 Welcome back everyone, the Cube live on the ground floor here in Chicago for KubeCon. CNCF CloudNative founded KubeCon and the Cube. Here I'm John Furrier with your host with Rob Streche, Savannah Peterson and introducing Joe Peterson tomorrow, our new industry analyst that's collabing with us. We've got a great guest here talking about VMware Tanzu. Cube alumni back from the old big day-to-days. Dan Besquat, director of developer relations at VMware Tanzu. Thanks for coming back on. Love the shirt. Thank you. Got to represent the Marvel Universe. It's been a long time. Welcome back to the Cube. It's been almost 10 years. Yes, as I said earlier, a little more gray now. Well, what Rob and I have been pontificating on the Cube and celebrating is the fact that big data promises are coming true now with AI. You're starting to see a lot more, those workloads having AI enabled or AI native. Certainly AI rappers are popular with open AI and these other models. So, super exciting time. I know Tanzu, you guys are in the middle of that wave too, the modernization of cloud. Give us the update on what's going on with you. Really starting to, I think we've taken a step back from developer, developer, developer and focusing on the application lifecycle. Develop, operate, optimize and really trying to enable that and still giving the developers something to make them happy and make their life easier. And let them code, but you're enabling another team to give that to them. What's the current state of the hybrid world now with containers and Kubernetes? What's your view of where it's at now? What's the current state of the union? I mean, it's a broad, broad universe. So, it's definitely an interesting time. I think from a developer perspective, it's probably a bit too much, right? That's so much to take in. I look at like the CNCF Kubernetes training for developers, it's a lot to it. I don't really believe the developers should have to know that much about what they're deploying on. They should know how to write their code, commit their code and move on. And to that effect, you're a contributor to backstage as well. And I think they've made some steps and there was some announcements with the Spotify code that they were giving to go from 70 steps down to three steps or something like that to get up and running. Do you see that getting even better to help give guardrails within, because I know you use that within the Tanzu architecture as well, right? Yeah, I definitely see the number of steps decreasing. So, one of the things we're looking at as we look at more of that next level up is that VMware Explorer in Vegas released the Tanzu application engine and this idea of application specific abstraction called spaces. And so it lets the platform engineers basically define governance, compliance and feature sets of a space. And then if an application developer wants to deploy, they know what the requirements they have, they can now deploy to a space and where that goes into your Kubernetes farm, it sort of depends on where those spaces live. What's your view of platform engineering? We hear a lot of DevOps versus platform engineering discussions. What are some of the things you're seeing in platform engineering that's jumping out at you in terms of what's critical about it and what areas need to be improved or filled gaps filled or developed? I mean, I think it's definitely a growing field. I mean, I came from the old pivotal company in VMware acquisition. So one of pivotal big things is growing that platform engineering team with Cloud Foundry and really seeing that take off in the Kubernetes world as well. So definitely see it as that it's sort of their job to ingest all of what we see around at this kind of conference, building the best platform for developers to move forward and letting them develop. I think some of the big pieces of it are right now it's security, security, security, right? So vulnerability scanning, S-bombs, just all of that security, all those security features. There's a lot of overhead in managing all that stuff and that seems to be the core theme, Rob. What's your take on the container management and the management side? Because if platform engineering is to enable developers, as you had pointed out earlier, okay, that developers develop and the platform engineering teams do their thing. Sounds like you were saying, right? So I mean, I simplified a little bit, but okay, what's the management side look like? How do you manage all this? Yeah, from a management standpoint, we've come out with something called Tanzu Hub. It's sort of this consolidation point from data from the developer platform or our cloud health costing analysis. So all this data can come in, Kubernetes management comes into that as well. So you have this central point where I can create Kubernetes clusters, create spaces that have costing information available immediately. So what's that costing application team? And you also, just remembering back to explore as we were there, you talked about, or there was the Tanzu Data Service as well. How is that helping organizations that are in a Kubernetes environment really move towards AI, I guess you could say? Yeah, so the data services are adding multiple features around the AI. So from a green pump perspective, that's using it as a vector database or they're adding definitely some learning, AI learning technology into it as well. I think the biggest piece of that is just sort of the infrastructure around the AI that we can help provide. The use cases for Edge come out a lot. It came out a lot at VMware Explorer. Tanzu has that aspect of it. What's the Edge look like now from a Tanzu perspective? What's the update there? From an Edge perspective, we're looking at a couple different ways. So you still sort of have that SAS management model, if that makes sense. We're also doing sort of a self-managed model for Kubernetes management at the Edge or deployment. So you sort of get to pick which everyone makes the most sense for you. You mentioned that costing piece of it. Is that coming from the CFO or is that more of technical costs? What's the cost equation look like? It's, in this case, it's cloud costs. It's not, it's financial dollars. Yes, financial, yes. Not technical debt. We're not getting anything. No, not technical debt. This application runs on these systems, these clusters. It's costing me $3,000 a week or whatever. The question I get a lot on theCUBE, Rob and I get this a lot from folks is, where are the areas that platform engineers need to pay extra attention to? Because it's, you know, you're enabling a lot of people to do work with developers. And if you miss a piece of it, or don't optimize for certain use cases, you could have a systematic failure or potential downtime. What are the areas that people should pay attention to the most in platform engineering? For me, I use the simple answer as it depends, but it depends on what your developers want and need. So I think for platform engineering, the most important thing is to talk to the developers, find out what they need, and start to bring those ideas back into your system and help build it out. And that's, a big piece of that is something like Backstage we mentioned earlier. So, we have the Tanzu Developer Portal, which is based on Backstage. And that basically lets all the data from the developers be surfaced in a single spot, right? So it's something they can use that's something platform engineering can provide. And kind of, I guess, springboarding off of platform engineering is the fact that with GenAI coming out and with all of these new tooling and everything else that needs to be there, what are some of the skill sets that you're seeing that are kind of changing or evolving as people get into more of these AI-based applications? Definitely seeing those data science type folks in the organization sort of moving a little more of that direction. From a platform engineering and developer space, it's probably more around just understanding all those technologies and what it means to deploy and train models and what types of systems and GPUs do I need to make this all work? Yeah, and you also have spring as well. I think it was announced was spring AI. And how does that fit into this entire? Yeah, so normally when we think of AI programming, it's a lot of Python, right? And Python has basically tools that allow you to write once and then run on different models. Spring AI brings that idea to the spring world. So now Java and spring developers can basically write to the spring API and then deploy different models on the back end and not have to rewrite their code. So really that spring value proposition. So real advances on that from the developer's perspective, the subtraction you guys are doing there. How's that going so far? What's the feedback been? Well, we just announced it to VMworld and Vegas and got great feedback there and we had a training session. It was just packed wall-to-wall. If one person got up, there's a person to fill that seat immediately. So it's been a lot of good feedback and a lot of good progress on the project since then. I really, AI does give the Java community a big boost in these new areas. A lot of glue layers kind of can, when you have these abstractions, you guys went to the application platform. It's interesting, some of these problems get solved. What are some of the key things that you guys are seeing with the platform you have that customers are resonating the most with? What's jumping out? What's jumping out at you and saying from a customer standpoint, what are they into when you look at that, the Tanzu application platform? I think the whole idea of a secure software supply chain, so the ability to plug in my CI-CD tools into this and then plug in the security tools I want to use. And I think the big thing with our model is, if something like a base image for your OS changes, the typical pipeline, I'm restarting the whole process. With our model, just that process, notices the change, kicks off, rebuilds the container, moves it to your container registry, and then you can redeploy it. So it's a model where each individual step can be kicked off based on some input. So time savings is a huge, huge benefit. Definitely. Yeah, and the other thing that comes up, you mentioned this earlier, I wanted to kind of bring it back Rob, to our earlier kickoff, we kicked off the show, keynote review, this idea of an end-to-end becomes a huge part of the spectrum of people thinking about the solution workload. Because at the end of the day, you want to just build and deploy your workload. Yes. A lot of other stuff going on, I'm under the covers. See a push. Yes. So now this end-to-end, how important is that end-to-end workload view and how can companies get their people trained more to think like that? What are you seeing for best practices or mindset shifts? I mean, I think it's very important because what we're seeing, and if you look at our state of Kubernetes report, it's like almost 100% of the companies saw value in Kubernetes, but then only 50% of the companies said, we have people that are trained to do this stuff. And that includes the developers, right? So if you're asking way too much of the developers, you're never going to get your code out the door. So the idea is we build this platform so that the developers can just code, just push their code. They don't have to understand any of the Kubernetes infrastructure plumbing. Makes their life a lot easier, right? If they want to get into that, fine, but most of the time they don't. So basically you're saying that the big blocker or the problem is that people just don't understand the infrastructure well enough. It's the complexity, right? We've never really asked the developer to understand the infrastructure like we are now. It's almost like we've gone backwards. Yeah. What way? By making them learn hardware? Yeah, we're exposing the infrastructure in gateways and all this stuff to the developer and they're normally just pushing their code, right? They've never had to understand this stuff before. Right, especially as they go towards microservices and things of that nature, and it would seem that, again, bringing kind of Spring and Tanzu together kind of tries to overcome that, right? Is that what you guys are aiming at? Is hey, you're going to go and build your MLAI here. Let me give you the tooling to go just do the coding. Yeah, I mean, we definitely have a huge install base of Spring, right? So as director of developer relations, so I have a team of developer advocates, and we're really sending that team out to major clients and letting them talk about Spring and the new ways to use things and AI, right? And developing that, basically developing that desire to bring, come to us for more information or even the products, so it's sort of a fun time. I'm doing a great job because I will tell the people watching here at VMware Explorer, we had theCUBE there. We weren't at your event with broadcasting, but we saw firsthand how packed it was. You guys ran the day before VMware Explorer, your Spring One show. Spring One. And it was GM packed, energy was high, people had a spring in their step, Rob, as we say, you know? It was the first year of combining it with Explorer, so it's interesting to see, there's two different crowds, right? It's a developer crowd versus operations crowd. So we weren't sure how that was going to work out, and it was beyond expectation. The cultural revolution of DevOps finally crossed the gas, I'm Rob. DevOps and platform engineering coming together. I talked to somebody out in the hallway during the conference, and there was a developer who said, hey, I talked to my, and went out to dinner with my operations team, and I've never really met them before. We're building relationships inside the companies that didn't exist. Awesome, well congratulations, big success, and again, platform engineering, we're huge fans of it. We see there's a huge opportunity to modernize the old IT, bring it in, and AI is a gift, and it's going to be an opportunity to help. Definitely. Set the table, and let developers do what they do. Develop, and the ones that want to get involved in the hardware, that's called DevOps, welcome to the party. Yes. The door's always open, right? You want to learn about the infrastructure. Definitely, especially with private AI, and we're seeing a lot of that on-prem. I mean, it definitely is. Yeah, I mean, that's AI sort of interesting, right, because there's a lot of these coding tools out there, but now you're starting to see companies a little worried about using some of those, like co-pilot and some of these tools. So that's what private AI is designed for that purpose. Let's bring it in, let's train it on our code, our documentation, so that when you need help developing something, it's coming from our historical developers. Yeah, we appreciate you coming on theCUBE and sharing. We're definitely be following more of what you're doing. We're totally interested in it. We're on the same page, so to speak. Last minute we have left. Put a plug in for what you're working on, key things in your job, Tanzu platform, and what's going on in Dev, Dev Relations. I think the biggest thing now is just getting people on board with that platform engineering on that train and understanding what it takes to build a platform that developers will love. So we're working on that, working with spring developers to make that a reality, and working to make our platform the best place to run spring, period. Dan Baskett, director of developer relations at VMware, Tanzu at VMware, great to have you on theCUBE. Get back with more coverage here at Chicago after this short break. I'm John Furrier with Rob Streche and Savannah Peterson and a lot more coming right after this break.