there are, there can be many reasons, right? They're not going to say simply, "We're laying off people," because that's a bad signal to shareholders, right? "Oh, you're not doing great." So you need to have a reason why, how that is a good thing, And it cannot be perceived as if we are, "Nah, we're not doing so great." And AI is a scapegoat right now. There, there is always a scapegoat depending on in which period of time that is happening. Right now that's AI. I'm not saying that some companies are not literally laying off people because of AI. I'm just saying that it's a scapegoat for those who need it
Darin
00:04:10.704
in Cisco, on that day when they laid off everybody and made the announcements, in after-hours markets, they were up 16%. So shareholders really loved that they got rid of 4,000 people that day
Viktor
00:04:22.979
I'm not in the mind of the CEO, so I cannot give you the reasons why they did it for real. What I do believe is that in large companies like that one, and this is going to be very unpopular what I'm going to say, you should be laying off people every once in a while. Because there is a certain level of complacency, And not everybody's equally good for the company. Somebody was good but is not good anymore. Somebody was never good, so on and so forth, right? You should be laying off people every once in a while. And but in this context, when I say laying off people, I mean replacing people, not laying off as if we don't need people anymore
Darin
00:05:01.325
because it's your job to have a successful business as the business owner or the business CEO
Viktor
00:05:08.767
Yeah. And the business changes, I could be the perfect person for the company where I'm right now, and then I might be a very wrong person for the company next year, right? Because the priorities shifted. What we're doing is now different. My skill set does not apply anymore. Or I was never good. That, that's a more common thing with me
Darin
00:05:31.881
there was a book that came out about, believe it or not, 20 years ago that was titled Fire Someone Today go back and read it. I would recommend reading it if you're, in case you're actually leading companies. It was a, a nice way to let people go. But it gave you the reasonings of why you would let people go. And I would say even 20 years on, the basics of that book, in fact, you can order it today, but it's, you can look at it at firesomeonetoday.com. It's by Bob Pritchett. A phenomenal book, short read. But it was about getting rid of the lower 5%, because why not? I mean, if we think about what's been going on, we're in 2026 now. We had a big hiring spree in '21, '22, '23, to an extent '24
Darin
00:06:31.226
should be doing that. In fact, I saw something today as we're recording. A couple of days ago, Microsoft laid off 4,800 people, mainly affecting the gaming division, so Xbox, all of that
Darin
00:06:48.828
but that's the next thing, that goes back to fire someone today. This is no longer part of our, what was our core business. And to sort of follow onto that, I also saw that they're splitting up the gaming studios to where they're not gonna be reporting into Microsoft anymore. Whether it's selling off, I don't know, but it's, it's this kind of thing. So getting rid of non-core holdings is what we need to do as well. So if you're caught up in that, it's hard not to take it personal until you've been through about two or three of them, and then you're just like, "Okay, this is the next one."
Viktor
00:07:19.812
Yeah, and especially, you know, there are cases like Xbox, that's long time coming. That's not out of the blue you know, we all hope for the best, but then hopefully people that worked in that division of Microsoft were ready for it
Darin
00:07:34.104
We did hear about Cloudflare laying off people earlier in 2026, uh, with the phrasing they weren't AI enough. It was 20% of the workforce, right? No more middle managers, which is, okay, do we really need middle managers?
Darin
00:07:51.193
You even talked about it a couple episodes ago. If at some point you've got your AI workforce totally dialed in, and you've gone from creating five features a day to creating 100 features a day, but the downside to that is you cannot validate to your level that you want validated more than five a day, so you have 95 So if you actually want to get this 100 a day, you've got to hire 19 more people, and you as a single person probably will not be able to handle managing 19 humans
Viktor
00:08:27.522
Correct. It's just that, what I'm going to say is almost certainly based on my ignorance, so ignore it, but I never understood "Oh, I'm a manager, I can't… Five is maximum. Oh, I cannot do more than five. It's so exhausting." I don't get it, that part
Darin
00:08:43.474
See, that's when it goes off the rails. That's an excuse potentially, not that capacity isn't able to get beyond that
Viktor
00:08:51.531
That also tells me that that person does not know how not to micromanage. Because the only way I can imagine that five people would fill in my every single working day is if I would be breathing behind their necks And assuming that my only job is to manage people that's not a split function
Darin
00:09:10.191
okay, maybe you can't do more than five. Let's say that, but still it's going to… You're going to need some people once you get beyond a certain number of people that just do management, people management. Because if the bottom of the rung of humans is managing multiple AI agents or AI workflows or whatever you want to call it, at some point somebody's got to help write the paychecks and help manage all the paperwork to be a company and all those things
Viktor
00:09:40.791
Oh, definitely. It's just that I feel that companies do not value enough people with autonomy or don't give them enough autonomy, and then you end up with 57 different management layers. Imagine that you're managing people, but people you trust, kind of, you're good at this. We talk about it, we decide what we are going to do, I will not call you back for a week unless you can call me back and say, "Oh, I'm blocked. Can you resolve this issue for me? Can you give me more tokens?" Whatever the request is. But other than that, yeah, go. Go, my son. Go do it. That's not valued enough
Viktor
00:10:23.110
Valued, yeah. Kind of That or not common, right? It's kind of, oh, nobody trusts anybody, right? Oh I, I need to have a person who will validate this person, but who will validate the person validating the person? Well, Then we need another person to validate that person who will validate the person who validates the person. It's just insane. Nobody's doing anything anymore. We all manage somebody
Viktor
00:10:48.496
would I manage myself, Darin, if there are five people managing me? Why would I do that?
Darin
00:10:54.464
Yeah, that inversion of managers to ICs is always a, a funny one to track down. But traditional layoffs, as they've happened, and most of the time happened to me, were due to bleeding cash. But now all these AIs r- layoff, or the layoffs that we're getting now, if they're attributed to AIs, are typically during a very high uprising of cash, of revenue
Viktor
00:11:21.174
If you're talking about specific companies like hyperscalers and AI companies, they are bleeding cash left and right. I don't know even how they can pay people that they have
Darin
00:11:31.838
Well, I think it's because they already have that much cash. When you have trillions in the bank or, multiple billions in the bank
Viktor
00:11:39.835
even that, how many times did uh, Anthropic raise so far this year? Already twice?
Viktor
00:11:46.055
This year. It wasn't it kind of like, oh, you raise every five years or three? Now we have half a year passed, we- they already raised money twice, not even mentioning how much money because it's obscene they don't have money. They wouldn't be raising money if they have money
Darin
00:12:03.213
can. In fact, Google is issuing some bonds to help build out more capacity. That's… So sometimes it's you issue debt bonds to do that kind of work, or, sometimes you just need cash flow
Viktor
00:12:18.471
debt is very different because for companies like Google, it's actually cheaper to raise debt than to use their own money, right? That works for them. But that's very different than raising money, like VC money. That means that every time you do it, you give away part of the company basically it's in Anthropic, and I'm using it as an example, it's in Anthropic's interest to give away the smallest possible amount of the company to VC, not the other way around
Darin
00:12:44.719
That's very true I'm thinking about this is just probably the next wave. Think about when cloud, Kubernetes, money was flowing that way And people were losing their mainframe jobs and everything else 'cause they were going to cloud. Is this just the next cycle of that standard wave that we're hitting?
Viktor
00:13:04.779
Yeah. It's just that the waves are accelerating. That's the major difference. They are the same. I even in this pod- how long do we run this podcast? Probably five years now?
Viktor
00:13:17.698
did, did a single year pass that there is not a big wave either of layoffs or hirings? It's one of the two every year. Exactly. It's, it's… We always get surprised, "Oh, they're laying off people." That never happened before. That happens every year It's either hiring or laying off because if everybody's laying off now, then next year they will be hiring. I guarantee that
Darin
00:13:41.047
Yeah. But the problem is, is as engineers in some companies, I'm gonna say all companies, but let's assume this is true for a moment. This is the funny one, the phrase that I hear now a forward deployed AI engineer. What the heck does that even mean?
Darin
00:14:06.087
I can't figure out what forward deployed means other than you're on the front line, and that's a professional services person. That doesn't get much more front line than that
Viktor
00:14:19.584
Okay, most other people you don't continue doing the same thing but change the role from network guy to DevOps guy or a girl without raise, Otherwise, why do you do it?
Darin
00:14:34.053
what we would do when you'd fill out your CVs and you would start listing out all of your tech tools, and it would be Visio and Excel, and you'd go like, "Really? What? I- Why would I hire you?"
Viktor
00:14:48.278
long time ago, that, that was when I was young and still stupid I was going through some hiring agency, something like that, and they told me, "Your CV is not big enough. You need to list more of those tools." I listed every single tool in Linux in spite. There was curl, there was cat, there was bat as well, there was LS. Man, I had, I had a massive CV, yeah, and I knew them all. There is VI and there is Vim as well Those are not the same thing.
Darin
00:15:18.986
did you list them beside each other though, or did you have them… Oh, okay, it was alphabetical order. That sounds great. So your CV now was padded with about five or six pages of commands
Viktor
00:15:34.913
tools, exactly. You don't say Microsoft Office, you say Word, Excel, and what so not
Darin
00:15:40.278
But that's the danger part of now the new flavor of that is they'll say Claude Codex open code, blah, blah, blah, blah, blah, blah, blah.
Darin
00:15:58.516
Oh boy, we're going off the rails here. But that's the other concern in this, is people say "Well, okay, if I'm gonna get a new job…" Thinking about me, right? So I'll go back to me. I was laid off September 2025. At the time I was 59. I did the math. I was able to stay retired, so which is where I'm at today still. I do some consulting, blah, blah, blah, but not working full-time. I feel for the people that are getting laid off that are 50, They still got kids maybe at home in high school or maybe still in college, and they still need real cash flow
Viktor
00:16:32.907
That's a tricky-- That's probably the worst age, 50, you're not retired and you're not youngster anymore
Darin
00:16:39.082
And unless you did really good investing in your 20s, which fortunately in this market over the past 30 years would be great, right? If you'd have been very consistent and if you did nothing but put it in the S&P, this is not investing advice, which I didn't do that because I didn't learn about all this stuff until I was in my 50s, like a doofus
Viktor
00:16:59.799
I learned about that, those things. I just always thought that it's so much nicer to spend the money
Darin
00:17:04.969
That's another episode that we can talk about for a different day Well, you have expenses and you blah, blah, blah, but then you try to justify things that aren't expenses. Okay. Anyway, what we've got to be concerned about is as engineers, we're being taught, okay, we've got to do the AI stuff now. Great. We've talked about this before. If I'm working as a primarily a developer, and now I'm speeding up, but my QA isn't speeding up, and even after that, operations isn't speeding up, then I'm just sitting around watching Netflix the rest of the time. Or maybe I'm starting my new tomato growing hobby, whichever the case may be. We seem to be stuck on tomato growing. But that's where we're headed, right? Isn't that what's happening out there?
Viktor
00:17:44.378
I still feel it's the same like always. It's a cycle, right? Here's my bet. We do predictions at the end of the year. I'm gonna, I'm going to make a next, a prediction for the next year right now. I'm not going to wait for the end of the year. There will be hiring spree next year It always comes after firings
Darin
00:18:00.647
Oh, of course. Yeah. Well, Actually, what happens usually after all the bonuses go out, everybody gets laid off right before the bonuses get paid, and then they start hiring again
Viktor
00:18:09.667
Yeah. there was a company I did consulting with, I'm not going to name them, and they had a lot of consultants, and they all get fired in December. And I was kind of, "What? Guys," kind of like, "Isn't this horrible?" "No, No, this happens every year. We're back in January. They just need to, close the books for this year, show the savings. And then next year, ah, next year is starting. We'll see what we'll do by the end of it."
Darin
00:18:36.679
some people are gonna say, "Yeah, but this job is important to me." That may be true. In fact, I can guarantee you it's true to you. Bottom line is, unless the business is your family, you're not family to that business And I would could even say, even if you're family to the business family, you're still not family
Viktor
00:18:57.819
Earlier today I saw, I must say now that I haven't double-checked, I just heard in one of the podcasts, so take it with a grain of salt, that actually right now they can observe higher percentage of new hires among let's call it AI native companies that are really doing good with AI than more traditional companies And if that's the case, that's the opposite of the story, this is because of AI.
Viktor
00:19:29.224
Yeah. yeah. Not only AI companies like Anthropic, but companies that are really doubling down on AI for real. Whether they're creating models or just using it, doesn't matter, right? And that might lead me to say that because of is the reason why some companies are really firing people, in big part because they don't know what to do with it
Darin
00:19:51.552
they're still trying to figure it out, whereas the people that know what they're doing with it, they see the value in the humans helping drive the AI
Viktor
00:19:59.284
I think that you said that as well quite a few times, I definitely did, that AI augments you, and then when you see uh, a large-- even more slop being generated now than with AI, that's because you were generating slop before. Now you do more of it. Well done. Yeah, faster. Yeah. we are now producing unnecessary things faster than ever before. Well done I can blindly appro- approve more PRs now with AI than before. When I say blindly, I mean, you know, my job is just to say approve without even understanding what the PR is. I can do so many more now than before
Darin
00:20:37.947
I will flip it around to where I can handle the PRs now. In fact, I had a PR today that came in and failed. I was like, "What's going on?" 'Cause I had Renovate open it up, and it was great. So okay, great. Why did this not pass? And I looked, it's oh, my Renovate rules don't allow it 'cause it was a major version bump, and my Renovate rule for that tool did not allow that bump. Great. That's what I wanted, right? That's a, That's a good… It's like, oh, do I really want this one or not? And of course the answer is yes, I do want that one. So I had to go through and clean it up. But I'm not a Renovate guru. But you know Who is?
Darin
00:21:20.518
But I know enough to understand, "Oh, okay, that's what's happening. make the change. Show me what the change was. Okay, great. that makes sense to me. Apply it. Let's go."
Viktor
00:21:30.995
I feel that that's the difference in between ex-having and not having experience is that now when Claude explains you something, you can say "Oh, oh, oh, yeah, that makes perfect sense." Or you can say, "I have no idea what you're talking about. Continue."
Darin
00:21:44.475
And both are valid. But then there's the third option of it gives you something and you go like, "No, no, no, no, no, no, no. You need to… 'Cause here's what I think and believe. You need to tell me why I'm not correct." And it'll go, "Oh well, actually you are correct. I was wrong. I was hallucinating."
Viktor
00:22:07.849
probably there is always a case of getting too much of something, but I don't think that we reach the peak. in a specific company, yeah, why not? But if you're talking about AI usage, I think that that's going to grow exponentially. We are, We are still at the very beginning.
Darin
00:22:21.566
Correct. So the follow-on to that is how do we truly measure it? Are we just gonna, pop in a OpenTelemetry collector and start piping everything to Grafana and be done with it? that seems fine, but so what? Yeah, great, we got another trillion data points. Now what? I don't know what we're gonna do with that
Viktor
00:22:43.791
I think that we will easily see like sometime in the future we will see that cost per employee of tokens is close to their salary
Viktor
00:22:55.691
Or beyond even. Yeah, kind of like, yeah, you spend this year 100K or 200 or even 300K on tokens. Yeah, you're fine. You're still within boundaries.
Darin
00:23:04.411
Okay, your point is great. You're within boundaries. You stayed under 500K based on your $150,000 salary.
Darin
00:23:13.681
Okay, great. That thousand salary plus all the benefits, let's call it 200,000 for the human and 400,000 because we stayed under the cap. So that's at 600,000. Now, let's assume that person was good. Not great, but good. So their total value is 600,000, 200,000 to the person plus 400,000 in token spend. The question's gonna become, okay, could I have hired two other people at 150,000 a year plus their benefits to 200 and gotten as much output from that three people without AI than I could with the one person with the big AI spend? How do we truly measure that? That's what it's gonna bo- boil down to, I think
Viktor
00:24:04.549
Yeah. That's going to become a thing. As the prices… I think the prices will be rising of tokens and the amount of tokens spent will be rising. Maybe prices not, but the amount definitely. So that will be the next big thing. Companies will reach the point where they will start questioning that. I still think that the answer is going to be… I mean, you probably cannot spend 50K a year yourself by trying very hard
Viktor
00:24:37.902
Okay, with Fable, yes, you probably could. But you would need to try very hard to spend more than 50K and which is already a lot
Darin
00:24:45.886
It is. And going back to Fable for a second, even if I spent 50K on Fable, I probably wasn't using it correctly, meaning I was probably using Fable for the wrong things if I'm spending 50,000 on it.
Darin
00:25:04.916
So that's where we as engineers have to learn of, okay, if I'm going to be using these tools, which ones are the correct tools? Which ones need to be advisors or orchestrators? Because if you go back to your agent deck, I'm assuming the way you've got things set up, certain roles, I can't remember what you call it in agent deck, what roles may use different models based on what they're doing
Viktor
00:25:29.078
Yeah. And actually I'm now g- I'm, I… One of my pending tasks is to switch some of the Opus models now to Sonnet The new Sonnet
Darin
00:25:38.390
The new, uh, Sonnet 5, yes. 'Cause it's not as good as Opus 4, but it's really close
Darin
00:25:53.958
But measurably better than where 4 was. So understanding how those different things work, this is the value of you today as an engineer, is knowing how to pull the right models off the shelf to use when you need to use them. Which tools do I need to use? Guess what? That's the same story as all along. We're just changing tool sets now
Viktor
00:26:14.778
How is that different than, let's say, managing infrastructure and then you can have clusters with 20% resource utilization, you can have clusters with 80% resource utilization? some are wasting money, some are not
Darin
00:26:26.847
And if that's okay with you, that's great. But if it's not okay with you, then as a business owner, what am I going to do? Now I'm back into the cycle of getting rid of people again
Viktor
00:26:38.937
y- exactly. You know what you do when you hit 20% cluster res- utilization, you have multimillion-dollar bill from AWS? The solution is to fire people
Viktor
00:26:50.567
No, just fire people. I don't know. The cost overall we're spending is too high. It's easier to fire people than uh, lay off people than to fix your clusters
Darin
00:26:59.982
That would be sad if it wasn't so true. Because some people can't figure it out and it's like, we can't get rid of AWS, so let's get rid of people so we'll be okay on our cash flow. We've talked about that earlier in the show. Whereas reality, if you'd spend a little bit of time understanding how much and how you're using a tool, AWS in this case being the tool, you'll find out if you're reasonably optimal or not. Am I hitting that sweet spot of 70 to 80% utilization? If I'm not, then why not? Those are the questions we need to be asking before I start firing everybody. Unless you're going to fire everybody because they were a transfer of people that only knew mainframes, so basically they brought their mainframe mentality and their client-server mentality to the cloud and then didn't know how to optimize from that point forward you
Viktor
00:27:54.067
You got it wrong, Darin. You got it completely wrong. Okay, so here's how it really works. When you need cash flow, you fire people, and then after you fire them, you try to figure out what's actually not optimal. And then you realize that you cannot figure it out because you fired people who can figure it out. And then you hire people, and then you realize that you're still bleeding money and you need to fire them again. Then you try to optimize, but you don't have people, and the cycle continues
Darin
00:28:34.837
It's always a cycle. So the question is, is can you either get back onto that hamster wheel or are you able to step off like I was able to do? If you were able to step off and it's like, "You know, I'm done," great, you're done. If you have to get back on, then what do you do if you can't find something? So let's play that out. Let's say you've gone through that hire and fire cycle a few times. It's like, "Man, I've been out here. I've been searching for a job. I've had the green banner or whatever up on LinkedIn saying 'I'm ready to work.'" And you're not getting a bite. You're putting out five CVs a day. Nothing's biting. What do you do after three, six months?
Viktor
00:29:14.242
there should be a six months job market okay, I just got fired from Microsoft. I'll be back there in half a year can I have a job here for half a year only?
Darin
00:29:23.180
That's where you go to work at fast food or something, I guess, or to the big box retail stores that are still open
Viktor
00:29:28.650
No, then you go to startups and go, "I'm a Googler. I know how things work at the
Darin
00:29:35.536
Wait, I know how to… I know how things work at scale. You don't say what you don't need because then you're actually able to get them to spend the more money and you look like a hero, but that's a
Darin
00:29:56.156
So you're saying now you were an IC, now you want to be a manager because you saw that managers are the best role to be in because they'll get rid of all the people actually doing the work before they get rid of the managers
Darin
00:30:12.080
Again, if it wasn't true, it would be funny Not always true. What are we supposed to do though? Because we gotta figure out what to do in this market. Unless you've started your own business and you're running your own business of course you have clients. Unless you're actually a one-person billionaire company and then great give us a call. We'd like to talk to you. How do we do this? Okay, we talked about the 50-year-olds. What about the people coming out of college now? They were sold this bill of goods, "Hey, go get a CS degree. It'll be great." And now you can't even get a job at Starbucks doing whatever Starbucks people do. And I'm talking about Starbucks, not the corporate company, but the actual place where they serve coffee What are people supposed to do? For people listening, Viktor's just shaking his head.
Viktor
00:30:57.478
Yeah, I don't know I honestly don't know. one advice I give to everybody, if you can Get the job for less money, even for free in one of the, magnificent seven companies, Google, AWS, all those things, because after that you're secure. Everybody wants a Googler
Viktor
00:31:20.789
That's probably the best thing how you can secure yourself, right? 'Cause if you work in a bank, then kind of who-- The, I think that getting fired as a software engineer from a bank is much, much worse than I don't know, Azure, Microsoft
Darin
00:31:37.038
I would say go work for government, but that just feels all sorts of bad in all sorts of different ways. S-
Darin
00:31:45.588
public sector is… U- unless you enjoy working in public sector. There are some people that they're built for it,
Viktor
00:31:52.174
The good thing about public sector is that nobody does anything and everybody has a li- a job for life, so that might be the pitch
Darin
00:31:59.815
We've already sort of mentioned it earlier AI augmentation, right? If you're gonna go ahead and it's like, "Look I'm in. What do I have to do?" There are some companies, in fact, I saw Sam Lambert from PlanetScale mention this the other day. It's like he had a guy that w- went through the interview process or I can't remember what it is. It was a tweet, and I saw it on LinkedIn as well. The guy was either going through the process or getting ready to go through the process. It's like, but he's vehemently opposed to using AI anywhere, like just completely a- against it. And and Sam's like, " Not to the interviewee, but just to the world of going, "H- how can you even exist like that today?"
Viktor
00:32:38.998
Exactly. That's the real crux of the problem, and uh, you know, people get fired after being for a long time in a company and with obsolete skills that's I think the key, right? Because there are certain skills that most companies don't need anymore. You could be DB2 expert 20 years. Man, you know it inside out. If you get laid off, I'm sorry, man. I'm really sorry for you
Viktor
00:33:07.371
Yeah, exactly. Either you migrate it from DB2 to something else and you don't need DB2, or you're still using DB2, but then you have that person that, that's been with you s- since the dawn of time. You don't need new people. So I'm not saying that nobody uses DB2, I'm just saying that nobody's hiring for DB2 Unless the person retired
Viktor
00:33:32.421
But even when, when people retire, usually then that's when people that were trying to do something and couldn't because of that person claiming that DB2 is the best thing since sliced bread that's when those people kick in like, "Okay, finally we can do this. He's gone."
Darin
00:33:49.481
One thing you could do is if you're looking for a job, pay attention to how those press releases, we were talking about the Cisco one at the top of the show. Pay attention to how it's actually worded and how it reacts in the market, because it's not necessarily just a layoff to get rid of bodies. It's an indicator of a rebuild happening within that organization, and that might be your place to jump in
Viktor
00:34:13.541
Here's alternative to that. Maybe actually you will know more how to navigate the world without paying attention to those in the first place
Darin
00:34:21.151
agreed. But like I said, if it just happened to come across your desk and it's "Oh, they just got rid of a bunch of people and it's bumping up, and they're not pushing it as AI, but man, they're getting rid of people that… But they're talking about using it," it's like that might be my fit." And again, going back to your thing, changing a title means a pay bump usually. And right… It's just like, wow, good grief, how many years ago now? 13, 15 years ago, data scientist
Viktor
00:35:05.565
Yeah, you just need to change the title. It's not about knowing what the thing is. Nobody knows what data science is
Viktor
00:35:14.385
Ah, maybe they don't know it either but they have the title. That's what matters. Let's take something closer to home. Nobody knows what DevOps folks are doing. We just know that they're called now DevOps folks
Darin
00:35:36.547
Oh, that would make sense, ' 'cause it f- it has a very military feel to it, so that would make sense that it's Palantir.
Darin
00:35:45.911
if you're looking for a job you're getting fed up because you're not getting any bites, you might just have to go do something to make ends meet. I had a friend that had to do that for a while. He was in tech searching for jobs like nine months. He finally just started delivering pizzas, and after about three months, he was making as much delivering pizzas as he was when-- in his tech job. He was hustling, but it was, uh… he had to do what he had to do. If you're close enough to retirement take a look at the numbers, maybe get a financial planner, this is not financial advice, and figure out if your numbers make sense. Or maybe you just ride it out because, again, we're in a hype cycle of hiring. We're not in the hype cycle of '21, '22, '23, but we're in one
Viktor
00:36:30.644
We're always in a hype cycle When was the last time that a hype ended without a new one coming up?
Darin
00:36:41.944
So bottom line is whether you're fired, laid off, or whatever the other word you want to use, rapidly reallocated. I think that's, that's how we're gonna name it from now on. Rapidly reallocated
Darin
00:36:57.421
Just don't give up. That's the big thing is don't get frustrated. Yeah, it's gonna be frustrating. Take the day, be mad, do whatever you gotta do. Go eat ice cream, whatever the case may be. I don't even remember what I did on the day I got-- actually, I do. I won't talk about that here. But actually, I did have ice cream. Let's put it that way. And just keep going. Yeah, it's frustrating. Yeah, it's whatever. If you're in the camp-- I, I will address this. If you're in the camp that's "Look, I am not going to use AI," then in this case you don't need to be using-- listening to this on your iPhone or your Android device or your computer, because there's AI happening. Not from us at this point. Actually, to prep the show, there was AI that was used. You'll probably, unless your car was produced before 2005 like mine, you probably have AI somewhere. Maybe you're using-- do you use anything like Waze or Apple Maps or Google Maps?
Darin
00:37:56.361
yeah, you're using AI. I, get it, and I do have empathy for people that's "Look, I am not going to use AI." That's great. You've chosen to not put a hammer in your toolbox. That's fine. That's your choice. But on my job site, everybody that's on this job site needs to use a specialized hammer
Viktor
00:38:14.990
the problem is when you make a choice that you don't understand the consequences of that choice, and every choice has a conse- has consequences, right? Uh, I don't wanna touch AI. I think that's a perfectly valid thing. You cannot be employed, unless you're already employed, then you stay where you are. I don't know, but, hey, why not? That's a choice uh, we do media and stuff like that, I don't use social, socials. I don't do LinkedIn, I don't do X. I d- I just post once for a video and that's it. I don't do any of the things there. And there are consequences, and I'm aware of them. And I said, "Okay," I kind of… "that's the price I'm, I'm going to pay." But I cannot complain how I'm not or how we are not exploding
Viktor
00:39:02.413
Yeah. It's a valid choice like any other choice, but we cannot complain afterwards "Hey, oh, we don't get enough clicks from X."
Viktor
00:39:15.423
I know. E-exactly, but what I'm trying to say is that we cannot we cannot complain about it
Darin
00:39:20.602
We can't complain about it. we use it as a notification channel and that's it. And
Viktor
00:39:29.748
Wait, you didn't post this morning what your, the color of your dog's poop on X?
Darin
00:39:40.922
I don't know how to play the game. I'm old. But to the peop- young people just trying to get into the market I wish I had good ideas for you. I think what Viktor said earlier in a little bit different way, try to get in somewhere, even if it's low salary or no salary, because having those human connections will probably get you further along in your career in figuring out what you want to do more than doing anything else
Viktor
00:40:10.874
Not only human connections, but the brand, right? Saying I work in bank XYZ is very different. I've worked, sorry, past tense, in bank, whichever bank, and saying I've worked in Microsoft, very different.
Darin
00:40:25.134
I'll even spin that one a little bit different. Let's stay in banks, and I will name off two banks. If I said to you, "Hey, I'm wor-" and no shade against either bank that I'm getting ready to list. I'm just listing them 'cause they've… what came to my head. "Hey, I work as a software developer at Bank of America." Okay, good. Hey, you're working at a big bank. Versus, "Hey, I'm a software developer at Capital One."
Viktor
00:40:51.959
Because, they do the funky stuff, people trust them. They promote themselves even within… I mean, it's similar to like, uh, so here, here's what I feel is a better example.
Viktor
00:41:11.154
Yeah. 'Cause they open source a bunch of-- You know about them doing stuff. I never met a person working, a software engineer from HBO on any of the conferences. I never heard any of the talks they made, I have zero understanding in how good they are at their craft. They probably are, but I have zero understanding. Netflix kind of, ah, is a potential employer. Yeah, Netfl-- Oh, yeah you do stuff. Yeah, Yeah, you're probably good. You couldn't be there if you're not good
Darin
00:41:40.225
exactly, but then also me as the Netflix employee would have to know who to apply to, because if I applied to Bank of America, that may not be as good as me applying to Capital One, right? Be- from Netflix, right? Because the hiring manager in Bank of America… Again, no shade against Bank of America. That's where I bank. There, I'll said it. the person at Bank of America doing the hiring may not know anything about Netflix because they're a new manager and they don't, they don't remember the heyday of Netflix
Darin
00:42:18.197
Y- that has six pages of, uh, how do you say it? Tools. Linux tools, Yes. Not commands, tools
Darin
00:42:28.703
Well, and that's the other twist too is if you're not customizing your CV for each place you are submitting it to, that's a big miss. I know it's more work. That's where AI can help you if you were using AI. Let's put it that way. Again, full empathy for people that don't want to use AI at all. I appreciate that. I am not one of those.
Darin
00:42:54.162
A little bit, but then you also were like, "No way." I was a few months behind. I had some other things happening in my life, but I was a few months behind. But now it's like I can't think how I would do a lot of the things I'm doing now, and think about the things I've actually shipped without the help of AI getting me to the point of, I'll say this, using AI to write up a bunch of task. If you're not familiar with task, task, task.dev? Is that what it is? Go to, Go to task.dev. Task files. Yeah, writing task files and having, boy, having Opus write those out for me, holy cow. I'm not using AI all the time, but I'm using AI to help me get to the point to where now I've got something that's deterministic that I can run. Could I have done it? Yes. Could I have done it as fast? No. Could I have done it as well? Maybe, but over time to me, it's just another tool, and I can use a tool for good and I can use a tool for bad. I'm trying to choose to use it for good. So if we haven't ticked you off yet I can't think of anything else I could say to tick you off. But it… I'm sure some people are gonna have some comments about this. So head over to the Slack workspace, go over to the podcast channel and leave your comments there