 Thanks so much for joining me for our drive-thru Detroit. I've done it a number of times now, but I think I'm starting to get the hang of it. But it's Bell Isle where we've been going and it's really, really pretty. So why don't you start, can you tell us a little bit about like what you do for Ford? Oh yeah, I'm like, thank you for having me first of all. So I have been along with the rest of the team, brought a team, we've been developing platforms generally, everything from Pivotal Cloud Foundry all the way back in 2016, to now Kubernetes, PCA, to GCP stuff, Terraform. Gotcha. Every data center platforms, so any hosting platforms we build, we operate, we manage. The big part of all of that has been enablement, right? How do we bring people to the platforms? Right. And those are the primary areas that I've been interested in focusing on. Gotcha. So what do you think you've been doing this for a little bit? What do you think is the biggest hurdle for someone coming into your environment? What do you feel like? I think the biggest hurdle for my point of view, and I think that could be, I think they will resonate very well. It's just a sheer amount of things people have to know. Yeah. I'm like previously we were talking about programming languages, and I'm like, what do you do? I'm like, everything has an API, everything has a CLI. Now, am I supposed to know the GCP APIs, or am I supposed to write my Java app that connects to a Cloud SQL instance? So the challenge has been is that, how do you steer those, I would say sometimes curiosity? Sometimes I don't have time for all of that. Right. Just get me, cut to the chase, give me this thing, and I will do, I think the change in mentality and the change in the adoption, things that people talk about, it has been very challenging. For example, for a long time, you just tell us what do you need. Right. Could be a spreadsheet, check, check, check, and then we'll turn around in a couple of months and here's your environment, tell us no, let us know if it's not working and invariably that it will not work. Right, there'll be something wrong. And then you go back, fix, and now what we've been doing in the last few years is that we are saying that here's the platforms, here's the complete self-service platforms. We are onboarding you and we will give you an opinionated stack to get started with it, whether it be pipelines or Terraform or what have you. But then you assemble all of these things together and we are expecting that you are in the SRE of the team to manage all of those things. That incredible mind shift in a very short time, it's basically like a night and day. Yeah, I still remember on one of my gigs, way back in the day, having to give the data center a 12-week lead time on being able to get a server because they needed to rack it. Exactly. Stuff like that. Just make sure it's connected to the right ethernet board. Right, which invariably it wouldn't be. But it's so ridiculous now that I can, it's funny, just last night, actually I spun up a machine so that some kind of external instructors who were teaching at BU, but they wanted to get their students onto Unix and Linux, right? And I had kind of forgotten about it and I was like, oh nuts. And so last night, like actually I built the machine and then I realized that I didn't like the way it was built. So I trashed the machine and then I built another one and then I basically logged in, wrote a script for user creation and then generated a bunch of users and then sent them an email with the details and I think it took less than an hour. Yep. And it's just like, this scale is just so different. And the other side of it is also very scary for a lot of folks, right? Oh my God, what are you doing? I'm like, one of the things that tell people we are in the car right now is think about you're in a car, up until right now you're sitting in a car with blocked out windows. Just crack it a little. Yeah, right, right. Just crack it a little, let the light come through and you will see that you have a controls panel in front of you. You are in charge. You could go as fast or as slow. You could be as cavalier about it. You can experiment with it. You can burn it down. You don't have to submit a ticket to burn it down. You can scale it up when you want it. And that act of that, you have so much control is mind boggling. It's very difficult for some people to just think that you can do that. Right, yeah, the allowance factor. Yeah, that's totally true. And I think, especially when you talk about kind of software developers who are kind of coming from a more strict engineering background, one of the things I talk about with software development a lot is like, when you're building a building, or a car, right, and you're an engineer, like it doesn't stay together if you don't do it right. Like the building falls down. Whereas like when you do software, you basically have endless rewrites. You have endless re-dos. And invariably you'll need them. But also it's kind of, it's a very quite different mindset than you're trained, I think, when you come up through college, or whatever, you're taught that it's like engineering. And you're like, and I always say, I think it's much more like writing. You rewrite, rewrite, rewrite, and then eventually you get to something that's good. And there's rules you have to follow, but if you're good enough, like EE Cummings, for example, you can ignore a certain rule, like capital letters, right? Exactly. You have to be really good to be able to do that. Yeah. The biggest thing is that I can have a broader set of guardrails, right? Don't do something that, hey, don't put my service on the public internet as an example, right? But within that bounds, as long as we can bake those things into the bounds of the platforms and the tooling that we are providing, then it provides the so-called safe space to experiment and express yourself in terms of what you can do. Well, and I think what's been really cool for kind of the opposite end of those types of people, right? It's been that with, you know, kind of even from the advent of like Paz, was like, now if I have an idea, I can just go try it absolutely without getting anybody else involved without really spending any material money. And just be like, hey, you know, I was annoyed by this thing at Ford, you know, and so I sat down and I was like, you know, maybe we could make this thing work this way and you can just try it. Absolutely. And then, you know, but when you tried it, you also built something is actually not production ready, but at least in the nearby space, right? Exactly. Instead of just a piece of crap that you have to throw out and build again. Correct. And that's such a mindset change. And it's so nice, at least, you know, for somebody like me, I would come up with ideas all the time and I'm like, oh, let me go experiment with this thing. It's so empowering if you think about it. And I'm like, it's a radical culture shift in the way the organizations are put together and the way organizations work. And that's been some incredible journey. We have been on past five years through that and like some of it work really, really well. Some of it taught us some lessons, like normally we used to hand hold people or do office hours one on one. Quickly realize that, you know, we can't do that for 10,000 developers. Right, right. Yes. Not enough hours in the day just to do that. So we have to find alternate means of actually empowering and educating and inspiring a thing and that's more of the thing. Well, I think one of the other huge problems is also the re-educating. Like as in, you know, things change a lot. And so not only do you have to kind of educate them in the first place, you also have to make sure they understand how to continue to learn, right? And where the next piece of change is gonna come from. And what's going on over here? This is interesting, maybe a race or something, I don't know. Police training. So kind of the cones? Oh yeah, maybe. Police training looks like defensive driving or something. Yeah, that's kind of cool. I've seen a number of neat things on this little tour. Yeah, I saw a soccer game. I saw like a race that was about to take place. I've had geese and seagulls and squirrels trying to hide in front of me. All right, on Lake Michigan, so yeah. Yeah, but this is a really pretty little island. Detroit River, it's going into the Lake Michigan. Yeah, it's quite nice. So what have you been working on lately? What's the kind of thing that you're, you know, kind of hoping to see you lay in sometime in the near future? I think the biggest one that I'm working, at least rest of the team, I have like a few things. How can we make what I've been calling and what generally people have been referring to as inner loops, right? So we all have been remote. And how do you provide uniform bookstations or developer spaces? Oh yeah, yeah. Pre-plumbed, so that you could just go in there and do what you want to do, but with security and provides a variety of catalog of things, and things that we are looking like, Eclipse Shea, for example, right? Yep, I'm a big fan. Yeah, so that's awesome. And then there are other technologies like Cloud vendors are doing, like Git Pod comes to mind, for example, right? So those are the kinds of things that I'm looking, we are looking into. How can we make that easy? The other thing is like, what other abstractions can we build on top of, let's say Cloud, both internal and public, and say that, hey, how do we actually do that? Try more abstract patterns that are easier for developers to get started. I think those are the major areas that we are trying to focus, primarily seeing that, hey, spending incredible amount of time getting people up to speed. Right, right. So the question is, can we actually reduce that barrier to the entry, right? That has been a biggest challenge and quite a bit of time has been, at least R&D time, we are being spent on doing that. Right, yeah, I think it's funny when you kind of roll out some of those development platforms, people don't quite realize how hard they are to build, because not only are you trying to build like an automated rolloutable platform, but you have to like pick all the tools, right? And that means you got to make a whole lot of bets and it's nerve-wracking, because especially, I talk about this with open source, with open source it's so hard to know am I betting on the right horse, you know? Yes. And if you're doing it on behalf of, would you say, some 10,000 developers or whatever? Some of it, yeah. That's real stakes, you know? Exactly. And so, yeah, I think it's very, very difficult. I've worked on a number of them over the years. Yeah. And it's super hard. The interesting thing becomes is the complexity of so many things. You have to actually coax and cajole so many things together. Yeah, yeah. You need to have, you know, what about security? I'm like, are we going to be everybody's route, everybody? Right, right. I don't know. I'm like, but I think in order to do all of those things, sometimes I feel that either, A, we are encumbering developers to be thinking a lot more things about it, or are we saying that we're going to provide an ecosystem where things are easier to get started with. I think that has been a huge focus that we have been trying to actually make it easier for people to get started. And I think it works in best interest of everybody, right? So, people are much more happy or satisfied. Things roll out much faster. You don't have to wait for a ticket and, you know, wiggly your thumbs expecting that somebody is going to pull off a magic somewhere. Right, right. Yeah, but it's a really hard line to draw of the where you want to be prescriptive, right? Versus letting people kind of like evolve their best practices. Exactly, exactly. Because we can't be very prescriptive in the sense that you can't, thou shall not do this thing. Right, so. Right, that usually does not work well with developers. But yeah, I totally understand. So, let's talk a little bit about Kube by example. So, I know that you've been using it a lot. A lot. What is it, like, what does it do for you? So, not so much like what's its pitch, but as much as like, why do you find it useful with your organization? I think the biggest part that Kube, for example, got, you know, the aha moment, there was like, hey, Kubernetes is huge. It's just huge, I'm like. And I can't take somebody who has just been doing some Java Spring Boot development and say, hey, you figure it out. The best part of Kube by example is that it boils down to certain basic primitives that you have to know. Sitting in a car, you need to know where the accelerated pedal is, how the steering wheel moves, you know, how do you turn on wipers. But those are the essential human, human machine interface kind of thing, right? So what Kube by example does is that there are other things, maybe there is an engine, there is a carburetor, injector, you don't have to know about it. So what Kube by example does is it boils those things down to few primitives that you understand those things. And then what we have done is that we have taken that to a next level. We said, hey, Kube by example, gives you this nice bite-sized videos and concepts. But how do you turn it into practice? So let's take an yaml example. Let's annotate it and think that what is this line doing? This is how you do a toleration. This is how you do a label on it. And this is how it actually works. And the idea is to take a simple example. In this case, we have taken part in, for example, that the CNCF uses and build whole this Kubernetes ecosystem around it, right? How do your HBS works, PDB works, network security policy works. So that way you can get the little bit theory and small bite-sized videos from Kube by example. And then you can immediately turn around and translate that thing to a working thing and get a feel about it. I think that gives you the sense of immediate gratification. You are in the age of wherein everything is do-dash or everything, there is an app for everything. But it kind of eliminates some of that uncertainty that you don't have to go figure out something like that. You are not spending orders of magnitude of time doing Google search or on StackTrace, StackOverflow and say that, hey, how do I do this thing? Right, right. So I think that has been a great learning tool, at least learning methodology that has been very successful, right? So learning by example is something that we found, at least internally, is it has made things easier. The amount of pushback that you get is easier. It's a lot less. And the best part, if somebody asks a question, so get repository, you can select those lines. Here is an example. Right above it are a couple of blog posts sort of referring back to the concepts that are called in Kube by example. So it becomes a kind of a loop wherein you kind of read it, do it, then read a little bit more, a little bit more. So it becomes an iterative approach to learning. Yeah, I mean, I think it's really important. One of the things I think that developers are kind of colossal or sometimes too often is the kind of theoretical part. And so, but it's very difficult to get them to go read a theory book, right? Yeah, it's too much. So if you can give them to them in pieces, I think it really does, it does help a lot. We were talking about programming languages before. It's like, it makes a big difference if you know the philosophy behind a programming language about how you accomplish things in that programming language. And I think a lot of engineers don't recognize that it's important. And yeah, so I was just working on the new learning path for TecTone. And so the first thing I do is explain CI, and CD again. Exactly. You know, and that, oh yeah, we have two of the same acronym that mean two different things. But all three of them will be used interchangeably to mean all three things. So if you're getting a little bit of that context, I think is also helpful because then you know how people are gonna talk about it and what they mean. But then you also wanna actually just go build a trigger in TecTone. Exactly. Yeah, that's really cool. Sometimes if you think about it, it goes back to, you know, ergonomics or things people talk about, man-machine interfaces. It's all the same if you think about it. I'm like, these are all the mechanisms that we have provided and we have abstracted them behind concepts and terms. Right. But if we go back and you know, think about how we learn language. Right. If we can actually connect with Tec in that terms. It becomes much more easy to eternalize those concepts. Right, right, yeah. Yeah, I mean, you know, I teach an introduction to data science classes. You know, I kind of run into this a lot. And you know, and so what we do in that class is kind of give this overview. It's almost like an overview of data science. But the idea is that student who walks in has had no programming experience whatsoever. Exactly. And they learn Python as a way to accomplish some other goal. Correct. And I think it really does help them kind of wrap their head around it. You know, even the ones that believe that they can't program, you know, or don't know how computer works and stuff. And you're just like, oh no, just do this, you know, a little bit of code here and it'll solve this problem over there. Yeah, I mean, particularly it's very satisfying. Yeah. Like when you see something, you did something, you see something running. Right. And no matter how stupid or how simple that might seem, but it is immensely gratifying saying that yes, look what a cool thing I did. Right, right. Yeah, 100%. I was talking about it with that exact thing with Josh Berkus yesterday, I guess, or sometime before now. And we were both saying how we have a couple of very minor PRs that were accepted into really big projects. And he has one in HA proxy. And we're both like, you know, like we didn't really do very much, but at the same time, it's amazing, right? Like you feel so good about it that you're like a part of this, you know, thing. And, you know, and then whenever you build something, it's, you know, I think, I don't, I think people don't realize like, you know, programmers are usually in it for the, the gratification of like, hey, I built that thing. And it works, you know? Yep. And it's even upright most of the time. Yes. That's the beauty, I'm like, I think, and I think the whole goal we've been trying to tell people and trying to change the culture within the organization is around that concept thing that, hey, you are in control if you're destined. Right, right. So, but you are not alone in it. Yeah, yeah. We want to help with you, but the important thing is that only investments we are asking from you is curiosity and real time. Right, right. So if you're curious, and if you are willing to invest in time, we are willing to meet and pair and partner with you to get you there. Right, right. And we will learn. We all will learn around. I'm like, no, I'm not an expert at everything and all of a sudden. Yeah, it's like every time, when I run teams in the past is one of the things I make every new member to the team do is add, edit or update the onboarding guide for the team. It's like because the person who's best available to recognize where the flaws are in the onboarding guide, the guy who's using it the most recently, right? And but I think it really does help to instill in the team, like a part of the community, because now you've all committed to the onboarding component. Exactly. But I think it's kind of in the same way. It's like, if you can kind of take the first step, I can kind of help you, you know, continue. But you got to help me take the, you've got to take kind of the first step. Exactly. And, you know, and then sometimes there's steps along the way that you also have to kind of commit to. But it does, I think it really does instill, you know, more of a community around the thing if you kind of are electing to be engaged. Exactly. It's kind of a loan, right? Yeah. So, you know, for a loan, you had to put some investments down. There's some collateral. So that means that you are engaging in it. Right, right. And there is a reason for you to engage in it. And the collateral that you're putting in your, in this case, is time and effort. Right, right. And what you're getting back is this immense gratifying sense that, hey, you could do something. Right, right. Yeah, totally. But we're also starting a bit with Becky was saying that, you know, you are all doing a lot of like active hiring of self-prodevelopers. Absolutely. We are hiring. And I think come see us in our booths if you are in the, you know, at KubeCon. Yeah. All the same positions are at careers.for.com as well. Yeah. And a lot of things that we are doing is all so much based on technology. Yeah. Well, I was kind of saying is that, like, I think, you know, you're a lot of, you know, developers may not think about the fact that how much software is going into, like, in the actual car these days. I think it's getting a little more obvious. But also, in how much software there is to, like, make the car. Oh, yeah. And there's a lot of really neat software there. It's all software from your drawing boards. Think about it. In America, we don't even make clay models anymore. Oh. Everything is, like, digitized and 3D rendered. So unless the final version comes out, people make clay models out of it, right? So everything is digital. Everything is data. Everything gets to evolve fast. And that's the beauty of software. Right. Yeah, every company is becoming a software company. Every company, I'm like, the other thing that I will say is that we live in the age of gadgets. Typically, we carry those gadgets in our pockets. This happens to be a gadget you sit in. Right, right, right. Exactly, right. So hopefully there will be a day where you can hail it. Right, yeah. Right, it'll go drop us something like, and it'll come get us, and it'll go do its thing. Hopefully the day is soon, and all of the day can only be realized if we have great depths. Right, exactly. No, I completely agree. I am a big fan of the, like, I'm waiting for, like, William Gibson's world. It's like, and what I was hoping I was contributing at Red Hat actually was another little step towards that. Because I think it's where I want to see, without the dystopia, of course, but as far as the technology and the, like, just every, you know, it's just kind of, it's just around. You know, we don't have to, we don't have to open a computer. It's just there. It's like pervasive computing. It just happens, right, so. Right, which I really think is very cool. All right, well, thanks so much for coming. We really appreciate it. And, you know, we'll be doing a panel together later. But, yeah, let's call it for there. Yeah, thank you. Thanks, thanks for having, and thanks for the opportunity. I'm, like, looking forward to be engaging with all of you online. Right, right.