LIVE DevOps Career Workshop

Key Takeaways

So uh I think we are live. This is the first ever Tech Forana live actually. Let me see if my camera is perfect. Uh all right. So let me open the chat here. Okay. Now I see hello from Usbakistan, Canada, US. Awesome. Okay. Um, I have some really cool stuff prepared for you guys. But before we get into it, I really want to kind of get the energy from the community because I saw a lot of people have registered, which is amazing. And it is not surprising for us because we see exactly the interest and demand for the type of information that I'm going to share with you today which is basically DevOps career. There's so much happening in DevOps world. So many things changing all the time. Um that sometimes we engineers think how you know it's probably time to standardize everything and not kind of keep everything uh changing all the time. Um but it is how it is and let's use this session to kind of get some clarity about the next steps, career um steps, career goals, how to navigate this this complex jungle of DevOps. Um I'm super happy for all of you guys who are joining. Um hello from Uganda I think. Hi from Georgia. Hi Beck India. Wow. We have we have the whole world. So, uh I'm right now in Vienna, Austria. So, in our time zone, it's 6 pm. Um so, I think I'm going to be going to bed right after the session because I think my approximate estimation is probably going to be like 3 to four hours including Q&A. And by the way, um I have a lot of slides prepared. I have statistics prepared. Me and my team research a lot about the job markets. uh we prepared a lot of uh data to give you some compressed knowledge but I think the most exciting part for me uh but also probably for you is the Q&A part because I want to hear your questions. We also received some of the questions before um that you guys submitted on YouTube community. So I'm going to try to cover as many of your questions as possible because I think that's you know the the core of the content today because I want to make sure that all of your questions get answered. Okay. Um so that means um we have people from again India, Greece, Tunisia, all over the world. That's amazing. Um hi from heat sink. [laughter] Um via Austria. Awesome. Um yeah, so including the Q&A, it's probably going to be three to four hours. That's an estimate. It could be shorter. Um probably not longer because then it's going to be like midnight here. Um, but we can get started already. Awesome. So, I'm scrolling the live chat and it's basically a lot of Okay. Um, so I'm going to tell you guys exactly when you can start posting questions so that your questions do not get lost in the chat. And Nicole is also on the on the live call and she's going to basically um immediately copy your questions. We're going to, you know, have bunch of them. So, we kind of we are going to try to categorize them so we can cover all of that. I have my water here in case I talk too much and my um my voice gets uh dry. So, I have everything prepared. Let's get started. Okay. Awesome. So, hope the slides are working. All right. Let's get to it. So, what are we going to cover today? What am I going to talk about? Um mostly we're going to focus on DevOps market the job requirements but also I'm going to do everything in a context of what's going on in DevOps what are the changes what are the trends what are the things that have become have remained constant because not it seems like everything is changing in the DevOps world because so much is happening and we're bombarded with this news but there are a lot of things that are staying stable and constant okay and it doesn't feel like this but in reality a lot lot of things like for example um a lot of the things that I was talking about five years ago on technology channel are still relevant sometimes even more relevant today um which means a lot of things are actually staying consistent so we're going to talk about that as well so you don't have the anxiety that you constantly have to keep up with with the changes of course we're going to address the impact of AI and again it's not as crazy as we uh are made to believe or it seems um from the outside. It's actually way more chill than it is and I'm going to dive into that. Um I'm also of course going to talk about uh career steps because we got lots and lots of questions from people who are junior engineers who want to transition into DevOps. You know how to uh what path to follow because I understand I was at zero as a DevOps engineer at some point and I remember looking in the future how overwhelming this whole thing was. So I'm going to try to dive into the road map and again answer a lot of your questions specifically about that. Uh and I also have some bonus tips about job hunts, how to prepare your LinkedIn profile, everything basically uh to to make it useful for you uh so that you can so basically by the end of this call uh I want you to have much more clarity about your specific next steps. Okay. Um, and I'm starting with DevOps market and some major numbers, but at the end of the day, you you don't care about the market trends. You care about what is it that you as an individual contributor, as an individual engineer have to do based on the data to um basically prepare for your next career step, right? What what does it mean for you specifically? So, that's what I'm going to try to focus on and use all these data uh to kind of feed into that. Uh so that's my main priority because I know that's that's what is the most interesting for us. Uh we still have uh we still have a lot of people joining seas service a lot of uh people from Europe, India, uh US, Canada. Awesome. Um okay. I try to keep the statistic as simple as possible because there are few number there are a lot of numbers but again let's narrow it down to what is the most important thing I'm pretty sure that a lot of you have seen these clickbaity videos on YouTube where people are like is devops dying you know is everything you know going to towards uh reduction in terms of devops should I still learn devops and this like kind of kind of making people people anxious because it's a clickbay title right people people want to kind of get that um intrigue uh from people and this is just one very official and a lot of people have shared this specific statistic uh from the official source that just in one graphic I mean look at this it shows in very clear terms that DevOps is not dying I mean we 2024 right that was last year um the DevOps market forecast right and and think of it what does DevOps market mean right think of all the DevOps platform forms all the technologies uh all the all the platforms that companies are using in order to implement DevOps and all all the tools that they're using. So the glo the total market share of all of that right the the estimate of how much that is worth all these technologies and and all these combined and look at the prediction in so this is nine years right not even 10 not even decade like 81 billion I mean look at does that look like devops is dying and people are needing less devops and obviously this trend is fueled by companies demands right because if if a lot of companies it's like a chain reaction or like a domino effect. If a lot of companies see value in DevOps and they um in in the practices of DevOps um regardless of the tools, they are going to keep implementing them. They're going to keep uh uh you know investing money in upskilling their engineers. There are going to be more startups in DevOps space who are going to offer resources because they see the demand. So that's kind of the domino effect of that trend going up. that means develop skills for those technologies is going to increase directly proportionally because obviously that that demand needs to be met somehow. So that this statistic should just show you DevOps is not dying. Okay, the trend is actually uh upward and uh because of AI it's even AI kind of fueled it and I'm going to get into that later. Um, so even though AI affected other engineering roles and kind of diminished their value, it actually did the opposite for DevOps and I'm going to tell you exactly why that happened. Um, and again tying it back to what does it mean for you as an individual engineer, right? Because you're not a company, you don't care about market forecast, right? So what does what does that trend actually tell you? Um and um and I remember like when when when I think about me myself as an engineer whether I'm you know I was working at engineering pro projects or consulting companies and you know software engineering teams on different projects of course I need to reverse engineer from what is it that companies are looking for right so what is it that uh the IT leads the the managers at companies are actually putting their focus on and again very very clear numbers I Let's start actually from here. So 2023 we had uh 25% of organizations that uh actively tried to bring like bring in DevOps in their companies and that means just just think about this thought process. So you have these IT managers uh not all IT managers are technical so sometimes they don't understand these terms right. So back then most of them didn't even know what DevOps was or they didn't really understand the concept. So it was it was more like you know the other company's doing it so maybe we should do it too right? So it was kind of copypaste um type of approach. So by 2027 there is an estimation again based on official official sources that 80% of organizations will basically say we need DevOps in our companies because we need to streamline our processes because it makes us uh competitive because our competitors are doing it so we need to implement this and that alone tells you that even those IT managers who have no technical knowledge they don't even like who don't understand DevOps even they already know that DevOps is mainstream and it's not kind Oh, should we do it? Should we not implement it? It's kind of like yes, we we should do it because our competitors will be doing it. It has become so mainstream. It's not even debate anymore. And again, I mean, you can like you can find all this type of type of um data online uh where you have a very clear comparison of how it was few years ago when I started uh teching on a channel and I started sharing my my knowledge on DevOps technologies. Um like a lot of company even even my project that I worked at there was usually one DevOps engineer in the entire company most of the engineers themselves like not even engineers actually understood what DevOps was. they heard of the term but the concept was still like a little bit niche and now if you talk to like just think about your colleagues think about people who are not like deeply technical everybody knows DevOps already right it has become this mainstream concept and it's not a niche anymore so we're not like we're very far away from you know devops being so mature that it now diminishes and and the demand is decreasing so we are in a very high maturity like we're going towards high maturity layer of DevOps. Um and um this leads to and this is this is actually probably statistic that is most relevant for you as an individual engineer because very simply put if companies put so much value in adopting DevOps because they have self-interest right they know that DevOps will make them competitive uh they can ship faster uh they can uh reduce the waste of their you know kind of kind of manual work that engineers have to so their engineers can can work on more productive stuff. Um so if companies are so convinced that they need DevOps but there is a shortage of talent that supply demand gap basically creates a tremendous opportunity for you if you really know DevOps skills right because that means that you have this massive opportunity because of the gap where companies are desperate for DevOps skills but there there is way less people or fewer engineers who can fulfill this gap right who have this um talent and look By the way, look at this number here uh or these statistics. This was a survey where companies were asked what is your biggest challenge. So you want to implement DevOps in your company. What is holding you back? What's the biggest challenge? So we had you know we have legacy architecture. It's really hard to implement DevOps processes when you have a monolithic legacy architecture when everything is kind of running on bare metals. You're not on cloud. You don't have the knowledge so to say in operation side. So first you have to kind of lay the foundation uh you know have a cultural shift where you explain to CIS admins you know you shouldn't kind of uh keep your secret scripts on your laptops and machines maybe we should collaborate on this um you know operation side and introduce infrastructures code. So you have all these issues but look at the number one DevOps challenge that they have for implementing DevOps which is skills shortage which which means let's say everybody's ready in the company everybody has signed up for it. Uh the the the engineers are uh basically ready to implement DevOps but who's going to do it? Like DevOps is not something that you just learn in a in a weekend and start doing it the next day, right? So the skill gap of someone who actually looks at the process and say how can we migrate our legacy architecture to the modern one. How do we move to cloud? How do we uh you know streamline our release pipeline? How do we even create a streamlined CI/CD pipeline? What tests do we need? You know how do we operate and run Kubernetes cluster? So these are skills that are missing in so many companies. And by the way, I remember um you know a few years ago when I was consulting uh companies, the the managers would constantly come to me and and say you know we we are have been looking for DevOps cloud engineers for a year and we can't find anyone so we need to upskill our existing engineers. That's how uh high the skill gap was and as you see it hasn't changed because the the demand is growing much higher and there are not many engineers who are who are become who are learning DevOps skills. Um and then we're going to dive into that who are those engineers what what those profiles are and so on. And I see some questions are perfect. So keep the questions coming. Uh Nikki is gonna actually filter them out. We're going to categorize them because we want to try to make sure most of the questions get answered. I will try my best to answer all of those. Perfect. All right. Um, so let's go to some interesting more interesting stuff. And this is this is where it gets really really um juicy and interesting for all of you and where most of the overwhelm probably uh comes in. And and by the way, write in the chat if this resonates with you. Okay. Um, so you start learning different DevOps technologies, right? So you start with Kubernetes and Docker um and uh AWS cloud and you know GitHub actions docker whatever and then you think first of all do I know enough to apply for the jobs and second of all which roles do I apply for right should I go for devops engineer then there is platform engineer then there is solutions architect then there is uh you know cloud engineer then there is s sur and then there are software engineers who are asking to do DevOps. So what what do I what do I apply for? There is an automation engineer. Um I think you guys see my mouse. I mean look at this uh array of DevOps jobs. Um yeah. So look at this like like this is a list of all the job titles that cover some set of DevOps skills. And when you know these bits and pieces of technologies dev technologies and you look at this picture or this list uh I understand why you you get confused because you don't know what to apply for you don't know what they are expecting and the interesting thing is that if you think about the perspective of companies and organizations they are just as confused as you are because they are looking for a specific set of skills but they don't know very often what title to give that uh skill set. They they have the same problem where they're thinking is is this a DevOps engineer role? Is it a cloud engineer? Is it an S sur? Should we hire software engineer who has uh these skills? Should we hire a test engineer who can build a CSD? So so they basically just randomly put out these these titles, right? And as I said, I've consulted a lot of companies who um have put out those titles and and I know how how their their thought process works and what their um point of view is. Um so let's let's clarify some of these points. Okay. And uh this this could be part of your DevOps job market analysis which is by the way something that you should absolutely do. Um so let's put it in two categories. Okay. So think of DevOps uh in general not DevOps engineer but DevOps in general as like a umbrella term. Okay. um like ve similar to when we had this fullstack engineer that was supposed to know every everything. So we have DevOps and then we have specializations within DevOps that emerged from DevOps. Okay. So DevOps engineer was one of the specializations. Uh there was not meant to be as a role but it's that's how it happened. Cloud engineer became another specialization of DevOps role. Dev sec ops engineer became a specialization. platform engineer SR. So think of this as under the umbrella of DevOps. So DevOps is much broader broader. Okay. So for example in uh in our DevOps boot camp we teach [clears throat] it's called DevOps boot camp right a lot of lot of people who who do the boot camp they can apply for DevOps engineer job they can apply for cloud engineer they can apply for platform engineer and for S sur because the tools that that you learn are very like they overlap in most of the cases. Um but because DevOps is so broad, you know, back then when I was uh when I was designing the curriculum for DevOps boot camp, I basically just put everything that was back then part of DevOps and that included um for example administration part of all the tools which is part of platform engineering. That included monitoring tools which is part of S sur job. for example, it included the cloud platform and the cloud infrastructure automation tools which is part of cloud engineer job. So all of that uh basically was automatically included and then it kind of emerged into the specializations where you have more like niche knowledge uh for example with each specialization so to say. So think of think of DevOps is one category with those specializations and then but that's that's probably like if you look at job market that's probably like 50% of the DevOps job or DevOps skill related job descriptions. Okay. So like basically the list of this uh job titles is just 50% of engineering roles that demand DevOps skills. So what the other 50% come from is this right here. So in in many cases uh in in lots and lots of companies because they are obviously more software engineers than DevOps engineers, right? What happened was kind of as a natural evolution was companies said we need to automate CI/CD pipeline. Uh we need to deploy to Kubernetes. Um who is going to do that? Software engineers. Right? We have software engineers. they're writing their application code already. So they should uh kind of extend their responsibilities and build an automate automated pipeline to deploy that application that they coded, right? And then they were like but where is that being deployed? Oh, on cloud and on Kubernetes. So they should extend their responsibilities and they should learn how to deploy their application that they wrote on Kubernetes uh and on cloud platforms. And who's going to manage that infrastructure? well they should manage it because that's their job to deploy it there right so very naturally it happened that software engineers um had to learn DevOps skills including Docker Kubernetes CI/CD and a huge amount of DevOps skill set is actually attributed to software engineer job titles okay so now if you actually search for software or let's say mid to senior level software engineer jobs you're going to see a lot of DevOps jobs DevOps um technologies listed in the job requirements. Um then you had another probably not as big of a proportion but very large portion of CIS admins um or network engineers who suddenly um were out of job for from those companies who said we're moving to cloud. So what is a network engineer who is you know kind of um working with the cables on an on-remise system need to do right so so they had to naturally move to cloud or you had CIS administrators who basically now had to work with DevOps practices to automate infrastructure provisioning uh to move to cloud and learn cloud infrastructure automation. So this became the the second biggest section and by the way uh if you have any of those backgrounds so if you're a software engineer CIS admin network engineer and you had to learn DevOps at your work uh as an extension of your existing skills please write in the chat and also share like what was your experience like and how that transition happened because we like we know from the data that we have actually like majority of our uh of the the people who join our DevOps uh programs since the beginning like the the number hasn't changed were software engineers and CIS admins like those were the the biggest groups and the third one that emerged in the last two years I would say were test automation engineers um and here it's a little bit more specific because with test automation engineers or QA engineers what happened was that uh AI came and AI was like I can write automated tests much faster much better than do so you know they they it kind of reduced the responsibilities and tasks of test automation engineers by half and there were a lot of QA um test automation engineers who were saying 50% of my job is now automated so what happened is that that kind of got cut off so they had to stack on additional skills to stay valuable and to kind of keep their um futureproof careers so the natural ext extension was uh the CI/CD specifically CI/CD because uh they were already writing automated tests running them. So it was where do I integrate those tests? How do I spin up a quick uh infrastructure to run my automated tests? Right? uh you know there are tests that you run in the on a static code but there are tests which are more important that you run on running application right so how do I spin up an environment where the application starts running and basically I I run the test against that and all of that skill set became a new extremely valuable add-on um extremely valuable add-on for test automation engineers and going back to the pointif like almost 50% of job market requirement for DevOps skill skill set um actually comes to those traditional engineering roles and only the other half is actually the specialized DevOps roles which means um if you only take the the DevOps specific uh or DevOps specializations as a market analysis you can multiply that by two and that's going to give you the real um perspective of how much demand there is for DevOps skills. And that leads me to the next point which is if you take one thing from this um live stream I want this to be one. Okay. A lot of people and I see this in questions as well uh in general think which role should I prepare for? Should I become a DevOps engineer? Should I learn from plat for platform engineer or I have all this um knowledge and skill set. Should I apply as a s sur or what is my my knowledge um applicable for which job title? And as I said job title usually does not reflect properly the skill set. So forget about the job title, forget the roles, okay? Just forget that they even exist. Think about the DevOps skills in the first place. Am I able to create a fully streamlined end to-end CI/CD pipeline for a software application that takes the the code that was just committed and deploys it all the way to at least development environment and then promotes it to staging environment right let's say not all the way to production but at least that if you can do that that is the valuable DevOps skill of creating building CI/CD processes is that that has a tangible value. Okay. Um think think about another use case which is can I spin up a Kubernetes cluster configure it with proper security um with because you know you guys know that Kubernetes by default in many cases is not production ready right it is working it is functional you can deploy your applications to Kubernetes but it's not production ready there is a lot that out of the box Kubernetes configuration does not have which is by design by the way um I had an interview with uh with Kel Kelsey high tower kubernetes was meant to be um a flexible simple tool and then we kind of complicated it by adding lots of stuff on top of it but per se it's not production ready so think ask yourself am I able to spin up a cluster a kubernetes cluster configure it with production grade best practices um that I can deploy my applications to automated way from CI/CD pipeline so Those are the DevOps. So that's the question you should be asking yourself. Do I have those skills? Right? Not do I know Kubernetes, do I know Terraform, can I build those processes? Uh or if I join um a company that is working with legacy systems, can I introduce an automated CS8 pipeline in the process migrating their you know basically mapping what they have to the modern tools? Can I spin up an infrastructure on AWS? Um can I do it in an automated way with Terraform? Right? Can I spin up a thousand node currency cluster? Can I spin up uh a thousand EC2 instances that are all connected uh but with backups in different regions? Like that's the valuable skill that companies are looking for there. Again, think about company's perspective. They don't care if you have a DevOps engineer job title. If you don't have the skill sets behind these, then even if you have that as a job title, even as a in your experience, it's not going to help you get the job because companies want the the skills. But even let's actually go even further a little bit further than the skills. Companies don't even care about your skills. Um, companies have their systems, they have their um, you know, software that they're building and they want to get it to the end users, right? with buck free, you know, secure and all this stuff and they have current systems. What they care about is an engineer joining the team who can uh optimize their processes, make things faster, uh reduce the manual effort, automate things in most cases, automate any anything that can be automated. Uh automate observability or implement observability. So you guys, your team knows basically that things went wrong before they actually go wrong and the users, you know, knock on your door and say, "Hey, your app just crashed. We can't use it." So basically be being proactive and finding those um issues faster. That's what they care about. The companies care about can you help us optimize our processes, speed up things, automate stuff, uh make our engineers lives faster because, you know, they don't have to do those those manual things. That's what they care care about and that then translates to DevOps skills and DevOps skills then companies try to at their best ability match to some random DevOps roles. Sometimes they get it right, sometimes they get they get it wrong. So always focus on what is the what is the skill set required to solve what problems for the companies and um actually and this is another interesting statistic uh which goes back to where we had it. Yeah. Uh as I mentioned most of the or at least half of the DevOps skill set actually is hidden in the software engineer job descriptions. And this is super I found this this statistic actually um super interesting. The the job role that required or that ranked the highest in terms of needing Kubernetes skill set was software engineer compared to look at the difference and again it tells you that there are way more software engineers than DevOps engineers which makes sense. Um but the DevOps skill set is kind of hidden here. Okay. And that's why when people say DevOps is dying, you shouldn't learn DevOps anymore. [snorts] Again, forget about the job title. The DevOps skills which obviously are becoming more and more demanded, they are in all those um in all these job titles that you're seeing here, right? They are in the software engineers have to know DevOps. DevOps engineers not need to know DevOps obviously and all these other um you know traditional or or modern engineering roles need to have the DevOps skills. So I want to change your perspective from thinking about do I need to learn DevOps uh uh you know engineer skills to what are the DevOps skills that I need to learn but also not technologies but what is it that I should be able to build. uh most of the employers does not understand DevOps role. I know uh that is absolutely right and I um they they don't understand DevOps role because it's it's not entirely their their fault because it has changed so much and it was so broadly defined from the beginning that they were you know it kind of had to go through few steps to standardize and as I said those specializations was actually one of the good ways to define okay now we have a cloud engineer that has a little bit more specific skill set. We have a platform engineer that has a little bit more specific skill set than this broad devops. But that means if you are like a full rounded DevOps engineer uh and like me you had to go through the pain of learning all these technologies and and skills because you had to at the beginning because there was just one DevOps engineer uh role then you you are basically you cover all these specializations automatically because you have you had to learn this these skills um and that's what I mentioned when I created DevOps boot camp I I basically just put everything in there that now translates to you know platform engineering to uh s sur to um cloud engineering because you were just a requirement back then uh for a devops engineer. Um [clears throat] so and that that basically just gives you another set of statistics of um uh so we we actually have a survey of um we have started this like four years ago in our programs and because I wanted to know like who are these people who are learning DevOps or who have to who are forced to learn DevOps and all the time like without exception software engineers were always top uh on the list in the rank uh who needed to and who were learning either voluntarily or nonvoluntarily those DevOps skills. So it remains till today that this is the largest group that has to have those DevOps skills and and um I'm I'm scrolling through your chat to see I think we do have a lot of software engineers uh in the session as our subscribers as well. So just some of you please write in the chat if you are a software engineer that was forced involuntarily to learn DevOps skills. Yeah, sis system administrators are um I would say second largest group um after software engineers um yeah and I mean you can you can basically find so many yeah that's and I remember when I was uh either consulting uh software development teams or when I was working myself as a DevOps engineer with with um you know microservices teams. I remember how uh how much software engineers hated Kubernetes because they had to work with it um in addition to their like software engineering tasks and they they were really forced to basically work with this tool that that was too complex. There was no abstra abstraction of Kubernetes. Um and um you know there was the times where software engineers were kind of forced you need to you know you need to learn cubectl you need to learn how to deploy to kubernetes you need to now learn how to you know automate CSV pipelines and I know it's not it was not the most um it wasn't the favorite thing for for those software engineers to do um okay so now that I hopefully changed your perspective from thinking about uh job roles and asking yourself should I learn that was engineering should I learn cloud engineering should I become this or that uh to what skills do I should I learn and th this is going to help you long term because what if let's say five years from now there's no DevOps engineer title let's say it all dissolves into the specializations and DevOps engineer title disappears there is no single job description you know of DevOps engineer which is very unlikely but let's say it happens right theoretically the skills underneath that now you know are part of DevOps engineer uh skill set they will remain right they will just be part of like different roles different engineering um uh yeah different engineering roles because what companies need to do in their IT systems and in their um you know um with their applications that they need to release to the end users that remains So the challenges and the problems the companies need to solve technically they will remain and the tools that are going to solve those problems also remain. Uh AWS is not going anywhere. Terraform is not going anywhere. There is a reason why these tools became so popular and so demanded because they solve real problems. So even if you remove the the job title entirely the skills underneath remain right. So now it's just a matter of uh thinking what are the tools that are most demanded and I learn um those skills those technologies and concepts and skills that allow me to solve specific problem which is can I build an entire CSD pipeline and what are this what are the technologies involved in that can I um administer and manage Kubernetes cluster or let's say it's already administered and it's already managed can I use configure um manage an existing Kubernetes cluster. Can I uh set up you know forget those two skills let's say you don't have those can I set up monitoring for the entire application ST uh infrastructure ST so and and what are the skills or technologies involved for that use case so think about use cases that are valuable for companies and they're not even so many of those like what I just listed are probably like half of those right u and then what are the skills needed for that and you learn those skills and if you're if you as a selfch check. Like if you're able to build those end to-end projects with those skills, then you are automatically valuable and you you can apply for whatever role uh the company um publishes as a job application that that will basically build that process, right? And this is a really interesting statistic which was um based on an analyzing 400,000 job postings and look at these statistics here. So we have AWS which is very reflective of so many companies now moving to cloud. Uh because again when I started uh creating uh videos on tech roomana I remember so many companies were still on premise they were not even hybrid like they were literally they had their own private clouds or they were completely on like you know operating on bare metals. Uh my first DevOps job I remember we deployed Kubernetes cluster on bare on our own um uh servers in Vienna in data center on bare metal uh virtual servers where we were literally installing the b the the docker binaries for the runtime the kubernetes binaries like every I needed to configure the networking everything directly on bare metal because we were not using any abstract we were I wasn't even using the CLI CLI to like the the you know um what is it the AWS CLI or EKS control like these type of abstractions it was literally just Linux commands and installing all the binaries directly on the on the servers um and that was not an exception like a lot of projects actually had that type of infrastructure I remember a company I was consulting for they they had hardware load balancers they were not even using software load balancers so there's A there has been a clear very fast shift to cloud which of course increased demand for cloud engineers and for those who wonder what is the difference between cloud engineers and DevOps engineers there's so many companies who are just using those terms interchangeably like practically if we really really dive in there is a clear difference because cloud engineer means you are basically uh more focused on you know cloud platform like infrastructure ure you know let's say you specialize in one cloud obviously so you know all the services they're uh important for setting up infrastructure for your application but guess what DevOps needs to know how to create infrastructure on AWS how to automate provisioning and how to you know um deploy your applications um on that infrastructure so there is a lot of overlap actually so as I said DevOps is like an umbrella term and then cloud is like a subset of it which is just the cloud part. Okay. Um Terraform this this actually surprised me a little bit but uh these two go hand in hand. Okay. Because if those companies who at some point later decided to move to cloud they also because Terraform was becoming so popular they obviously realized okay if we are moving to cloud let's do it properly. So let's use uh infrastructure as code to move to this cloud infrastructure so that uh we're not just taking our on premise mechanism or framework of doing everything manually and just moving that to cloud. Let's do it properly with automation as well. Okay. So Terraform and AWS uh demand at the at the same level makes sense because Terraform is basically used for infrastructure automation. Um, [snorts] and considering AWS is still the most used, the most widely used by far, um, cloud for most projects, it makes sense that the use of or demand for Terraform, um, you know, followed suit basically. Um, and another interesting thing on this statistic is look, Terraform is an automation tool, right? Python is an automation tool. Bash is an automation tool. Jenkins is also an automation tool because it automates the the release um of applications. Um obviously Enzible is automation. So on the top you have just one after another you have so many automation tools and the the you know it shows very clearly that it [clears throat] wasn't just that companies moved to cloud or move to Kubernetes but it was that companies were seeing obviously a lot of value in automating things instead of engineers doing things manually which was a very standard way of doing things before um and I mean Kubernetes and Docker I think we're not surprised to to see that on the list. And Azure is second largest and mostly used uh cloud platform. You also have GCP. Um the good thing about cloud platforms I would say in general is if you learn AWS like if you let's say if you specialize in AWS um as a DevOps engineer cloud engineer whatever if you switch to Azure and GCP like 70% of the underlying concepts are the same the names are different the UI looks different but like if you know the concepts you will know exactly when you switch to Azure Okay, where's the virtual machine? Where's the equivalent of VPC? Where's the equivalent of whatever AWS service? Because they have literally the same same services or same storage uh and compute services. Um and yeah and then you have you know GitHub actions which is you know in in DevOps boot camp I at the beginning I included Jenkins and we had in the last years we had so many people who were asking why Jenkins you know there are so many new tools you know why not GitHub actions which I know I agree with you GitHub actions is cooler uh GitLab CI is cooler than Jenkins Jenkins is outdated um you know you have these plugins that are maintained by some people and you have to you continuously maintain that um well there is a market demand okay so Jenkins I mean look at the difference Jen GitHub actions is extremely popular and widely used especially anyone who is um starting who's creating their own first CSV pipeline I'm pretty sure in 99% of the cases they will not use Jenkins right if they have the choice they will not go for it but it is still the number one most highly demanded skill skill right or or technology in DevOps because CI/CD being the core of DevOps and Jenkins is the implementation of the CI/CD. So it is still extremely widely demanded mostly because of these large companies who still use Jenkins um you know and obviously it's working for them. So the the pain of using Jenkins is not higher than switching to an alternative tool. So it remains and I believe I personally believe that it is going to stay uh like this for a foreseeable future. Um, and obviously if it doesn't, if at some point everybody just, you know, moves from Jenkins to GitHub actions or some other or some other coold tool uh comes along that replaces Jenkins, obviously we would completely scratch everything and update the entire curriculum because and and that's the thing like I like maybe with an exception of Kubernetes, but I don't I don't really care about any of these tools like that a lot of people get attached to like I love Terraform or I love whatever GitHub actions like they're means to an end, right? They have a purpose. They solving a specific pro problem. So I don't really have any attachment on any like super crazy preference of for any tool. If something better comes along uh or very pragmatically if something is extremely highly demanded uh because of some obvious reasons then that's the tool where you know you should put your focus on and attention uh to and there is no nobody who is telling you not to learn alternative tools and I think actually uh learning Jenkins and then learning GitHub actions or gitlabci those two are very similar gives you a much richer overall knowledge of CI/CD concepts than just learning uh GitHub actions and especially just learning the the most abstract uh abstracted um version of GitHub actions, right? Uh only using I don't know the built-in or out of the box features and not actually doing the the little bit low-level stuff. So, I like I like starting with the complex low-level stuff because just enriches your knowledge and gives you much more depth and more layers versus just going to one tool and just saying this is the best one. I'm just going to stick to it. Especially if that tool is easy because of an abstraction because you don't you don't know as much what's going under hood under the hood and it doesn't give you as deep of a knowledge. So some sometimes those you know painful technologies are good because they they force you to learn a little bit more under the hood. Um before I move from this slide I want to I wanted to highlight two uh tools Linux and Git. Okay. Uh do you remember I told you software engineers are the ones who are uh who need to learn DevOps the most probably uh sometimes even more than DevOps engineers like in terms of numbers they're they actually outnumber um when I was so I have a software engineer background when I was working as a software engineer in multiple projects um 90% of software engineers in my team including me by the way did not know Linux um I don't know if that's your experience as well. By the way, if if you were a software engineer, you did not have to work with Linux and most of them didn't care about Linux or learning Linux. Now um if you think about Linux in the context of DevOps like I think this should be actually put uh even even higher in the in the context of things because best scripting you need to learn uh Linux uh if you're working with Jenkins like you want to install tools um uh you know on on the Jenkins agents you want to manage Jenkins agents uh you want to configure or run commands in the Jenkins pipeline you need to learn Linux because most probably you know it's going to be Linux based the entire cloud is you know mostly based on Linux right the the EC2 instances um I mean Kubernetes docker like you know if you're working uh with Kubernetes if you're working with Docker uh yes I mean there are some desktop tools and abstractions but in most cases you're going to need Linux knowledge enzible modules you're executing Linux commands in many cases right so I think Linux is kind of became non-negotiable already and uh and this is interesting so the Linux was the became the non-negotiable for software engineers and git which was a software engineering tool became non-negotiable for CIS admins and the reason for that was CIS admins um we again have I saw in the chat we have a lot of CIS admins actually which is not surprising because I know they're they are second largest group that moved to DevOps. Um they basically um so put it to put it in context uh they were very used to um just having these scripts on their uh you know laptops or in their not just laptops like they they would have these machines that they like when I was for example managing the Kubernetes nodes and directly sshing into them. I remember I had like scripts and you know uh when the colleagues would log in like we would share those uh scripts and you know it was like very um decentralized right uh and there was no like there was no coll culture of collaboration on code and scripts between the CIS admins. So when we were managing the Kubernetes nodes, uh the the servers, the virtual um computers, you know, the or the infrastructure configuration, there was no culture of collaborating on those configuration changes. It was like everybody just executing their own commands and scripts. So with the infrastructure as code, with the Terraform, we're basically saying, you know, you're not sshing into servers. You're not executing some weird scripts from your own machine. You're not even executing commands. Everything is coded as a script in terraform, python, enzible whatever bash scripts and it's shared with git collaborated on just like application code and we are working on it as a team and all the changes go through the you know automation process like CI/CD pipeline which by the way gave us githops right so it was like a reverse of the so CIS admins now they had to work same way as software engineers so they had to learned git for example. [snorts] Um all right so enough of those stuff. So I have um actually I think this probably would be interesting for a lot of you because I saw it in the questions before that you guys submitted um the DevOps uh like in general uh jobs across countries you know across continents across companies um that require DevOps skills. I specifically focused on DevOps jobs, but as I explained, um you can if you search for the skills like if you search with CI/CD, CI/CD, Kubernetes, whatever, you're going to find jobs that do not only include DevOps engineer in the title, but also other ones, but we can focus on those ones. Um let me let me check some of the questions here. Yeah. Uh the I just exposed the CIS admin live. Yeah, I'm going to talk about envelopes and some like new trends. Um by the way. All right. So let me Oh, that's that's what's coming after impact of AI. So, and by the way, um we can keep this quick, but I just wanted to show you I mean I encourage everyone who is attending um to do market research in terms of like it doing this what I'm going to show you right now gives you so much data and so much um so much better understanding of like what's going on on the on the job market in general. Uh like just just analyzing and comparing like 10 different um uh job descriptions. You could be devos engineer, could be cloud engineer, could be software engineer that requires DevOps skills. you're gonna get so much data and so much deeper understanding of things and and get some kind of vibe of like what's going on generally uh on the market and how are companies thinking about this. And always when you do this type of analysis, please always think that companies usually in most cases when they post the job descriptions, they probably don't don't know for sure that the job title is right. Maybe an HR manager wrote the uh entire thing and they just copy pasted some template. uh probably the job description uh and requirements are way too exaggerated and the title may be misleading. Okay, so always assume this is not the perfect job description. The companies don't always know what they're doing. Um and it will it will put things a little bit in perspective. Okay, so let's let's just go through um a few of them and I want to I want to show you um this is actually great news for you guys. So imagine if you were an engineer and you learned a specific technologies and set of tools and uh in one company let's say and you got practice and you got experience in that and now you were looking for another job and you were browsing through different job descriptions and every company requires completely different stuff that you have not used. Let's say you know one company has Jenkins another one has gitlab and there let's say there were like hundred other tools for CCD and every company just wanted something else right it would make your skills very difficult to transfer from one company to another but we're literally in the most luxurious position as DevOps engineers specifically as DevOps cloud you know the deal right the the specializations that Most like these are literally just random picks that I that that I chose from the research in most of the DevOps engineer job description you will find 70 to 80% of the technologies which are the same like they don't they literally do not change genkins AWS GitHub actions terraform and then you have like cloud form and this is like for every single category let's say CI/CD uh AWS you usually have like one alternative. Okay, it's not like 10 or 100 alternatives. You usually have one alternative like cloud form is an AWS alternative like AWS service as an alternative for Terraform. I personally think Terraform is great