---
title: 'AI May Replace DevOps, But Not for the Reason You Think'
source: 'https://youtube.com/watch?v=jcrLdUPHQZQ'
video_id: 'jcrLdUPHQZQ'
date: 2026-08-04
duration_sec: 3455
---

# AI May Replace DevOps, But Not for the Reason You Think

> Source: [AI May Replace DevOps, But Not for the Reason You Think](https://youtube.com/watch?v=jcrLdUPHQZQ)

## Summary

The video addresses the widespread fear that AI will replace DevOps engineers, presenting interviews with practitioners and experts to argue that AI is more likely to augment than replace these roles. It emphasizes that AI's current limitations in accuracy and the growing complexity of infrastructure create new opportunities for DevOps professionals.

### Key Points

- **The Fear of AI Replacing Engineers** [00:02] — The video opens by acknowledging the common fear among engineers that AI will automate coding, testing, and infrastructure deployment, making their skills obsolete.
- **Anna's Career Transition** [01:12] — Anna, who worked in medical devices for eight years, switched to tech during COVID after seeing bootcamp success stories. She chose DevOps because it seemed future-proof, especially after ChatGPT emerged.
- **DevOps as Future-Proof Career** [03:21] — Anna explains that DevOps is a growing, challenging field with constant new tools, and she believes AI will not replace it anytime soon, but rather benefit experienced professionals.
- **Roi's Perspective on QA Evolution** [04:05] — Roi, a test automation engineer with 13 years of experience, notes that QA roles evolved from manual to automation to CI/CD, and he anticipates further evolution with AI and infrastructure as code.
- **Expert Insight: Brett Fischer** [05:05] — Brett Fischer, a Docker and DevOps educator, states that even the best AI models achieve only about 70% accuracy on GitHub issues, which is insufficient for production infrastructure.
- **Why 70% Accuracy is Not Enough** [07:14] — The video explains that in DevOps, a 70% correct Terraform configuration or Kubernetes cluster will not work, requiring human judgment to validate and fix AI-generated code.
- **Historical Pattern: Automation Increases Scale** [09:31] — Brett describes how each automation wave (from physical servers to cloud to containers) increased the number of resources a single engineer could manage, from 10 servers to 10,000 containers, without reducing jobs.
- **Key Concept: Automation Empowers Engineers** [11:24] — The video emphasizes that throughout DevOps history, automation made engineers more powerful, not obsolete, and AI is expected to follow the same pattern.
- **Kelsey Hightower on Open Source and AI** [13:04] — Kelsey Hightower discusses how open source democratizes AI, citing DeepSeek as an example, and argues that humans will continue to share and build, preventing AI from being controlled by a few companies.
- **Who Controls AI?** [15:35] — The video highlights that open-source tools like Kubernetes and Docker are foundational, and AI will be accessible to all, requiring engineers to deploy and manage AI infrastructure.
- **Two Worlds: DevOps for AI and AI for DevOps** [16:19] — Brett explains the distinction between building infrastructure for AI (DevOps for AI) and using AI to improve DevOps workflows (AI for DevOps), both requiring DevOps skills.
- **Opportunity in AI Infrastructure** [18:43] — AI startups need infrastructure for deploying models, managing GPUs, and setting up monitoring, but they lack DevOps expertise, creating high demand for engineers who can bridge this gap.
- **DevOps Engineers Who Use AI** [20:53] — Brett predicts a greater need for DevOps engineers who know how to use AI tools than for AI engineers themselves, comparing AI proficiency to knowing Terraform.
- **Joseph's Advice on AI Threat** [22:36] — Joseph, a staff-level security engineer at PayPal, advises that infrastructure provisioning and security will remain human responsibilities, as AI lacks a shared responsibility model.
- **Israel's Journey from Civil Engineering** [25:26] — Israel, originally from Nigeria, transitioned from civil engineering to DevOps in the UK, motivated by visa sponsorship opportunities and the high demand for DevOps skills.
- **Recruiters Find Israel** [28:44] — Israel did not apply for jobs; a recruiter found him on LinkedIn due to his certifications and keywords, illustrating the shortage of qualified DevOps engineers.
- **Interview Question on Pipeline Optimization** [29:57] — Israel impressed the CTO by explaining Docker layer caching and dependency caching to optimize a slow pipeline, demonstrating conceptual understanding over memorization.
- **Two Job Offers Before Finishing Bootcamp** [31:45] — Israel received two job offers, one with a contract sent within an hour of the interview, highlighting the high demand for DevOps skills.
- **Patrick's Advice: Be a Problem Solver** [33:08] — Patrick, a former network engineer, advises that companies want all-around problem solvers, not specialists, and recommends building one big end-to-end project.
- **DevOps as Cross-Functional Role** [34:29] — The video argues that DevOps is inherently cross-functional, making it less replaceable by AI, as it requires understanding the entire application lifecycle.
- **Rodi's Transition from QA to DevOps** [35:24] — Rodi, a QA engineer, learned DevOps skills like CI/CD and Docker to stay relevant, as AI can automate test writing but not the infrastructure around it.
- **Israel's DevSecOps Implementation** [37:48] — Israel improved security by ensuring containers don't run as root, using least privilege, and scanning for hardcoded credentials, leading to a 10% salary increase.
- **Strategy: Hands-On Projects** [40:48] — The video emphasizes building complex end-to-end projects, not just following tutorials, to gain the depth needed for real DevOps work.
- **Strategy: Portfolio Plus Certifications** [44:11] — Israel's combination of certifications (CKA, Terraform) and a public portfolio on GitHub/GitLab made him visible to recruiters and validated his skills.
- **Strategy: Continuous Learning** [46:19] — Successful engineers like Israel and Anna keep learning after getting jobs, which leads to raises and career growth, making them truly future-proof.
- **Investment and Payoff** [47:58] — Learning DevOps requires time and money, but the payoff is high, with graduates landing roles earning £50-80k in the UK and seeing a return on investment within months.
- **Market is More Competitive** [50:17] — Anna notes that job interviews are now more challenging and companies are more selective, making real depth and practical experience essential.
- **Summary: Why DevOps is Future-Proof** [52:23] — The video summarizes that AI needs infrastructure, increases scale, is a tool not a replacement, and creates new specializations like DevSecOps and MLOps.
- **Call to Action** [54:59] — The video urges viewers to position themselves now while demand is high, offering a free DevOps roadmap and demo projects as resources.

### Conclusion

The video concludes that DevOps is not being replaced by AI but is becoming more valuable, as AI creates new infrastructure needs and expands the scale of what engineers can manage. It encourages viewers to invest in learning DevOps with hands-on projects and continuous upskilling to thrive in the evolving tech landscape.

## Transcript

right now. Will AI replace my job as an engineer? And I see this fear everywhere in comments, in emails, in LinkedIn in comments, in emails, in LinkedIn messages. And I get it. AI is automating
code generation. AI is automating writing tests. AI is even doing infrastructure deployments in some cases. So naturally people wonder should I even bother learning coding or DevOps or cloud engineering or will AI just
make these skills obsolete in two years? And here is what I discovered after interviewing multiple people like DevOps engineers with different backgrounds who just transitioned into the field plus industry experts in the field. And the
industry experts in the field. And the answer honestly surprised even me. And I created this video because I did not just want to have theory of me explaining some of the concepts. I wanted to have real people with real
jobs with different backgrounds, real salaries and industry experts all giving different perspectives from their environments for a complete and full
picture. And let's start with the fear everyone has. So this is Anna. She worked in medical devices manufacturing for eight years and then decided to switch to tech during covid. >> Yeah, I think it was
it was a cumulative of things uh that I have already been working in the medical devices sector for quite a long time. It was like seven eight years. I also have worked for different industries and I saw that the type of job it was kind of
saw that the type of job it was kind of the same you know even change companies >> and I also it was the time during COVID that also boot camps were very famous you know and I was seeing a lot of stories on the internet from people that
they completely uh switch careers you know after doing a boot camp >> and I was really really intrigued by that and uh yeah and I decided to to bet on myself, you know, and do the same >> and it worked for me.
>> Yeah. Now, listen to Anna's reasoning for choosing DevOps specifically. This is really important. Honestly at the beginning I was preparing a bit for everything you know that's why I did data science data engineering you know
data science data engineering you know um but uh then I realized that uh yeah DevOps uh was like a very strong trend you know >> and then after chat GBT came out you know and all this revolution I saw that
yeah DevOps it's one of the more uh careerproof you know future proof careers uh that exist >> and I'm very happy you know that I ended >> and I'm very happy you know that I ended up in this field and also yeah it's a
field that it's always growing you know it's uh challenging always you know you will never get bored you know >> because there are always new tools appearing you know and you have constant constantly be learning
>> did you catch that after chat GPT came out she researched which careers were future proof and DevOps clearly stood out but why let me show you something that will make this crystal clear.
>> Yeah, exactly. Um, and as I say, I think this is this and as I say, I think this is this career is very futurep proof. I think AI is not going to to replace it anytime soon. So yeah it's only yeah it's only
benefits you know in the future because you have more experience more knowledge and also there therefore you get a better pay you have more options you >> call you you know uh recruiters yeah
know >> now Anna is not alone in this concern let me show you Roi he's a test automation engineer with 13 years of experience and he saw something happening in his field that made him
happening in his field that made him realize that he needed to act fast. Okay. Interesting. Did you also make the decision a little bit based on like what the future job requirements would look like? Like did you consider that in that
mix? >> Definitely. Yeah. Because as I see um what I well what I see in the field with QA is that initially you had manual QA then all of a sudden QAS were expected to do automation then all of a sudden
CI/CD so I can already well at least in my opinion I think that the next step would be even more because yeah technology is moving forward so QA also needs to move with for example with AI or with infras code these kind of things
you see the pattern these are not people running away from AI. They are running towards the skills that work with AI. And that's the key insight that I want you to understand. But before I show you why this works, let me introduce you to
the expert perspective. I interviewed Brett Fischer. If you don't know him, he's been teaching Docker and DevOps for over a decade. He has his own courses, his own podcast about AI and DevOps. And I asked him directly, is AI going to
replace DevOps engineers? And his answer is going to change how you think about this entire topic. So pay close attention to this. And even the best models that we have, the absolute best models at the top of the rankings are
models at the top of the rankings are only going to fix about 2/3 maybe if you're lucky three4s of GitHub uh issues. So if you so we we have this thing called SweetBench now and all the models are listed there and they're
constantly competing with each other for the top rankings. But even if you paid the most money the hundreds of dollars for the high-end uh clawed models or the for the high-end uh clawed models or the high-end GPT models you are at best
going to get about 70% accuracy which for a software developer 70% means it >> Yes. [laughter] >> Uh it's not going to function at 70% function eventually there'll be so many bugs right so it does I think there is a
where they're picking up steam and then they realize that we're still many years I think away from these models being so trustworthy that we can just pass on our
our our requests and not have to look at the answers um I think that that's a real so so the idea of like an AI writing the PR and then an AI uh reviewing the PR and approving the PR whether it's for software code or
develop or devops YAML I think we're still many years away from that um for worried about jobs and >> and interesting I think most of that stuff is really hyped up and don't believe anything in AI if they work for
because they're living >> they're not living in reality with the rest of us they're living like 10 years in the future or 15 years in the future and >> okay This is huge. Let me break down
what Brett just said because this applies directly to you. Even the best applies directly to you. Even the best AI models today get about 70% accuracy. That means 30% of the time they are wrong. And in infrastructure, in DevOps,
production environments, 70% correct means a lot of things break. So you cannot deploy a Kubernetes cluster that is 70% correctly configured. It will not work. This is fundamentally different from other fields. In content writing,
from other fields. In content writing, for example, if AI gets 70% right, you can edit the rest. But in DevOps, in automation stuff, if your Terraform configuration is 70% correct, your infrastructure will not deploy or will
just be misconfigured. Your CI/CD pipeline will not run. You will have to go through and fix all the issues and in addition validate whether the running or functioning code is actually correct with best practices and your security
policies will not protect anything if they're just 70% accurate. So what does they're just 70% accurate. So what does this mean for you? It means AI becomes a tool that speeds you up. But you still need to understand what's happening. You
still need human judgment to review the generated code. the YAML Terraform code. suggested the right Docker configuration or the wrong one. The skill that becomes
valuable here is not can you write YAML from scratch and learn the syntax of Terraform by heart. The valuable skill now is do you understand infrastructure
well enough to know when AI got it right or wrong. And that understanding that comes from learning the fundamentals and concepts and the purpose behind each
concepts and the purpose behind each technology in DevOps properly. not from watching a few YouTube videos or doing some shallow Udemy courses, but from actually building projects hands-on, like doing it yourself, breaking things,
fixing them, troubleshooting them, understanding why they broke, and most importantly, building complex projects and not just one Terraform script that creates an EC2 instance. Now, here is where it gets even more interesting. So
in all those generations of us moving from one one evolution of infrastructure whether it was a PC or servers or whatever each time the CIS admin grew. So I in '90s I could only really manage 10 servers. I don't even think I had 10.
manual. There was no scripting or automation. Open source wasn't really popular yet in terms of typical enterprise infrastructure. And as it open source grew, as the tooling grew over those decades, we went from and I
have a chart that shows like one CIS admin to 10 servers. Then with with a 100 servers. Then with the cloud, it was one CIS admin with a thousand servers. Then with containers, it it sort of went beyond servers and you
thought about workloads and it went to one CIS admin or DevOps person could do 10,000 containers. And our tooling got better, our automation got better. And each time it was all about a single individual managing a bigger and bigger
fleet. So I don't look at that as DevOps lost jobs or in ops lost jobs. We just were able to manage more. And it turns out so far at least in my whole career that companies and organizations are in
infrastructure as you can give them. Like they they will take as much automation. They always want to run more stuff. I've never seen a team have an inq like their their ticket queue be zero forever. Everyone's got bro bro bro
everyone needs more st more more things they want to launch more apps more copies of the app to make it stable more for capacity more more de more disaster recovery locations where we can spin it up on the fly in case this place goes
expanding expanding and I don't think AI is going to cause that to stop I think AI is just going to cause us all to manage more this is the most important concept in this entire video so I even suggest that you write this down
throughout DevOps history. Every new automation technology made a single engineer more powerful. It did not replace engineers. It made the individual engineers more powerful and productive. That has been a pattern. So,
one admin went from managing 10 servers to 100 servers with virtualization, then to 100 servers with virtualization, then to thousand servers with cloud, then to 10,000 containers with Kubernetes. Each time people feared this will take away
my job and my career safety. But what actually happened? Companies expanded, applications, they needed more infrastructure, they opened more data
centers, the automation did not reduce jobs, it increased what was possible per engineer team or per individual engineer. And AI is actually showing the
same pattern. AI will help one DevOps engineer manage 50,000 containers instead of 10,000 or even more. But companies will not fire engineers. They will expand their infrastructure even more. They will build new AI
applications that need deployment. They will need more monitoring, more more compliance. So the demand does not shrink. It actually grows. It changes its mode. So you will not be doing the
same thing as you were doing before. But it doesn't mean that the number of engineers with whatever skills become demanded will shrink. And you want to be the engineer who knows those new skills when the old ones get outdated. Who
knows how to use AI to manage infrastructure at scale or containers at scale or application deployment and development at scale. Now, let me show you what Kelsey High Tower said about this. And Kelsey is someone who has been
at the absolute cutting edge of cloudnative technology for over a decade. >> One thing I love about open source is the fact that it has a way of correcting,
right? It has a way of balancing the scales a little bit. Um, and there's >> Mhm. >> Oh, it was so great to see this because everyone thought that the endgame was Open AI, ChatGpt, and Nvidia. That's it.
everyone else would just fall in line. End of story. And then Deep Seek comes out and everyone's like, "Oh my god, whoa, whoa, whoa, whoa, whoa." And these guys like, "Yeah, look, maybe they copied some parts, but who cares?"
>> This is the power of open source. We take the best ideas and we democratized them. And so when they did that, it reset the thinking a little bit, right? Whoa, whoa, whoa. Looks like open source is going to be competitive here. Nvidia
isn't the only game in town. And going around the world, I'm seeing other countries getting inspired that, yo, wait a minute. We can build things too. We can create things too. We can compete too. We have good ideas. And so the way
I'm thinking is that humans will do what they normally do. They will show up and billion people in the world. I don't think we're going to rely on just like a happens to everybody else. That's insane. And so as as great as this AI
train is moving, let's not forget the momentum that we've been building behind open source for the last 30 years. And I think if I had to bet, I think the inertia from open source, the appetite for humanity to continue to share and
care about each other. We see this with recipes, right? People come up with an amazing dish, they cook it for other people, they figure out the recipe, and they share with each other. And I don't think that stops when it comes to things
with operating systems. We've seen it with protocols. I don't know why we won't see with AI, especially once we democratize like the hardware and the rest of the stack. So my bet is most people are good. Most people want to do
good things and if we give them the tools to do that, they will. >> Kelsey is talking about something deeper here. The question is not will AI replace us. The question is who controls AI? And in the open-source world, in the
DevOps world, we have a history of democratizing powerful tools. Kubernetes was open sourced by Google. Docker was open source from the start. Terraform, Enzible, all these tools that we use daily, they're open source. This means
AI will not be controlled fully by three companies. It will be accessible to everyone. And someone needs to deploy these AI models. Someone needs to manage the infrastructure for AI applications. Someone needs to set up the pipelines,
the monitoring, the security. And that someone is usually an engineer in many someone is usually an engineer in many cases right now. DevOps engineer, MLOps engineer, whatever role that may be. In fact, let me show you something that
makes this even more clear. Brett told me something that completely changed my perspective. There are actually two separate worlds forming right now. DevOps for AI, which is building infrastructure for AI companies, and AI
infrastructure for AI companies, and AI for DevOps, which is using AI to do DevOps work faster and more efficiently. And both of them need DevOps engineers. Let me show you. >> Yes. And in fact, that preceded
everything we just talked about, right? Because I think we we as ops and devops into one, I think we understood that we were going to have to start building or managing GPUs and GPU clusters and Kubernetes on
GPUs and like that we were going to maybe have to do that as just a new piece of hardware really in infrastructure for decades. you know, we've been managing the same memory, networking, compute, uh, disk, you know,
uh, we've been managing that stuff for, you know, forever since the PC was invented. And we haven't really had a to worry about we haven't really had a to worry about TPMs and TPUs and GPUs and AG APUs, I
think, are another one that's new. And these are new types of hardware. So for three years now, ever since really um like a little bit before Chad GBT, but specifically because KubeCon is an infrastructure conference. So it's
really about how that conference is how do we do DevOps on infrastructure. It took us a while to get functionality even to do that and then and to get the all that stuff and then how to know the difference between servers with GPUs and
we make sure that we get the we assign the right places in the right locations how do we schedule all these things how do we you know and that was even if you just look at the most recent KubeCon probably 50 to 100 talks on that very
topic and there's it turns out we now have years later we now have lots of examples well docu doumented examples, tons of KubeCon uh talks that you can look up on YouTube for free about
building infrastructure on top of these specialized pieces of hardware. And then also how do we what does it look like to run a rag yourself, right? What does it look like to deploy a model in a container image? Let me explain what
Brett is saying here because this is a massive opportunity that most people don't see yet. Every AI startup needs infrastructure. They need to deploy their models. They need GPUs. They need
Kubernetes clusters that can schedule workloads on expensive GPU hardware. They need monitoring to make sure their inference endpoints are responding. They need CI/CD pipelines to update their models. But here's the thing. AI
companies are full of data scientists and machine learning engineers. They know how to deploy them to production. They don't know Kubernetes or Docker. They don't know how to set up proper monitoring or logging or security. And
this is creating a massive demand for engineers who know how to do that. understand both traditional infrastructure and the specific needs of
AI workloads, which usually is referred to MLOps. And even if you're not working at an AI company, every company is starting to integrate AI in their workflows. And that means more infrastructure, more services to deploy,
more complexity to manage. And all of that again needs infrastructure automation, DevOps expertise, which gives you this nice additional database of context to your your AIS. you learn how to run the AI models in the
background using uh we all these little Python scripts and things like that that can now fit into containers. So that's the AI engineer that it turns out. And the AI engineer that it turns out. And then the DevOps engineer now if we're
going to use AI for DevOps doesn't have to know any of that and can just use the models and and type in system you know prompts and build scripts and And I think all three of these could live in separate silos more or less and
and they could they maybe share a little internal guild or or group together that just talks about the AI stuff in general. But you could you don't have to AI engineer to stay relevant in DevOps and you don't uh but if you want to um
you can do that. However, I would say I do think for a very long time we are going to have way less need for AI engineers coming soon. Then we're going to need DevOps engineers who know how to use AI. I think that that like we always
we we have this saying now that you know AI is not going to replace your job. A to replace your job. Well, that that does apply in DevOps that if you are about you know to me using AI for DevOps is just like saying I know Terraform.
knew Terraform, you were you were bleeding edge. that tool. >> Yeah. >> So, in the future, or if not very very soon, maybe 6 months a year, you're
you're going to start seeing DevOps engineer jobs that list you know how to cla code. This is the critical takeaway. You don't need to become an AI engineer. You don't need to understand machine learning algorithms or neural networks.
What you need is to be a DevOps engineer who knows how to use AI tools. Just like you need to know Terraform today, you will need to know how to use cloud or CHBT or cursor to write your infrastructure code faster. The skill is
not building AI. The skill is using AI to do DevOps better. And that requires you to understand DevOps first. You need to know when the AI generates correct Kubernetes YAML versus incorrect one. You need to know if the Terraform code
makes sense or creates security vulnerabilities. So AI makes you faster but only if you have the foundation to evaluate its output and that foundation is DevOps knowledge or the knowledge of the concepts and technologies behind.
Now let me show you again real people with real experiences who are living this reality right now. First, let me introduce you to Joseph, a staff level introduce you to Joseph, a staff level cyber security engineer at PayPal with
15 plus years building cloudnative platforms and now works at the intersection of DevOps, security and applied AI. So, I have two final questions. Uh the first one is a huge topic right now and I would really love
topic right now and I would really love to have your um insight here and also advice from you to uh a lot of the you know again like a lot of junior engineers who fear it and they feel threatened by it but also a lot of
experienced engineers who are maybe in a little bit more outdated IT roles like networking network engineers you know data science um you know CIS admins and so on. uh because you also have a unique perspective of not just you know
perspective of not just you know engineer but you're also uh a managing a manager of uh cyber security engineering uh at a large company. So what is your career advice to people that are feeling threatened by AI? And many times they're
asking themselves, is it actually worth for me to invest next two to three years learning uh something that may get outdated in the future because AI is going to take over any every single job including the IT roles. What would you
including the IT roles. What would you tell those people uh in a way that they tell those people uh in a way that they actually believe rationally um that they are not completely threatened and going to be replaced by uh AI?
Yeah, definitely there are couple of spaces definitely I think uh we can contribute a lot with all the other AI productivity tools right so security and the infrastructure side of it right the way we ship even if it is a it runs on a
bare metal server somewhere on a VM or somewhere right so infrastructure provisioning or managing the infrastructure uh that is definitely not going to change it's going to be there for a long time I will say and security
Right? We know uh we know about the shared responsibility model when it comes to AWS. Right? So you know what AWS will take care and what we need to take care what is the application deployment team will take care. So uh
these AI tools right now there is no shared responsibility model at all like it generates based on what it is uh trained [clears throat] right. So maybe the responsibility and ownership side of it is definitely going to be there for a
long time. Even though we buy different tools and other things, end of the day, the tools will be managed by us. >> Now let me introduce you to Israel. He's originally from Nigeria. He studied civil engineering. Then he moved to the
UK and decided to transition into tech. And his story shows exactly why DevOps is so futureproof. >> Okay. Uh so I need to say this that I
>> Okay. Uh so I need to say this that I came from not like an IT background like um I did civil engineering on my on my like back in Nigeria because originally I'm from Nigeria and um I went through the university and after my graduation
study civil engineering I you know something within me tells me that this is not for me like I feel that like there's no way I can continue you know going studying civil engineering and one thing I said to myself that I want a a
skill that is like a global skill because when you think about civil professions they're kind of you know specific to one country because it looks journey of trying you know front end development the back end the devops but
you still get it you did the course like what made you still persevere and be like I'm still going to continue learning um and finding out how to get into this field and not give up and say maybe it is not for me maybe I should
engineering like what made you stick to it long enough? >> Okay. Uh so one thing about me is that um when it comes to um so I have like that kind of resilience spirit you know. So what actually kept me not giving up
So what actually kept me not giving up all this why was uh you know I had the vision that okay if I give up because then I was in the UK and you know how visa restriction is and you know and all of that it means that after my studies
if I can't get a job that's going to sponsor my visa that means I need to go back to my country and that was a huge motivation for me that I can't go back to my country without you know uh you know having a company that will sponsor
me and you Being a DevOps engineer in the UK alone gives you you know that that opportunity that leverage for any company to sponsor you because um DevOps engineer are they highly [clears throat] skilled and they are highly paid and if
automatically you beat the salary threshold for company to sponsor your visa. So becoming a devops engineer alone gives you has already put you way above other professions and put you in a better position for you to be sponsored.
Israel went through front-end development, back-end development, tried multiple path, but he kept going and eventually he found DevOps. He did the CK certificate, then he enrolled in our devs boot camp and before he even
finished the boot camp, something happened. So, let's talk about that. So you basically uh get through like twothirds of the boot camp and did you start looking for the jobs because you felt ready already or how did the the
interview come about? >> Like it's is like my journey is just know how to say it but the truth is that I didn't even apply for the job like being that I was coming from a boot camp the thought of even applying for a lead
deos engineer role is just you know far away from me. So I would I didn't even have the the boldness or the confidence to to apply. So I think um the recruiter the talent recruiter for my company sent me a DM on on LinkedIn you know but then
not even posting anything. I just posted my CK certification and then I did Terraform certification that was what I did and there was nothing special about just added some keyword on my LinkedIn like DevOps engineer, listed some tools
on my bio and that was it. So let's stop right here because this is critical. Israel did not even apply for the job. The recruiter found him and this is happening more and more in DevOps. Why? Because there is a shortage of qualified
DevOps engineers. Companies are desperately looking for people who understand Kubernetes and Terraform CI/CD cloud infrastructure. The demand CI/CD cloud infrastructure. The demand is so high that recruiters are actively
hunting for candidates. But notice what made Israel visible. He had certifications on his LinkedIn. He had keywords. He had proof that he knew these technologies. You don't need 10 years of experience. You need proof
which is visible that you can do the work that they are hiring for. And here's what happened in his interview. Pay attention to this because it shows exactly what I have been saying about AI not replacing these jobs.
>> And one of the question I was asked that actually gave me that selling point like actually um you know um pick the interest of the CTO was you know if you have a pipeline that is taking too long to run for like 10 minutes. How do you
you know shorten the pipeline? How do you optimize the pipeline to make it run very fast? Because sometimes developers feel frustrated when pipeline is taking too long, you know, to to run and all of that. And that was one of the stuff I
How to optimize a pipeline by using caches, by caching dependencies, you know, uh leveraging docker caching layers in the in the docker file all of that. So that was what I explained that to the like I explained it very well. I
I told them how docker leverages you know layers. So as I then if you want to build an image it's better for you to place those layers that don't changes changes more frequently at the bottom. So I was able to explain this
conceptually and all of that. So the the CTO was just nodding his head and you know that was my first task as well when I joined you know I was trying to optimize the pipeline to make sure builds are many very fast efficient
builds are many very fast efficient using caches dependencies and you know etc. So you see what happened there? The CTO asked about Docker optimization. Israel explained the concept of Docker layers not from memorizing the syntax
layers not from memorizing the syntax but from understanding how Docker works. And this is exactly what I said earlier. The AI can generate a Docker file, but can it optimize it for your specific use case? Can it understand your build
process and make it 10 times faster? That requires a broader and more complex understanding from actually building projects, not just watching videos. And
to this. >> Okay, let me share something with you. So I need to say this. I landed two offers, not just this one. So I landed two offers. The first one with was with a company um in the company was based in
here in the UK. That was actually a devos that was a junior devops, right? It was a junior devos role and I I smashed the interview even before the end of the interview. They were already asking me like how much do they need to
offer me to take my CV off from the market that they don't want me to apply for any job. They just want to take me and there but because I was naive I was looking for a job. I was so fast. I just want to get something. So I I said okay
want to get something. So I I said okay I needed something between 30 to um 30 to 40,000 thereabouts. the okay they're going to send me the contract and just 1 hour after the interview they sent me a contract to sign so two job offers
before finishing the boot camp one company sent him a contract within 1 hour of the interview this is not luck this is market demand very clearly and
by the way Israel is now a lead devops engineer so he passed his probation after just one month instead of 3 months he got a 10% % salary increase after his performance review. Why? Because he keeps learning. He didn't stop after
getting the job. And let me show you another example. This is Patrick and he was a network engineer. >> Okay. Yeah. Good. Good. Good question. >> Okay. Yeah. Good. Good. Good question. Uh and I think uh yeah. So first of all,
a lot of companers are not looking for specialized people right now. I mean the big companies small still look for specialized people but the big companies look for people all around there that would take any task that they have some
Python knowledge they have some AWS knowledge maybe some Asure uh they are looking for basically problem solvers. They are not looking for software engineers or DevOps engineers anymore. They just want the work to be done and
that you are have this personality of of delivering the the the work they need. So if if I was just starting out out, I would probably join your boot camp
would probably join your boot camp again, build one one big project and try to learn uh a lot of different different different stuff to be just all around to to have >> uh yeah to to to be able to ship
something from from from the beginning to an end. >> Patrick said something really important here. Companies don't want specialists anymore. They want problem solvers who can work across the entire stack. And
this is perfect for DevOps because DevOps is inherently crossfunctional. You work with developers, you work with infrastructure, you work with security, you work with networking, you understand the entire application life cycle. When
the entire application life cycle. When AI comes in, someone who only knows one narrow thing becomes much more easily replaceable. But someone who understands the entire system, who knows how different pieces connect. That person
actually becomes more valuable, not less. And AI actually makes this role more important because now you are managing more complexity. You are deploying AI models plus traditional applications. you are managing GPU
infrastructure plus the regular compute, you need that broad understanding. Now, let me show you Rodie. He's been a QA engineer for 13 years and he saw exactly
what I've been talking about AI starting to automate testing. So, what did he do? really enjoyed, let's say, the more technical things. Uh, and not to discredit test automation because I really like and enjoy test automation
frameworks they really overlap with each other. Like the other Cypress is is very similar to Playright. Of course it has some discrepancies between one and another but when I was really doing the technical stuff that I didn't know I
wow this is actually interesting and challenging because I don't know. Uh so that was definitely one of the triggers for me and I'm not saying that test automation is easy because every problem has its own uh difficulty let's say
has its own uh difficulty let's say >> but I felt that it would be for me a skill that would be very useful and helpful to to get to uh get to know and learn because I can use it in my dayto-day. Um but next to that it was
jobs I really need to be challenged to stay motivated and to uh yeah move forward because I'm quite ambitious. >> What would you suggest them to learn as the very first step or like the the simplest entry point into DevOps?
simplest entry point into DevOps? >> Uh I would say uh pipelines for me together with Docker. >> Rodi identified the exact skills that make him future proof. CI/CD pipelines and Docker. Why? Because test automation
and Docker. Why? Because test automation alone can be automated by AI. But setting up the entire testing infrastructure that actually requires much more complex and broader DevOps knowledge. When AI writes your tests,
in a pipeline. Someone needs to set up the test environments using Docker. Someone has to design those systems and optimization of that architecture. Someone needs to configure when the
Someone needs to configure when the tests will run, how they report results, happens with those results, how to make them easily consumable. That's DevOps. And Rodie saw this coming. He learned DevOps on top of his QA experience. And
now he's a freelancer as a test automation specialist with DevOps skills. and he's definitely not worried about AI replacing him anytime soon because he's using AI as one of his very strong and efficient tools. Now, here's
something really interesting. Israel told me about how he's actually told me about how he's actually implementing defication
expanding and not shrinking. Did you get any feedback from your uh employer, from your manager about your because it seems like you were you know actively trying to improve the processes without definitely having to tell you like hey
Israel like can you can you do something like you are self motivated to learn you are self motivated to implement those learnings like what was your feedback from the the managers >> yeah so so on my performance review I
added that to um because we have like a performance review every every first quarter And being that I was on my provision was supposed to last for six >> I my review [clears throat] had to elapse for like one year. So in the
annual review I added that as one of my um how I improve the security posture of the companies by you know making sure that containers are not running as root users in production um to ensure that we are we are using the least privilege you
know for our deployment and also giving uh proper arrowback permission to service accounts that are running um stuff in Kubernetes you know and also implementing um scanning our containers scanning our pipelines you know you know
there's any hardcoded credentials and and all. So I added that to uh my uh what's it called? I added that to my uh PDRO and after that I had a pay rise PDRO and after that I had a pay rise which is like 15% I think no sorry 10%
of my base salary. So all of that actually added to my performance review and yeah so it actually helped. So I had a feedback from my manager was super happy. The CTO were happy. They were so happy about my my work rate and the
value that I've added the value that I've added to the companies you know um >> Look at what Israel is doing. He's not just deploying containers. He's making them secure. He's scanning pipelines for credentials. He's implementing proper
permissions. So this is deficults and this is one of the fastest growing areas right now. Why? Because AI applications handle sensitive data. They need security and every company integrating AI needs someone who understands both
DevOps and security. AI will not automate this away. In fact, AI creates more need for this because now you have AI models that could potentially leak training data. You have more APIs to secure, more services communicating with
each other, more a tech surface. So every new technology creates new security challenges and devs secops engineers are actually the ones who can solve them by automating security checks and security audits and Israel got 10%
salary increase for doing this work and it is a proof that companies pay for skills that protect them that give them very tangible value and security automation is definitely one of those skills. So now you understand why DevOps
is futurep proof. But how do you actually position yourself to take advantage of this? Let me give you the exact strategy based on what worked for these engineers who participated in our Davos boot camp. First of all, you need
hands-on projects, not tutorials that you can passively follow, not just certificates, like actual projects where you build something complex end to end. captures this. >> The DevOps boot camp. Um yeah, it was
super useful for me because I I don't have a computer science background, you know. >> But also I find that uh they don't really teach DevOps, you know, uh if you study computer science. So the only way
study computer science. So the only way to really learn it is either by the job, you know, if you have the experience or through a boot camp like yours. >> Yeah. Or self-study. But yeah, if you self study, it's a bit difficult to
really show, you know, that you have the level enough, you know, to work later as >> Yeah, that's a really good point. So, you can't have a traditional degree in
DevOps. That's that's current at least currently it's not possible. So, you have to have it either from self-arning or from experience um or like a like a boot camp or a professional program. This is important. You can watch YouTube
videos. You can buy Udemy courses and those are good for learning concepts or like very very high level stuff but they don't give you the depth that you need to actually work as a DevOps engineer. Like what are you going to do at the
first day on your job as a DevOps engineer when you get hired if you never build something that resembles the actual real life project. And why is this happening? Because these resources teach tools in isolation. You learn
Docker in one course, then Kubernetes in another video, then Terraform in a third course, but you never see how they connect together in a real project that is way more complex than just deploying a few EC2 servers with Terraform and
configuring some basic networking configuration. So in our DevOps boot camp, my goal was to make sure that the projects are completely end to end. are deploying an actual application to Kubernetes with a CI/CD pipeline
automation with monitoring with proper security configurations. You are building exactly what you would build in a real company in a real project. So you're actually prepared for that job. And that's what employers test in
interviews as well. They don't ask you what is a pod in Kubernetes or what is a service or what is the difference between them. They ask you how would you deploy this application to production and if you've only watched isolated
tutorials you will not be able to answer that. Israel told me he used concepts from the boot camp directly in his interview. Anna said that the Terraform and Kubernetes modules were the foundation for her current job. And
Patrick said that he would build one large end toend project if he was starting over. One solid complex project teaches you more than 10 simple tutorials because you have to make decisions. You have to troubleshoot when
things break. You have to understand why you're using each tool, not just how to you're using each tool, not just how to use them. Second strategy, portfolio plus certificates. Israel had both. And that combination got him two job offers
before he even finished learning. So I had just uh my little devops experience like when I was I told you earlier that I was shadowing my friend in Poland on cap Germany I just added some few things and I also butress on what I'm actually
project so I added my project on the project session with a link to my GitLab. Yeah I think GitLab. Yeah. >> So you included the the the project portfolios as well from the boot camp. >> Yeah. Yeah. in the GitLab and also added
some I also added my GitHub portfolio as well because what I do is that whatever I learn using GitLab I try to do it in GitHub as well just to see how it works. So I added GitHub GitLab and I also added my certifications just my CK and
and Terraform your portfolio is proof certifications show that you have passed the exam but a portfolio shows that you can actually build things and here's what most people don't know. Employers actually look at your projects. Anna
told me that they asked her specific questions about her configurations from her demo projects that she sent along. They wanted to know why she configured things in a certain way. And that proves that employers are interested in your
project portfolio and they want proof that you understand what your project is about and you didn't just copy it from somewhere. So your portfolio needs to be somewhere. So your portfolio needs to be on GitHub or GitLab publicly with proper
structure readmi file explaining what you built and why with clean code structure easily understandable for whoever is looking at them with actual working configuration not just empty files and combine that now with
certifications CK for Kubernetes maybe AWS certificates and this entire thing validates your knowledge together with projects they make you a strong candidate so you can stand out. And this is especially important if this is your
first ever DevOps job because you don't have a work experience. And this combination is what got Israel noticed by recruiters. He didn't apply, they found him because he had visible proof of his skills. Third strategy, keep
learning even after you get the job. This is what separates good engineers from great engineers and it's what makes you truly futureproof when you do not something. I will go back to that question. So when I joined the company,
it was like my was it called my probation was was to will last for 3 months but then even after 1 month I already passed my probation. So as I then I decided to um you know obviously the one thing about me is that I don't
like being in one place for a very long time. So I want to be upskilling. Israel time. So I want to be upskilling. Israel passed probation in 1 month instead of three. Then immediately enrolled in the deficops boot camp. Why? Because he
knows the learning never stops. Anna is preparing for the Redhead certification while working full-time. Rod is building additional projects and writing blog articles. Patrick is exploring AI tools and automation. Do you see the pattern?
This continuous learning is what really keeps you ahead. When AI evolves, you just evolve with it. When new tools emerge, you learn them. You stay curious. You stay hungry while you still have that comfortable job security.
Companies pay more for engineers who grow. Israel got a 10% raise after just one year. Anna says that recruiters contact her regularly now because of her continuous upgrade of her skills. Grody went freelance because he had options.
And that only happens if you keep improving and upgrading your skills. Now, let's talk about the investment because learning DevOps properly takes time and effort. It's not a weekend course and it's not a quick fix. But
here's what the payoff looks like. That was actually one of my questions I wanted to ask. So, uh let's zoom in into that moment like the day that you decided to to enroll in the boot camp. So, you did you studied the curriculum,
it matches the requirements of the job. So, you see what technologies are taught, which projects. So what was your thought process? So when you decided this is because it is a considerable amount of money when you don't have the
of kind of sure that you're going to get the job after it. So investment pays off. So what was your thought process exactly of deciding you know what I'm thought process was like the first thing I did was when I went through the road
map I saw all the tools and then when I compare it to most of the job descriptions like jobs and everything everything mashes especially the terraform the kubernetes the anible and AWS and those you know highly those were
high in demand as then you know in the UK and when I saw it I said I I I think you know I think it's the investment is actually worth it. Israel did market research. He looked at job descriptions. He compared the boot camp curriculum to
what companies actually need and he saw that they matched. And this is very smart. Do not just randomly learn technologies. Learn what the market demands. Right now that's Kubernetes, Docker, Terraform, CI/CD, cloud
platforms, monitoring, security. These technologies are not going away. Companies have massive investments in Kubernetes clusters. They have thousands people who can manage this infrastructure. And yes, it requires
investment. It requires time investment. Our boot camp takes six to sometimes Our boot camp takes six to sometimes longer if you're working full-time and learning part-time. It takes money investment. It costs money if you want
to get quality training. But compare that to the payoff. Israel lended a lead DevOps engineer role. And if you look at the job market in the UK, that's 50 to80,000 per year. Anna landed a DevOps role in
Switzerland. Patrick is working at the big tech company. Rodie went freelance and can charge premium rates now because of his edit skills. So the investment pays for itself like the first few months of your new job. And then you are
come. And here's what you need to understand. The investment is not optional anymore. the market is getting more competitive and evolving every day. Anna said this perfectly. >> Yeah, I find that I mean I cannot really
say because I I didn't experience it before but this is what I read that I find that now as the job market it's very challenging. Uh yeah the process it's more and more difficult and there are more rounds than they used to be and
also the difficulty you know on the interviews it's much higher than what it used to be. The interview process is harder now. Companies are more harder now. Companies are more selective. They are testing deeper. They
want to make sure that you can actually do the work, not just talk about it. And that means you cannot shortcut the learning anymore. You can't watch a few YouTube videos and expect to pass technical interviews. You need real
depth like real projects and real understanding. And you cannot build this deep technical understanding with simulated lab environments. We're not actually deploying stuff in real environments or with some shallow Udemy
courses or even YouTube tutorial. Even including our YouTube videos and courses because they're just not deep enough and practical enough to prepare you for an actual job. During technical interviews, they ask you stuff, ask you questions to
validate that you're ready for an actual job. And that takes structured learning, not random tutorials just chained together. A proper curriculum that builds knowledge step by step and projects, complex end to end projects
that force you to connect concepts together. And you also probably need support from the community and actual engineers when you get stuck. And this is why graduates from proper boot camps like ours succeed because they invest
like ours succeed because they invest the time and money upfront in building those skills. They build the foundation and then they have careers that last decades, not just years. Let me summarize everything you learned today.
DevOps is not being replaced by AI. It's a very clear and solid statement right now. It's becoming even more valuable because of AI. And here is why. As a summary, first of all, AI needs infrastructure. Every AI application
needs deployment, monitoring, security, CI/CD release pipeline because the AI models get updated and deployed with new versions and someone has to build that infrastructure and that someone is usually a DevOps engineer or engineer
with those DevOps skills. Second, AI increases scale instead of reducing jobs. Throughout history, automation made engineers more powerful. one engineer managing more servers, more
containers, more complexity. AI actually continues that pattern and companies expand because they can do more now. They need more engineers to manage that They need more engineers to manage that expansion. Third, AI is a tool, not a
replacement. No matter how powerful and self-thinking and creative it is, it's still a tool. You still need human judgment in many cases. You still need someone to verify the AI got it right. You still need someone to deploy the
production and take responsibility. AI gets about 70% accuracy right now and in infrastructure that means it fails. You need the knowledge to fix what AI gets wrong. Fourth, new specializations are emerging now. Devsops, MLOps, platform
engineering. So AI actually creates new roles not eliminating existing ones and all of them need DevOps foundation. You heard from the real people today. Israel from civil engineering to lead DevOps engineer in under two years. Anna from
medical devices to DevOps in Switzerland. Patrick from network engineering to big tech with a much more challenging and exciting job. Rodie from QA to freelance with DevOps expertise. They all made the same decision to
invest time and effort to learn DevOps properly once as a foundation and then reap the benefit of it for years and decades to come. Not from random tutorials. I mean, they tried, but they still needed structured training with
real projects to really get that knowledge. And now they have careers that are growing and expanding, not shrinking. And the experts even confirmed it. Brett says that we're many years away from AI replacing DevOps
work. Kelsey says humans will control how AI is used, not the other way around. And both of them confirm the increasing demand for DevOps engineers or architecture roles in general who have this skill set. So here's my
question for you. Are you going to wait and see what happens or are you going to and see what happens or are you going to position yourself now actively while the demand is still high and companies are actively looking for DevOps engineers
because the window is open right now. Companies need people. The shortage is real. But as more people realize this opportunity, competition will increase. If you want to see exactly what technologies these graduates learned,
curriculum gave them the foundation to lend these jobs, we actually have two resources for you that you can check out and see for yourself. First, you can download the DevOps road map that shows you exact learning path, the
technologies you need to know, the sequence of how to learn, in which order to learn each one, and it is the same road map that Israel used to research whether DevOps was worth learning or not, and it's free. Link is in the
description. Second, check out the demo projects from our boot camp to see the actual hands-on work that teach you the indepth knowledge, not just simple demo
projects. The projects that got Israel two job offers. The projects that Anna showed in her interviews. The projects that gave them confidence to answer technical questions. Also free to download from the link below. And if you
decide you want structured training like they had, our DevOps boot camp is available. But even if you don't enroll, use the road map, use the demo projects, start building, start learning without actually learning the practical skills
because whether you learn with us or somewhere else, the important thing is that you start the opportunity is here. DevOps is future proof and AI is actually making it more valuable, not less. So use that opportunity. The
engineers who understand this will actually thrive in the next decade and I will keep making videos or give you updates on where the trend is going so that you're always informed. You need to decide which one you're going to be.
decide which one you're going to be. Decide to actually really lock in and learn that those foundational skills to accelerate your career growth or you know kind of stretch it out and feel stuck or like running in circles for the
next month or years. So download the road map, look at the projects, make your decision, and let me know in the comments what your biggest concern is about transitioning into DevOps. I try to read every comment, and I'll help you
figure out the right path for your situation. As always, thanks for situation. As always, thanks for watching and I'll see you in the next