 Hi, this is your host of Limhartia and welcome to a brand new episode of our series TfHire topic of the month a.k.a. T3M and this month's topic is platform engineering is DevOps there and our guest today is Joe Duffy, CUN founder of Bloomy. Joe is great to have you on the show to discuss this topic today It's great being here. Thank you Before we jump into this topic I would ask you what kind of culture shifts that you're seeing are only happening used to talk a lot about DevOps I think one key common trait is especially in the innovative Organizations, they're applying a lot more of an engineering approach to the way they approach their platforms and the way they approach their infrastructure they think with a software engineering mindset versus, you know, maybe a Couple decades ago, we would think more about, you know, scripting or point-and-click, you know, click ops sort of solutions We definitely see people building systems using software engineering And related to that empowering developers like putting them in the in the driver seat But with the right guardrails the platform team puts in the guardrails because you know, hey a developer may Not know all the ins and outs of how to configure, you know database So it's not accessible over the internet or they might not know how to run their workload most cost effectively That's where the platform teams experience and their expertise still empowering developers but With the right sort of guardrails in place I think is a key challenge and and trend that we see as well. How would you define that for me new thing? I think platform engineering is Really a new new trend that we're seeing Especially in large organizations where you need to have a center of excellence for how you do cloud infrastructure Cloud engineering best practices and the platform engineering team typically sits between maybe this the operations more IT centric folks and Developers and their goal is to really be like I said center of excellence establish best practices put in place the systems That allows the organization to scale and ultimately just empower the rest of the organization to ship faster With confidence compliance cost controls all the things necessary in sort of a modern enterprise Looking to go to the cloud. How different is platform engineering from DevOps or let's also say as sorry if you make a win diagram Are they overlapping or they are totally different, you know, it's fair on their own I do think they're slightly different, but they're they're very much related. I think platform engineering sort of grew out of the DevOps movement and you know DevOps was all about applying engineering discipline to the way you're doing operations And sometimes putting developers more in the driver seat and platform engineering just takes that sort of to the next level We see a lot of different configurations for how teams do platform engineering and not everybody does You know some folks and that's where SRE is very common You know if you're just a back end service team, you know an amazon style two pizza team Let's say you may not have a platform team You may you may not need one it may be more distributed than centralized And in that case, you may have an SRE a site reliability reliability engineer Whose whose job is to be embedded within that service team and make sure there's operational excellence and And that's sort of model. So we see a fair bit, you know Of difference between the ways people approach it and frankly the way they approach it over time You know what works for you as a small organization a small company Maybe totally different than once you get to the scale of thousands of of employees A lot of these labels are used we talk about super cloud that we talk about, you know But in the ground reality since you also deal with customers They try to solve their specific problem And that's why either they need the teams or they need products, you know That from unique can be seen as a product or said DevOps can be seen as a practice We look at SREs. So what is happening in reality? How much of these labels actually matter? What kind of you know, cultural transform is actually going on The reality is platform engineering is a real thing, you know And we've seen an acceleration in teams moving to that model And I think it's for a few reasons one, you know, the complexity of the cloud is just growing day by day Especially for organizations that have to do multi cloud or maybe they're thinking about kubernetes or CICD for their not only applications, but also their infrastructure There does need to be a center of excellence for how that gets done in a large organization But another trend we see that that is Putting pressure on teams to adopt platform engineering is is empowering developers to be more in control of their own destiny To have a way to self serve the infrastructure So if a development team needs a new database or a new microservice environment They don't have to file a ticket and wait for days or months They can just go self-serve it on demand as they need it and that's really important in a modern organization especially if the cloud is part of The product, you know If it's a large-scale product like you look at something like snowflake You bet they practice, you know Something that looks like platform engineering and they do empower developers and that's definitely Increasingly common not just in these innovative new companies, but also in the fortune 500 and global 2000 as well How would you define developer experience in today's clouds and the cloud world? Yeah, I think there's a lot of different kinds of developers in the world And so there's not just one developer experience that I think about But I think about, you know, put yourself in the developer's shoes. What what are they trying to build? You know, they're the ones that are creating end user application value They're they're shipping value to your customers. So if you put yourself in their shoes You want to just make them as productive as possible doing that thing You know adding business value and sometimes that's, you know Back-end systems engineers that are building distributed applications that are comprised of containers and serverless In that case, hey, make it easy for them to spin up new microservices Don't put unnecessary bottlenecks in the way a good developer experience is really just getting out of the way And letting the developer accomplish what it is they're trying to accomplish That but that's so in that case you might give the developer A way to consume infrastructure using infrastructure as code However, maybe your developers are front-end developers or full stack. They're building, you know Angular apps. Well, in that case, maybe they don't care as much about touching the infrastructure There's a lot of different kinds of developers in the world You might you might have actually both in your organization. So there's not one size fits all I think developer experience to me is all about just Making developers life as easy as possible making them productive and help helping them get their job done faster What should be the right approach? What would your advice then people hear these terms? How they should approach it that hey, we need to embrace it not embrace it. We don't even need any of that Yeah, it's a good question It's very difficult too because you know, there's it seems like especially in the cloud space There's a new AWS service shipping almost every week and you know a new term like platform engineering coming up You know every every year or so and I think the the key advice I would give is These things were created to solve problems and it's important to identify Do you actually have the problem that these were invented to solve? Before just going and blindly applying it, you know platform engineering Like do you need a center of excellence because your organization's at the scale where it makes sense to centralize that expertise You know kubernetes, are you trying to ship containers? You know and scale them at high velocity with reasonable levels of complexity Likely across many environments and potentially multiple clouds. Okay kubernetes It's probably a good solution for you But I think it's important to just distill it to to first principles and then assess Do I actually have the problem that this solution was created to solve? And if not, it's okay to kick the can down the road and and revisit it, you know over time Like I said, what works at certain scales stops working down the road and that's okay And I think it's important to kind of assume that change is constant Because you will have to evolve and adapt the topic is you know platform engineering is DevOps that Last year at coupon. Also, there are some boots, you know, which DevOps is that What is your opinion is DevOps really dead? Can it really die? I think absolutely not. DevOps is alive and well I think platform engineering was really born out of DevOps and in fact, if you were to draw a lineage I would say platform engineering is heavily derived from the DevOps practices Um, and I think it really inspired the way we approach platforms in some sense It's sort of DevOps taken to the next level Um, and you know, maybe we'll use the term differently going forward You know, I think you'll see people emphasizing platform engineering, but in terms of DevOps being dead like I don't foresee DevOps days going away anytime soon I think people are still going to use the phrase for many many decades to come And I think we all should be thankful for DevOps Instead of sort of throwing it under the bus prematurely Is my take Joe, thank you so much for taking time out today and talk about this Interesting topic and thanks for sharing your insights. And as usual, I would love to have you back on the show. Thank you It was great. Thanks for having me