Guest host Nick Polce, Business Agility Practice Lead at Accenture, speaks with Mirco Hering, Global Offering Lead for Developer Experience and Platform Engineering at Accenture. The conversation, which was recorded live at the Agile Australia Conference, dives into the differing career models that can help engineers thrive in an Agile environment, with practical advice for leadership and project managers alike. Mirco also addresses one of the biggest preoccupations of the moment, Gen AI, and its strengths, weaknesses –even its environmental impacts.
Guest host Nick Polce, Business Agility Practice Lead at Accenture, speaks with Mirco Hering, Global Offering Lead for Developer Experience and Platform Engineering at Accenture. The conversation, which was recorded live at the Agile Australia Conference, dives into the differing career models that can help engineers thrive in an Agile environment, with practical advice for leadership and project managers alike. Mirco also addresses one of the biggest preoccupations of the moment, Gen AI, and its strengths, weaknesses –even its environmental impacts. And you can find out more about how Atlassian and Accenture are helping businesses succeed at scaling enterprise agility by surfing to our home page http://accenture.com/atlassian
Scaling Enterprise Agility
Episode 3: Dev X & Platform Engineering with Mirco Hering
Nick Polce: Hello and welcome to Scaling Enterprise Agility, a podcast brought to you by Accenture and Atlassian. It's all about how businesses can be more adaptive and responsive to the ever increasing rate of change around us.
I'm Nick Polce, your host, and Business Agility Practice Leader at Accenture. And we're recording here from the Agile Australia Conference in Sydney. Today I'm joined by Mirco Hering to talk about all things DevX, platform engineering, and agile. So welcome Mirco. It's nice to have you on.
Mirco Hering: Thank you, Nick. It's a pleasure to be here with you. Looking forward to the conversation.
Nick Polce: Awesome. Would you like to start by introducing yourself professionally and personally for us?
Mirco Hering: So, my name is Mirco Hering. I look after the software engineering practice here in Australia and New Zealand. And I'm also the Global DevOps Lead for Accenture. So I've been with the firm for 20 years, always looking for the right ways of supporting organizations in their search for productivity, for solving customer problems, and really bringing agile, DevOps, tech, all those things together for our customers.
Nick Polce: Okay. It sounds like you've done a lot over the past 20 years. How did it all start for you?
Mirco Hering: So I actually, funny enough, like before I joined Accenture 20 years ago, I did a master's in artificial intelligence kind of before it was cool. So obviously, as you can imagine, there's a lot of stuff coming back now that, you know, I have full of my old university books, but, um, after uni, I ended up joining Accenture really just for a year of consulting.
And then I realized that it's actually a pretty cool place where you're doing consulting with organizations, helping them solve their problems, and I'm a problem solver. So for me, it initially it was what we called development architecture, then Agile came up, then DevOps and SRE, then cloud, and it's all just tools, like, you know, like my tool belt get bigger and bigger and bigger and I could have more complex problems. So I've been doing this for the last 20 years. I've, worked with amazing people in the firm and in the industry. I made many friends and I still love how we are all collaborating and you can see it here at Agile Australia to just solve the problem. But there's no holding back, there's so many problems to solve that no organization can top it by themselves.
Nick Polce: And what was the one point moment in your career when it was like, there's a better way to do things? This agile, this DevOps way of working is the way we should be doing things going forward. When was that moment for you?
Mirco Hering: So it was probably about, 10 years ago or so I always, always believed in engineering. I was coming as a deep engineer, but for IBM before I went. Back into my masters, um, I always believed in engineering, but the DevOps side was natural. But I was sitting in a team where we were trying to do, uh, a very big tech implementation.
And there was this guy that I hired in as a contractor who talked to me about Agile. And every muscle in my body rejected it. You can imagine, I'm from Germany, so very kind of structured in thinking. And this guy kept telling me, just explain to people the problem. Let them solve it themselves. Let themselves organize.
Nick Polce: What is this guy on about?
Mirco Hering: Yeah, exactly. I'm like, are you crazy? Like, this is going to be mayhem.
Nick Polce: Yeah.
Mirco Hering: But I worked with him for like two years. And kind of came around to it. and then honestly, like just seeing the pride that I had in the team picking up things and just doing it without me telling them.
Not walking around checking on the project plan and stuff. Just really kind of like, yeah, that is a magic potion. Um, turned out it was a bit harder than expected. Yep. It's not just, let them loose and let them solve problems, but certainly like from there on, I kind of realized that Agile is, is the absolutely core, ingredient for what we need to do.
Nick Polce: Yeah. That's awesome. And now technology is becoming more and more at the forefront of businesses, yeah. Most companies now, technology companies, if nothing else, the engineering capability is getting more prominence in organizations than they have typically done in the past. How are you seeing that play out?
Mirco Hering: Yeah, there's the old saying that every organization is a software organization, right? I think that, that is true. and I had a couple of conversations here in the hallway track with people, when we talked about - we seem to be losing quite a bit of that interest in this kind of coachy coaching, right?
And I think there's a huge focus back on, on just engineering and delivery. Um, and honestly, if you think about it, we created incredible capability over the last few years. The cloud, like open source, microservices. Now take AI into it, take, you know, the machine learning stuff that we already had a while ago.
So there's a lot of extra capability. Five years ago, 10 years ago, you had like quarterly releases. I could ask you as a release manager, what are you going to do? And you were like, version six of this product, version eight of that product. You could name it. You can't do that anymore. There's so many more components, so much quicker.
If someone tells me they can manually govern IT, I don't believe them. Right. There's just too much going on, too many components, too many, not synchronized release cycles. Between your hardware, your software, your own customer solutions. I think there's a real need to find technology answers to that, that allow you to worry about solving the business problem while the technology side is abstracted away.
For me, the thing that has come up over the last two years is platform engineering or developer experience as an answer to that. It is, the developer should focus on solving business problems with software. And all the other bits we take away. So the security approved templates, the tools that you should use, you should need to make them all go away.
Nick Polce: So how do you unencumber them from the organizational anchor to deliver?
Mirco Hering: There's this beautiful term that I had never heard before like three years ago, cognitive load, right? Which is really what drags people away from solving their problem and worrying about the red tape, the document they have to fill in, all the other stuff in the organization that prevents you from doing things.
And I think the term cognitive load can really describe what we're trying to get rid of, or reduce. Like, how can I focus you on understanding your customer, solving their business problem, and everything else sort of disappears, well, gets abstracted away from you.
Nick Polce: So really at the heart of what you're talking about is, how do we get our engineers working and productive as quickly as possible?
Mirco Hering: Yeah.
Nick Polce: In the spirit of agility, what have you seen in that space, for engineers to deploy code, for instance?
Mirco Hering: I started as an engineer, so I feel with the engineers, they have to do all this other stuff that has nothing to do with writing code. Right, there's only two things a developer should do: understand the problem, and solve the problem. and everything else should go away. and unfortunately that's not necessarily where organizations are, because they have all these different tool sets, and it's tricky to solve. We had many, many, many, many attempts at creating enterprise by toolset, right?
And yeah, but they were so restrictive that then the developers tried to find ways around it. I've been to the cloud, created their own cloud instance, or, you know, created their own SaaS instance of a software. So finding that tension and the right space between centralizing and having common patrols, but the flexibility for developers That's really the trick.
And where we are now is in that kind of platform engineering movement, where people start to agree on reference architectures. And saying, we have to think about this as a product. Our product that we built, is the platform that our developers use, to solve our customer problems. Alright, and so as soon as you put your developers in the spot of a customer for an internal product, you really change the mindset.
And as soon as you start thinking, developers are not here to make mistakes and I have to restrain them. I have to really consider how can I get them to do their job the best way. It's just a really material mindset.
Nick Polce: And how do we give them the tools as well, right? To win the day.
Mirco Hering: And the visibility, right?
So that they know, like, okay, this piece of code that I put together has these vulnerabilities, or has failed these tests. Because... Nothing is more frustrating, and you and me have recently been in an organization, where it takes weeks for feedback. I write, I develop, I write code, and then six weeks later I get feedback that there's a defect in it.
Right? At that point you've moved on.
Nick Polce: The world’s moved on.
Mirco Hering: I'm already solving different problems. Right? And for a long time, we had a situation where an organization had a security team, for example, that sits separately. So, you write code, the security team, three months later reviews something, and then comes to someone else and says, hey, you need to fix this.
Because it's impossible to find at that point who actually created the code. It's completely out of context. So creating that visibility and allowing developers to really solve the problem, understand how good they are, and creating that mastery, and if you're from Dan Pink, the ability to really see how you're getting better at something requires you to get far too. And that's what we need to establish, and all the tools should be for that purpose.
Nick Polce: I agree wholeheartedly. The engineer is obviously a pivotal organization, we're establishing that. What's your recommendation to executives to create a world class engineering capability? What do they need to think about?
Mirco Hering: I think one thing that we really need to think about is the career experience of developers or engineers. Because, I think for too long people have looked for... Nearly like a one stream career model where, you know, you're an engineer, then you become a team lead, then become a project manager, a program manager, an executive.
And realistically, we should start thinking about different career models. There are people who enjoy that, and they want to, worry about the business side of it, managing teams. But there's others who are just incredibly good engineers. We need to look for ways, and we do this in our firm, to really establish kind of more senior engineering levels, right, individual contributors who work on those reference architectures that I mentioned before, who create blueprints, who create accelerators, who are really engineers. It also goes the other way. We put engineers into a slot that they own platform engineering, which is very much a product management role.
Nick Polce: Yeah.
Mirco Hering: Engineers don't necessarily know what is best for engineers. As funny as it is, right? So you need to really have that as a completely different way. So you have engineers really core for your product, for your customers. You have the people who worry about your engineering community and enabling them, both technology wise, process wise, governance wise. And then you have the people who go down more of the business path to actually have the commercial governance around it and really solve, well, budgeting and funding and all those kinds of things.
Nick Polce: We might shift gears now a little bit and talk about DevOps. It would be remiss of me not to, being the global DevOps lead. And you've literally written a book on it. DevOps for the Modern Enterprise. Do you just want to share a little bit with the listeners around what that book is about, or what the intention was in writing that book?
Mirco Hering: Yeah, absolutely. One of the things that I found incredibly frustrating, and I love the DevOps community, but a lot of the early DevOps books I picked up were basically books that said, this is how it should be done differently. Like, you know, more culture, break down silos. Uh huh. I agree. Now what? Right. And so a lot of the books didn't describe what you can do.
And so the reason I wrote my book was every chapter at the end has exercises you can do in your organization. It's the same stuff I do with my clients, and I do internally. And I wanted to bring this really, practical aspect to it. And basically really like a workbook. Like you have like, you know, a bit of theory and then you have some exercises you can run with your team. So you can bring it into action. You can learn from it. and that has certainly been my intention. The other thing that I was very, very passionate about is trying to find a way of writing a book that is timeless. And that is really hard in technology because there's new trends coming up all the time.
But the underlying strengths haven't changed. Right? We're trying to optimize the same process. We still have to stand up infrastructure and deploy applications, do testing. So really staying in that kind of middle space so that I'm not having to worry about, is FIV still a thing? Or is, nowadays platform engineering, is that still a thing in three years time? So trying to stay timeless and just focus on core problems.
Nick Polce: I was going to ask you wrote that before COVID, coming out of the pandemic, would you have changed anything out of the book?
Mirco Hering: I think the book is still relevant. And I see that one of the absolute pleasures is to see people that just today, someone popped up from Columbia with like a snippet of my book on LinkedIn.
And one of those anecdotes that I put in the book of my personal experience, I'm like, that is awesome. that makes you feel really good. Of course, I would change it, right, the book was written at a point in time when I still believe, um, that decentralization is a stronger force and now realizing that decentralization creates just so much noise that you have to be really careful with how far you go.
Nick Polce: It's that balance, isn't it?
Mirco Hering: Yeah. It's that balance. So I think I would spend a bit more time on governance and What you can do to create governance that is stable, but flexible enough. I was just writing an article for a journal in Germany, and I talk about this kind of breathing in, breathing out.
Right, so you breathe in and you let experiments run. And then you come back and say, okay, which of these worked, which one didn't? And we standardize a bit more. You prune a couple of the branches. Right, and then you do that exercise again and again. So that process of how you govern that.
It's probably what I would spend a bit more time on.
Nick Polce: And then how do you do that in a regulated industry? Many of our clients that we work with, or organizations out there, are heavily regulated. Have you given much thought to that breathe in, breathe out model in a regulated environment?
Mirco Hering: There's this beautiful thing at the moment, which is called governance as code and it's trying to do pretty much attestations in code, so that you don't have to worry about validating and confirming that you have done the right checks and controls. I think it's pretty new, and it feels like it's still a bit cumbersome. but I know some of the guys who are working in that space, and that's fascinating to me.
I mean, down to that they've written the language, and there's a book called, Investment Unlimited, that tries to describe some of that. It's definitely possible, but you have to think about it, and you have to create this kind of common layer of language, of process, that everyone has to subscribe to, right. And I think what has made this a lot harder is that everyone is working from home now, right, so, a lot of the attestations need to come from a process and not from Nick doing a review of my code because, you're not sitting next to me anymore.
Nick Polce: Makes it a lot harder, that's for sure. You're obviously out there in the industry as a thought leader in this space. You present a lot at conferences around the world. Are there any questions that you've got that have made a real impact on how you think about things going forward? And have there been other talks that you've listened to that have, got your attention over the past six months since you've been getting back out there and, uh, listening to, what people have been thinking about while they've been locked away and squirreling away in the pandemic times?
Mirco Hering: I spend a lot of time at the moment, really, realistically with this Gen AI topic, to be completely honest. And I find it amazing because if I go on my LinkedIn, I would say 80% of the posts have something to do with Gen AI, right, good or bad. but just the people who really deeply think about the impact of that and not the hype side, not the, it will take 50% of our engineers away, but really on, on market for us as a culture, I find fascinating. And so I'll give you two examples.
There's this kind of arms race on AI where we have AI create a content and because you can't consume that anymore because it's too much. You have AI basically consuming it and synthesizing it to the point of what you need.
So you have this arms race, which. It's the technology arms race, but also it is just the amount of energy it consumes and the amount of kind of just us having to be aware that technology is not the answer by itself. And you need to be ethically responsibly using that, knowing that the cost of that is basically lots of energy, lots of fossil fuel being burned for you to do a, like a chat GPT send an email to my boss that I'm sick. Right? Which consumes, 100x or 1000x the amount of energy than a Google search for a template.
Nick Polce: So there's an environmental angle here to the whole AI movement.
Mirco Hering: Absolutely. And then really a responsibility about that, So that you're not just using it for every little thing, which is what's going to happen.
Nick Polce: Which you're doing right now.
Mirco Hering: Exactly right. And just really understanding that you don't think that it doesn't add actually value. And then there's a cultural aspect as well. They very often talk about this in the context of law. I think I used this example recently with my leadership group, where in the early 1900s, if you would do a merge of the top 10 companies in the world, the contract would be 40 pages.
With the advent of the computer, it became 400 or 1, 000 pages because you had copy paste. With Gen AI, it's probably 10, 000 pages. So just the amount of contracts and content that can be created and consumed. Means we're just creating more and more fluff. And so, like, you don't have to worry about going to a lawyer for a lawsuit just let Gen AI basically post it for you, and you have this whole, like, social system will get stressed that they didn't need to in the past, right? Because you have all these transaction costs have become lower. You can create more and more content. So it's a very, very interesting space.
Not necessarily directly related to my space. But, oh my God, we're going to do the same for software. Think about that, to create more and more average software, really fluffy stuff, all that consumes more consumption cost on the infrastructure, etc, etc, right? So, really, that to me is the next barrier of something that we really need to think differently than we do right now.
Nick Polce: Fascinating. We can talk about this for a while, but we're just getting the wrap up here. So, Mirko, is there anything you want to leave the listeners with today? We've talked about the importance of engineering, we've stepped into DevOps. Is there anything that you want to leave with the audience?
Mirco Hering: My main thing for everyone is, take everything that we've done with Agile, DevOps, SRE, Platform Engineering, if you're trying to solve problems. And I would really like everyone to think, go back and, this is not about the tool. This is not about the message. This is not about a specific practice. It is about you understanding your specific problems and finding answers to that, use your whole toolkit to solve your problems. And if you need a friend, like tap anyone on the shoulder in the industry, talk to them, get an external viewpoint and then go after it and solve your problem.
Nick Polce: Awesome. Okay. Thank you very much.
Mirco Hering: It's been an absolute pleasure. Thank you so much.
Nick Polce: Cheers mate, bye.
Outro: You've been listening to Scaling Enterprise Agility, a podcast from Atlassian and Accenture. You can learn more about Agile Australia, where this conversation was recorded, as well as the work Atlassian and Accenture are doing together by using the links in our show notes. We'll be back with more conversations soon. Follow us now in your podcast app, and you won't miss an episode.