Is Automation with AI the Key to Navigating the IT Talent Shortage?

Ryan Clements:Automation is key and I think we've all found that out really quickly. The skills shortage is real and it's what a lot of businesses are coming up against these days–that it doesn't matter what platform or what automation tool you use, you just need to start somewhere on your journey, so that you can automate the rudimentary tasks, so that the people that you do have, the experienced ones, or the unicorns that you might have, can get on to those more difficult tasks, if that makes sense.

You can't get through a day without hearing about AI these days, and for good reason. AI helps us solve a problem quicker, so we can get back to what matters. And if we're not using it, we're behind the curve. If there's a skills shortage, we want to keep our best people. We want to give them something that they can put on their resume, such as generative AI,"I have experience there DevOps, "I transformed ClickOps to DevOps and we did that with Ansible Automation Platform." Those are real tangible things that make people happy, rather than turning that gear over and over and over a hundred times a day.

Marc LeBlanc: This isSolving for Change, the podcast where you'll hear stories from business leaders and technology industry experts about how they executed bold business transformations in response to shifts in the market or advances in technology.

In every episode, we'll explore real-world strategies and technologies that fuel successful evolution. I'm your host this month, MarcLeBlanc.

In this episode, we'll explore automation and how it's helping organizations to do more with less as the economy shifts and budgets become tighter. We are very fortunate to have Ryan Clements here to talk to when it comes to this. Ryan is an ecosystem solutions architect with one of our trusted partners, Red Hat.

Ryan's enduring passion for open-source software and his admiration for Red Hat charted his course. He made a pivotal move to Red Hat a few years ago, driven by the desire to help businesses harness the power of open-source software. Thank you for joining us on Solving for Change, Ryan.

Ryan Clements: I appreciate you having me on and it's fantastic to be here.

Thanks, Mark.

Marc LeBlanc: I wanted to start off the discussion by talking about some of the things that are happening in the industry right now that are driving the need for increased automation, particularly in IT. I think we're all feeling the effects of a skills shortage across the industry.

There was an IDC 2023 Worldwide IT Predictions report that suggested in 2024 shortcomings of these critical skills and training efforts byIT leaders will prevent 65 percent of businesses from achieving full value from cloud, data, and automation investments. What are your thoughts on this, Ryan?

Ryan Clements: That's a good... that's a great article and a great question.

I'm going to have to put on my slightly bigger glasses for this one. You're asking the hard questions already. You know, Marc automation is key and I think we've all found that out really quickly. The skills shortage is real. If you go on and I was a manager before this job, I managed a large data center.

So I had to hire and find people as well and it was really hard to find the right person, and we called it a unicorn. And then when we found that, we had to find more; that had to scale out. So, we realized that we had to do more with less because it wasn't just a matter of we weren't getting the, right requisitions for people. It's just that we couldn't even find them.

So we found that we needed to start automating with Ansible Automation Platform. And it's what a lot of businesses are coming up against these days–that it doesn't matter what platform or what automation tool you use, you just need to start somewhere on your journey so that you can automate the rudimentary tasks so that the people that you do have, the experienced ones or the unicorns that you might have can get on to those more difficult tasks, if that makes sense.

Marc LeBlanc: It does, and I'd like to ask a follow up question. There's a couple packed in there. You know, I like the messaging around the start somewhere and I'm wondering if you have any thoughts or insights around what's that blocker to companies or people getting started? What's that first hurdle they have to overcome?

Ryan Clements: Well,I'll tell you, it's not the technology. And this conversation comes up a lot and I think it really surprises people what the hurdle is because nobody ever thinks about it. It's the culture. Hundred percent the culture. You can buy all the technology in the world. Money can buy literally everything... Well, almost everything.

But it's those soft things, like culture, that people really don't think about when they're bringing a new technology in. They think, Hey"I have a pen. I'm gonna get a better pen that automates all the writing." Well, that's a simple analogy, but really people aren't going to use that new pen if they think that their job is going, they're going to automate themselves out of a job. And that's just not simply true.

So, there's a lot of pushback when we start talking about a new way of doing things, new charters in the job description, even maybe changing the job description. So, that's what I come up against and that's what I'ves een in a lot of companies that you can buy all the technology you want, but if you don't start with the culture, and have people adopt it as a new culture and a strategy, then it's a really hard journey and it doesn't have to be that way.

way

Marc LeBlanc: I agree. It always has to start with the people and the culture behind it. You know, I think some of the things that we see as far as getting started, within my own teams, I often put automation as the first thing I would like someone to skill up on. And the reason being, when they learn automation, they all of a sudden realize all these other things they enjoy to do are a lot faster, a lot easier, letting them learn and move quicker across other projects. You know, I think the other element of this with automation, we're hearing a lot around gen AI, and the tools that are emerging to help IT teams move faster and automate.

We hear about Ansible Lightspeed, we hear a little bit aboutIBM Watson Code Assistant, Amazon Code Whisperer, there's a number of others.How do you see that progressing and feathering into, some of the automation efforts we've been up to until today?

Ryan Clements: You know, it's a fantastic question. And, hey, you can't get through a day without hearing about AI these days.

And for good reason. AI helps us solve a problem quicker, so we can get back to what matters. And if we're not using it, we're behind the curve. I used to teach in school, I'm a part-time professor in security andLinux and automation. And, one of the things these days, that teachers are starting to, to... They don't want students to use AI. Well, I think that's kind of like fighting against the tide. Just as when we Wikipedia and the internet came out, we didn't want them to use that. AI is going to be used everywhere, let's be frank about it. And generative AI such as Ansible Lightspeed with IBM Watson X code assistant is fantastic.

If I had this when I was an automation engineer, I would have got from point A to point B much quicker and, doing it with best practice and doing it with being able to do something consistently the same time, every time. So, AI is fantastic and we really do have to look towards using AI to augment what we're doing in a company, because I guarantee it that your competitors will be and everyone else will be, just give it time.

Marc LeBlanc: I'm curious, I'd like to ask a little bit more on the culture side, I think that's an important piece. You know, I often hear when people are discussing gen AI and companies adopting, there's a couple of different tracks that come up, but the first one that you hear most prevalent is, "AI is going to replace me.It's going to make my job obsolete." What do you think is behind that? Do you think that's a real risk or is there something else that's likely to emerge?

Ryan Clements: It's nuanced, but in reality, when people say that they think they're going to automate themselves out of a job, there's two parts to that.

Number one, that is actually a real concern of people. They've been doing something for decades...I had a Linux engineer that I managed, and that Linux engineer could get on aSEV1 call at 2am in the morning and fix that thing before I could even hop on as the manager. It was fantastic, I loved it. But when we came to automate things, when I realized that I couldn't get more people to scale out as the company was scaling, we had to turn to automation, I had to solve that.

I said, "We have to do something." So we looked at automation and I learned it and I started implementing it, and rallying the team. And I thought, "Wow, this is it. Everyone's going to be so amazed and they're going to love this." How wrong I could've been. This engineer who could do everything, their main concern was I'm going to automate... He thought he was going to automate himself out of a job. And I said, "Nobody automates themselves out of a job. I've got a laundry list of a million great things that you can do besides expanding /var/log at 2 a. m. in the morning.Come on, we can have that automated or we can have like a junior level NOC admin to do that, so you can get your rest so I can, you know, task you on the greater projects that need to be done."

So that's one part and the other part is, you have a lot of people that have done things the same way, you know, the last 20 years and it goes back to exactly what I was saying before, but this takes a different angle: they don't want to upskill. They are comfortable with how they've done it in the past and they don't want to learn a new automation language or anything like that. They don't want to take part in that culture. But Ansible is, for example, I'm just picking on Ansible because I'm Red Hat, we love Ansible. It's a very easy language to learn. It's YAML language, it's everydayEnglish. It's fantastic. We even have Ansible Lightspeed to write it for you now, a generative AI. So, that really takes that out of there.

You can teach someone Ansible Automation Platform in a few minutes, and you can teach them to use Ansible Lightspeed pretty much with light speed, right? It's a matter of seconds. You write what you want to do in everyday English, just like ChatGPT or something like that. And BOOM, it writes the code, the automation code, which isn't hard. But it, it writes all that stuff for you. And it saves a lot of time. So the barrier for entry to learning new skills used to be, you know, go up like this, you know, it took some time to get an idea of how to write things best practice. That has been flattened now with generative AI.

And it's really fantastic.

Marc LeBlanc: I'm wondering if you could comment just on the thought... So, we're talking about how, gen AI makes developers' lives a lot easier. They can move quicker, they can find those reference points a lot faster. Why would businesses not just go all in on this? What is the value behind keeping that senior resource driving those AI conversations?

Ryan Clements: We're going to have to pick that question apart a little bit. So, let's start with why wouldn't companies go all in on generative AI? Well, let's be honest, they have tried. A lot of companies have tried, let's be honest. Like every company that has, you know, when ChatGPT and all the AI had an explosion last year, there was a lot of people who tried.

There's a few reasons why they didn't adopt ChatGPT specifically. And I'll leave you with this, one of my favourite things to say about ChatGPT is, and let me just take a drink of water here. The best thing about ChatGPT is that it gives you the most, or the best answer from StackOverflow. The worst thing about ChatGPT is it gives you the most popular answer from Stack Overflow.

And what that means, if you don't know what Stack Overflow is, it's just a place where people can post code, and it's typically not enterprise-level code. So, there have been a lot of blogs from the Ansible BU and just tests from myself and everything that found that, sure, it writes code and, you know, it can do a good job to kind of get you there, but it doesn't write enterprise-level. It doesn't know your company. It doesn't have context behind your company. So that was kind of stopped in it's tracks.

ChatGPT was really good for, you know, writing assignments for your little Python assignment or something like that. Or home, you know, like, home-built projects. It really didn't take in consideration the enterprise-level security and all the stuff that needs to go along with that.

And I had a long time ago, one of the best... Probably the best, security director I've ever known. And we'll call him David to keep him anonymous. But David said no to everything. And if the security operations director is good and worth their salt, they're going to say no to literally everything unless you can prove to them that it's not a risk.

And AI is a risk when it's going out to random places and you don't control that data that you've given it. So, where does it go when it goes to ChatGPT? What do they do with it? You know, it trains a model, right? Do you really want your company's information training that model? Probably not. And I know there's a lot of policies that when AI came about that said, you are not allowed to use ChatGPT or any of these models that we don't control.

With generative AI, Red Hat Lightspeed and IBM Watson X CodeAssistant, I'm really amazed, not amazed, but happy and thrilled to see that you control your own data with that on the IBM cloud and you can tailor that model to learn context about your corporation and organization and how you guys like to code. So, you have full control, you can delete the model, you know exactly where it's going. You can opt in to train their model, if you want, or you opt out. You have control and you don't have that with some of the public large learning models that we've all learned to adopt into our lives.

So there's a big security risk and concern from organizations over what people are using on those large learning models. So that's arelatively long question... answer to that question that I picked apart. Does, does that resonate with you?

Marc LeBlanc: It does. You've touched on a number of things in there.

I think it's important that people are always questioning as we're exploring these new tools that are emerging, where is your data going?It's so easy to pop online and find a new AI tool. Let's pump our credit card in there and let's go. But you really need to think about that data sovereignty, not just within your own country, I'm talking about within your own organization. Who has access to that when it goes out into the internet?

I'd like to change gears. We were talking a little bit about culture and what that looks like. AndI think that's a big, important part of the conversation. You can't talk about bringing some of these into a company without touching on what is the impact to your culture and how is it going to be embraced?

But the other part that always comes up is how do we start integrating these new workflows into our existing processes? And I'm wondering, maybe you could share some thoughts or insights around, you know, if there's a company that's evaluating or considering generative AI, how do they know if it's the right solution for them?

Ryan Clements:Automation is not just we didn't automate and now the whole company automates.It never happens like that. If you try to do it like that, you're going to fail. Unless you're a very small startup, because that culture and everything takes, it's a lag time. It takes some time to catch up. I like to describe a lot of these initiatives where automation or generative AI or adopting AI into current processes to help them. Or even app modernization from VMs to containers. I like to just describe it as a journey right? Just like if you decided that you're going to, start today in your home city and walk around the world, you don't do it instantly. And we're not going to do that incorporations and organizations as well. Like I like to say, the bigger the ship, the harder it is to turn, the slower it turns, right?

So, start small. And we'll go back to the automation thing.Start small and look at some small use cases of how we can go ahead and integrate that into a current process. But you have to start doing this across all of your silos, all of your, like NetOps, DevOps, Operations, Security,Development. And they all have to pick one thing to do and start small.

The trap that I see that a lot of people fall into is DevOps.Well, it's always DevOps, right? They get the new cool toys and they think,"Hey, I want to try this out and they do an awesome job. It's fantastic. It's brilliant. It's like it would rival any solution that some of the fan companies would have these days, but they get so advanced with that solution and it's not until that they're really far down the road that they start to try to share it with the other silos. And it just too far gone. The other likeNetOps and Operations are just too far back. There's just now such a big separation between what they're doing with automating something in one silo or using generative AI or anything, and their policies compared to what the other silo has done or not done.

So, it's important to look at it as a full strategy. You know, it has to start from the top and then everyone just pick a little project and share ideas have, a group meeting every couple of weeks to say, "Hey, what have you learned?" And now a part of generative AI, it's the same thing.

Generative AI is goingto help us write better code because IBM and Red Hat have gone a step beyond, not just by keeping your data private, but looking at the code that the model wants to suggest for the things you want to do and verifying it as a best practice. So, when you start writing with generative AI, you start writing the best practices that have been put forth by Red Hat and IBM.

You're not going to get two different solutions five seconds later. So, what that's going to start to do is it's going to start to transform how everyone writes code, Ansible Automation throughout your organization. And so it's just not a matter of writing code quicker, but it's a matter of writing code better.

And that's going to help anyone from seasoned Ansible Automation engineers, and trust me, I'm talking to you! You've written probably a million lines of code, Mr. Senior Ansible Engineer, but you've got a lot of bad habits. And we all do. And maybe you don't, but the idea is you want to write the same code that a junior would write. Because they need to understand it too.

So, it's a journey. You start small, you start little projects, and you start doing best practices there and working with that, and then you expand it out. You start looking at the code that you have written in the past, if you have any Ansible Automation code, and you want to start reviewing it with generative AI, you can shore up that code and get it consistent.

So that a junior can come in and say, "Hey, this senior engineer that was here five years ago wrote this and it's fantastic, but I don't understand it." To someone coming in and saying, "Oh, wow, I understand it because that's exactly how Marc wrote the code."

So, does that help? It's a journey and you got to start small and it really changes the culture and starts making things consistent overtime.

Marc LeBlanc: It does. So, I'm going to summarize super briefly, but I heard you touch on a couple of things. One was start small, find people that want to do it, start sharing those ideas so you can break down those silos, but then you said something I think is interesting. I'd like to explore it a little further.

You said, make sure you're starting from the top. And I'm wondering if you could give some more context because I have an idea of what you meant, but I'd like to kind of hear what you're thinking the top is.

Ryan Clements:Absolutely. And I always draw from my experience as an operation manager. I was always a champion for Red Hat, even before I came to Red Hat. That's why I wanted to come to Red Hat, I love it here. And I loved my previous job as well, but I really knew my heart was in Red Hat and my boss knew and everything.Besides that, what I want to say is when I was a champion in my previous jobs,I would try to always get something that would help my data center. So help my, guys do more with less or, or solve a problem. And it was very hard going from my lowly manager level upwards to director and VP and senior VPs... It was really hard. What I always got was, "Hey, you guys just want more toys, right?" "No, not really." And no matter how many, you know, Gartner reports or whatnot, or how many times we talked about it the idea was,"Hey, you guys are, doing all right with what you have, right?" I mean, you know, why do we have to spend money? You guys just want more toys."And that was a very hard way to sell my upper brass, or, any organization adopting new technology.

I could have got a very small environment for the people under me, but I knew that wasn't going to work. That was just going to create a big silo and we were going to end up in a big problem. Eventually what worked was, the VP heard about how fantastic this would be and suddenly it became their idea and it poured down And then we were able to get these things.

There's other reasons why you want to start at the top. Not because it's hard to go from the bottom up, which we all know. But when you start at the top, you start with the word strategy and culture. Because that's what those guys are focused on, let's be honest. You know, managers are focused on getting the work done, making sure that the people can get the work done and they have the right people for the job. And then the VPs and the senior VPs andCTO, CEO, they're focused on the strategy and the culture to help that strategy work.

So, if you can convince the top that automation or adopting generative AI. Or even if we're talking about containers. So, modernizing to a hybrid cloud infrastructure, taking VMs to containers. Then that strategy is communicated downwards to all the VPs under then all the directors under, and then all the managers under, and it becomes a company-wide strategy rather than some manager's tactic to solve a problem.

Does that make sense, Mark?

Marc LeBlanc: It does make sense. I think my take, when I think about it... When you talk about starting from the top, you really have to be thinking about somebody that understands the business problem that is trying to be solved. Otherwise, it very much is: this is a tactical plug for a tactical program.

So, if you go up high enough, we sometimes we think, if you're thinking about like cloud adoption, for example, you really need to understand somebody that knows what the application is doing before you make a single decision about cloud. If you started down that path without talking to someone that understands the business problem, you're already off track.

I'd like to shift gears a little bit. And I'd really like to kind of dig into what's this really all about? I think as technologists, we often think really quickly about automation, "Well, it's going to be faster," which is an element of it. I'm wondering if maybe you could speak to, what business problems are we really solving with some of this, this automation with some of this gen AI?What's happening and driving it, do you think?

Ryan Clements: Well, if we want to talk at the very top, we are being more competitive, that's a business problem that everyone, every CEO goes to bed at night thinking about,"How could we be more competitive? How could we be ahead of the curve?" But if you want to go a bit lower, now we're talking like CFO. CEO thinks about this too, but how can we save money, right? That's what it all comes down to. How can we turn a better profit for the next quarter? All these technologies, they have a great ROI, and it's not too long for that ROI to be realized. I think Ansible Automation, the slide that I've seen recently is a667 percent ROI. I love and hate that number. I hate it because it doesn't really tell the whole story.

Yes, you will get 600 and whatever percent to ROI on an average, but there's so much more there. And as a manager I realized that it's so much more than just shows up on a balance sheet. And that's what a lot of business objectives are trying to get is, let's save money. Let's turn a better profit.

But when you have tools that make people's lives easier, the people who are doing the work and managing the work under at the manager level, on the data center level, on the IT level, you now start to have happier employees. And that's what not a lot of C-level suite people realize is that, sure, it's going to save money, yes, And it's going to have a return on investment. But, what's intangible that doesn't show up on a balance sheet–wellI haven't seen it–is employee retention. Is happier employees and we all know that the happier the employees are, the happier the customers are, the better the service is going to be. We all know that.

So, if you start to invest in your employees, invest in the people that are doing the actual work, you're not only changing the culture on a very high level, you're changing the culture on a very where the rubber meets the road level. People are happier. People are more apt to helping other people out because they're not so bombarded with work. There is a million different business objectives that we could talk here about, Marc But those are the, the few that I really like to center in on. Because that's where my experience lies.

I've seen it. I've seen the front line. And I've seen the cost savings. And I've seen, the competitiveness. It's being able to make a better service than your competitor because you can maintain that five nines or six nines now, and the service is repairing itself. Does that make sense, Marc?

Did you have any other business objectives you wanted to touch on? I know I could talk here for hours, but I wanted to focus on those for you.

Marc LeBlanc: I think that's a great point that, yes, there is an impact to the balance sheet. We're seeing some sort of an ROI that's measurable. Fundamentally, I think we all understand the repeatable nature of automation.

I think that there's conversations around reducing human touch points for more secure product as well. But it's important and it's interesting you mentioned it–it didn't even cross my mind that this is where you're going to go–was the human happiness element. It's so unrecognized and I don't know what a great way to measure it is. I'd love to dig into that in a future episode, as well. How do you measure the impact of that culture change that you can bring around?

Ryan Clements: You know what, this ties into one of the first questions you asked. There's a skills shortage and people are going to go where they're happy. So you have to keep the good people, the wizards out there that can literally go to like anyplace and ask for a new job and you know they can. You know in your organization who those people are, right?

And it's not just a matter of dollars and cents, that's a lot, you know, salary. But it's a matter of how happy they are. How do we measure that? And so if there's a skills shortage, we want to keep our best people. We want to give them something that they can put on their resume, such as, generative AI, "I have experience there." DevOps, "I transformedClickOps to DevOps and we did that with Ansible Automation Platform."

Those are real tangible things that make people happy, rather than turning that gear over and over and over a hundred times a day. And they're making a lot of money, but at the end of the day, their eyes are glazed over, and they go home, they plop down on the couch, and they know they have todo the whole thing tomorrow, right?

If you can make those people happy, they stick around, and then they tell their friends. And when they tell their friends, now you get more people interested in working for you, and that's really what makes companies amazing is the people that work for you.

Marc LeBlanc:Absolutely, thank you.

We're just about out of time here, so I just wanted to recap some of the points we covered today, Ryan.

Thank you for joining us. We've kind of gone over a couple different ways looking at what that skill shortage means, what companies can do to address that gap.

Embrace some of the automation, embrace some of the new technology with gen AI, but understand there's a cultural impact that you're going to have to unpack and understand why people are feeling a certain way. We talked about how businesses can get more involved in bringing on these automation products and projects by looking at small wins. Get the company, the teams chatting and sharing notes, demonstrating what they've done with tools like gen AI, like Ansible Lightspeed, like the Watson X Code Assistant.

And really at the end of the day, we're talking about how companies are realizing value. Understand that it's all tied back into some business problem, helping them move faster. And, not forgetting that your employees are happier when they're learning new technologies, embarking on these journeys with them, and it should be part of the company's strategy.

Ryan, thank you so much for joining us today and sharing some of your insight. It's been a great conversation.

Ryan Clements:Thanks, Marc. Always happy to be here and have these conversations. You asked some great questions and there's a lot of different ways that we can go about it. And a lot of people understand the general questions that come back. But you and I, we dig into those questions that, or these answers that, people might not think about and it's very interesting. So thank you very much, Marc.I really appreciate it.

Marc LeBlanc: Thank you for listening to Solving for Change. If you enjoyed this episode, leave usa rating and review on your favourite podcast service. And join us for our next episode.

‍