Techstrong Gang – December 17, 2024

Hey everyone. Oh my. We've got an AI skills gap.

We're fighting on the Chinese. Next. You're watching Text on Game.

Good morning everyone. Happy Tuesday. It's the week before Christmas.

It's almost the end of the year. We should all be jolly, I would hope. I hope our gang members at Jolly today will find out.

It's Alan Shimmel for Textron, and we've got a great tech, strong gang for you. As usual, we have our usual kinda Monday gang. It's where we record Monday for Tuesday.

It's our Tuesday usual lineup. Let me introduce you to them. If you're not familiar with our Tuesday gang lineup, first of all, joining us.

He's back home from London and wherever else he's been in the world lately, but he's home high in Colorado. I don't know, but probably. Anyway, fu VP analyst, Mitch Ashley.

Hey, Mitch, how are you? It's a Rocky Mountain High. Yes, I'm back in Colorado.

You're a Rocky Mountain High. Alright. Good day everybody.

Um, joining us from the outskirts of Cleveland, Ohio. Cleveland Rocks. You know what?

He's pretty wound up today. Go easy on him. It's our friend and founder of, uh, tech Field Day, Steven Foskett.

Hey, Steven. How are you? It's rainbows and unicorns here.

Rainbows and unicorns. Boys. Just remember even with rainbows and unicorn unicorns, someone has to clean that poop so they get skittled.

That's me. That's me. Yeah.

That's Why you make Skittles. Yeah. Welcome.

Um, joining us also, he's back home after gallivanting around trying to find the next Yankee phenom, uh, our chief content officer, Mike Ard. Hey, Mike. How are you?

I'm good. Hey, If you got a ball player who's over 30 years old and is on the end of his career, by all means, send him to us. Where the place for him?

No doubt about it. We're back. Okay.

We'll over overpay for them too. And we, and we'll overpay for them. Exactly.

Some things never change. But that being said, bringing array of sunshine and, and decorum to our gang members today. Joining us here in studio with me, it's our sustainability and all things green editor and analyst, Bonnie Schneider.

Hi, Bonnie. How are you? I'm doing great.

Thank Good to be here. Ellen, It's good to have you here with this gang out that we, we need you. Um, anyway, folks, we've got a busy day today.

Let's jump right into things. So are we fighting an AI skills gap? This, this just sounds like this is at central casting tech story 34, right?

First, it's, it's all the rage. It's this, it's that. And now we have a skills gap and there's too many jobs, not enough people with training.

We need more teachers. Damn it. Mike, what's the story?

So, to your point, that's kinda exactly it. And everybody and his brother has a survey out saying, we need more training and skills. And I'm scratching my head a little bit about this because I feel like a big part of this issue is just that the software and the process is cumbersome and people don't quite take to it.

'cause it's not naturally part of my workflow yet. It doesn't sit inside my app where I can easily call it. And it's not like kind of embedded in everything.

It feels like a little disjointed and people don't know what to train for, right? Earlier this year everybody was saying, I want to be a prompt engineer. Now everybody's like, oh, well, we're gonna have AI agents that take care of that task.

But now I don't know how to orchestrate the AI agents and I don't even know how to plan the workflow for that. Nevermind, orchestrate it. And so I feel like, you know, I don't want to beat people up for lack of training, but I kind of want to say to the vendor community, Hey, you know, we made a lot of promises here, but maybe we're still two steps short of this thing in terms of what it's gonna take for mass adoption.

But Alan, I know that you're a big fan of ai, use it all the time. And yet we have conversations all the time about, gee, how come more people aren't using this stuff? Yeah.

Well, let me give you a cautionary tale. It was gonna be a very simple thing. Move from one email.

Don't No, no. That's a different tale, that's a different show, different tale. But let me, let me, um, let me say this first.

Got involved in the web around 19 96, 96, maybe 97. No, it was 96, 95, 96. Netscape was in beta.

And I, I was not a coder. I'm not like Mitchell Mitchell went to school for computer stuff and he was a computer dude. I went to law school and, you know, I had punch cards.

The only computer class I took was at St. John's University in college already. And it was punch cards.

So I didn't really know much. The web comes out and it rocks. My world rocks me.

I'm like, wow, this is the greatest thing ever. 0 is out. And I'm like, what are all these computer nerds getting worried about this stuff is easy.

I learned HTML really quick. You wanna put something in bold, you know, bracket B, close bracket word high slash bbb. I made it bold.

I could do anything. I'm a coder. And I proceeded to build a bunch of websites for friends and family and whoever we would want one back fed as early.

And we started charging for it. And I'm like, this is a fantastic thing. Anybody, you know, I know the magic formula.

I'm a, I'm a software engineer here at Coder. 0 and JavaScript and JS Script and Python and PHP and Java itself. That one, you know, you and, and I was, I was quickly over my head and realized, let the coders do that coding, I I should do the business piece of things.

It was a great experience being a coder for those three short weeks. But, uh, it was more than three weeks. But it was a great experience.

But it's, it's the way of the world, right? AI came out and people said, be a prompt engineer. 0.

Quite frankly, any monkey can probably do prompt engineering or HTML one oh coding. You didn't have to be a software where it's, we are going to, as this matures and it's very immature, it's going to be time for the real skills with a z kind of people to come in here and, and run the show as, as, as is normal in technology. I think this is the, this is the usual path, right?

It gets more complex and people need a, now there are friends of mine who started out as HTML one coders and, and learn CSS and learn scripting and, you know, can do stuff like that and, and, and are still coders today. They're software people today. So there'll be some of those in the AI space.

But this is what you expect. To me, this is, this is, come on, captain obvious stuff. And we're in an immature space where it's advancing rapidly.

And, and I don't know exactly what the skillset's gonna be to be an AI engineer next year or three years from now, but I'm pretty sure being a, being able to type what you wanted to say in a prompt doesn't, isn't gonna qualify you as such. Yeah, that's, that's a real good point, Alan, because truly the somebody who had phenomenal AI skills this time last year, do you think any of those skills are relevant this time now? I mean the, the, this, the whole, the whole field is advancing so quickly that the only way to have really usable AI skills is not only to have the skills to start with, but it's to continually stay on top of things.

That's awfully difficult to do when things are advancing quite so quickly as they are right now. You know, what are companies looking for? Are they really looking for prompt engineers, like you said?

No, I don't think they are. I think they're looking now for people who can put together inferencing applications and, uh, you know, connect with vector databases and, and that's completely different. Entirely different.

So I'm not surprised that there's a huge skills gap. Well, like, like security. You know, Alan, when you say we don't have enough security people, oh, what kind of security people?

It's a very wide broad, you know, topic. And so is ai, right? Are you talking about model training?

What kind of model training is an expert system model, uh, uh, machine learning MA model? Is it a gen AI model? What kind of model is it?

And what skills does it take to do that? I I think the, the other thing is the devil's in the details. Working with AI is not like writing procedural language.

Uh, and you know, whether it's Python or anything else, it's, it's very much of a kind of science iterative with, with, yes, it's got algorithms, but there's also a lot of data manipulation. And you're constantly going through a cycle to train a model to do something or write algorithms that will get the right kind of output. So it isn't like, you know, if you add column A and column B, that will give the balance of your checking account.

No, this is a, you've gotta tune this logic to get answers that will be useful. And I'm not even talking about generative ai, just meaning machine learning. So it, it's a different paradigm.

I mean, I remember, I mean, I got into AI right outta college and I kinda had to stop and just put aside the normal development languages I'd learned and say, now what is this? 'cause it's very different. It's not the same thing.

Mm-hmm. I think it's gonna be the, ultimately I need people who think about workflow. And part of the issue in my mind is that so much of what we all do is muscle memory.

That we forgot what the process is underneath it. And then when we get into the process, we discover that there are more exceptions than there are rules. And it's hard to kind of, or put something together that the AI agent's gonna fit in, right?

Because ultimately it needs to sit in alongside humans that are hopefully supervising this stuff. But I, I, I don't think we have the thought process in the business side, nevermind the tech side yet to actually kind of wrap our heads around what it is we need to do here. And I think a lot of people are stuck as a result.

You know what I love though, honestly, the, the almost predictability of it, right? That, that, that crossing the chasm model never disappoints. It, it, it remains true no matter what the next thing is.

It remains true. And another thing about this, you know what, this is gonna set off a whole bunch of AI certifications and AI training and skill sets and, you know, we, we, we've seen it. It's just, it's that same pattern that is just, it repeats, it repeats, it repeats.

And every time someone sits here and says, oh, this, this, this, we have a skills gap. Oh, this looks like something new. It's not, we had it with security.

Like Mitchell said, we had it with web programmers, we've had it with database people. We, we've, I mean name it network folks. This is what drives it.

Right now. You're going to, there will be the rise of some AI certification cert that's gonna be the be all end all for a little while at least. And everyone's gonna rush.

Someone's gonna make a lot of money coming up with the stack of ai. That's What we need is the lamps Stack of ai. But you, I'm, I'm not sure cics, uh, the C-I-S-S-P of ai.

Mm-hmm. I will, I have no doubt there will be certifications and people pay big dollars to get 'em. But I'm not a hundred percent convinced they're gonna even be necessary.

I think it's more about, um, the way you think about something and just the, the, the creative mindset that goes with that. And I'm not, I'm not sure that if I hire somebody with a certification in ai, that they have the right mindset to begin with. Is this, so what, they Took a test predictability of it.

I could say the same thing about security, DevOps, everything else. Mm-hmm. But at the end of the, God bless you, Steven, at the end of the day, at the end of the day, like that old commercial says, you gotta show me the sheep skin.

Right. Do you or is that dating myself? Well, I think we also have to let the, um, we have to let this lamps stack em emerge and evolve.

Yeah. I mean, you know, you can't train people on something that isn't even set in stone or even not even set in stone, set in mud. I mean, there's just, you know, before somebody can have a useful AI certification, they have to, there has to be a useful, uh, standard, like you said, the, the lamps stack of AI there, there're we're getting there.

You know, if somebody watching this saying, man, what should I do? Um, you know, hop on hugging face, start getting involved, you know, see what people are doing, see what the predominant models are, the predominant deployment methodologies are, you know, start getting involved. But it's just too early to say, this is the AI stack.

There is no AI stack, You know, I'm sorry again. No, I was just gonna say that I think depends on who you ask and the audience and which you're, you're talking to because you have an IT team. They're gonna be a knowledgeable about ai.

But let's say it's a smaller company and the IT department is just a couple people and they're deciding who they're gonna bring in. If this person says, well, I have a certification in ai, even though, you know, it's a, it's a moving target. I think that the people that don't know anything about it are gonna be like, wow, that's great.

You know, we don't know anything about it. I just think that we're at a time now where even a little bit can impress people, Um, without getting ahead of ourselves. Trick.

Mm-hmm. The next segment, your guests said exactly that, you know, this is early years and, and there's an opportunity. So I'm reminded of when I, I was one of the co-founders of the DevOps Institute along with, uh, Lisa Schwartz and Jane Grohl.

And, um, and look, we got it when we first started the DevOps, uh, institute, I was picking arrows outta my back like a porcupine picks quills, right? The people were livid, including my, I made some great friends as a real result. John Willis, right?

He hated us. Patrick dubois. He said, cry me a river, right?

When we did this. And, and it was just for what you guys said too early. It's too early.

We don't have best practices yet. And what we said is no, we may not have best practices yet, but there are emerging practices. And what we wanna do is help shape those emerging practices into best practices and make sure that they are being somehow codified and taught so that the next generation of DevOps people will have that as a foundation.

And what they go from there will depend on how the market evolves and how the the discipline evolves. Evolves. I think it's the same thing about ai.

The tech moves so fast, and especially AI is moving so fast. If you wanna sit here and say, well, I'm not gonna do anything until it, it's set in stone, or even mud, as Steven said, you're gonna miss that boat. You gotta jump on that moving train.

And the faster it moves, the harder it is to jump on. So the earlier you jump on, the better off you'll be. Make two sense.

I think there's fun. There's fundamental skills too. I mean, again, we're, we're using a broad brush to say it's too early.

We've been doing machine learning for a long time. You know, there's standard, you know, very, very commonly used pie, torch, psych kit, all kinds of stuff. Packages that we use.

And the common denominator is statistical analysis, right? You need some data skills to use those. So I think you can work on base skills.

'cause guess what? Model training is the same thing, or not the same thing, but requires some of the same skills for generative ai, right? You are training data going into a model, or you're, you're front ending a model with data through whatever mechanism rag might be using.

So it's a lot of data manipulation, data analytics skills that you can benefit from. So, you know, I, I understand what you're saying, Bonnie, about, you know, in, in the land of the blind, the one I person is king, right? It's like I have a AI certification, great.

You come over here, you own a book, good, come on over. Um, you, there's, there's more to it. And you don't have to jump into the deep end of training generative AI models.

Matter of fact, don't do that. Most people don't wanna spend their time doing it. That's not where you should be investing your time.

Um, there's a lot of areas where you can do work, do do useful work today. And there are best practices. There's tons of training on.

I just, my favorite is PyTorch. 'cause it's just so heavily used. It, it is everywhere.

TensorFlow. I mean all, all these things have been around for a while. Let me end this segment 'cause we're already over with, with this thing though, there's a difference between someone who wants to be a full-time AI engineer, for lack of a better word, or a full-time AI only person, versus someone who wants to be, let's say, a marketer who leverages AI to be a better marketer or a coder, right?

A coder who leverages AI to do better code a a ops person or, or what have you, uh, to use. You know, so I don't think you need to be the be all end all, you know, kick butt AI pro to leverage AI and what you do in your job now. And that's really the power of ai.

It's not just the people who are gonna be the AI gurus, it's how you use AI in your job to be a better you. And I, I think that's the important thing to, you know, the, the, the promise of ai. It's a nice way to Answer, a nice way to end that.

We're gonna take a break here on Textron Gang, and we're gonna talk next. Bonnie actually has a, a interview coming up with women in ai. You're watching Textron Gang Modernize your business to fuel innovation and elevate customer experiences with the builder community.

Hub AWS and its partner network provide essential tools for transforming applications and infrastructure to fully leverage the cloud. Discover free trials, in-depth demos and essential resources to empower DevOps engineers and developers to deliver value faster and more reliably. Visit the builder community hub to learn more.

Welcome back to the tech strong gang. We were just talking about AI and as the only woman at the moment on this panel who loves what ai, uh, this is the perfect time to talk about women overall. Now, when we're looking at women in ai, there's a lot of disparity, depending if you're looking at the US or or globally.

I spoke to a woman who really understands where we are with women in AI and where we need to be. Her name is Melissa Steyers, and she's the head of global growth for MIA ai. And here's a little preview of the interview I did with her.

And we are a global AI academy that equips forward-thinking companies and non-technical talent with AI education. We have a really big mission, which I love sharing. It, is to empower one of billion women with AI skills by 2030.

Now we train both men and women. We love all, um, and we are considered one of the leading voices for upskilling women in AI and driving inclusive, innovative global global training around the world. That's incredible.

So where do we stand now? What is the representation for women in ai? I I in the US I guessing globally.

Oh my goodness. So this is, this is a significant challenge. Globally, women make up about 22% of AI professions according to uno.

And this underscores this really urgent need for action to ensure that inclusive innovation and inclusivity is there for, as we are designing these technologies that are shaping our future. The, the disparity is even more pronounced when we consider that by 2030, the World Economic Forum projects that they're, that 85% of the jobs are gonna be here. They haven't been created yet.

Of those jobs, 80% are going to be women's roles. So it is critical to skill and re-skill women in AI and emerging tech with just everything that's happening. We are going to be an AI driven world.

So what are the barriers for women? Why are more women not embracing ai? I know women, I love ai, but there must be some, uh, challenges and barriers and what are they and what are you doing to kind of break through them?

The barriers are, are some of the barriers that women have faced for decades and century. They're multifaceted. They're, they range from, of course, the lack of mentorship and representation.

And then there's of course, the systemic biases that hiring and the promotion processes. There's, there's just a range. And, and we really need to be intentional in addressing the, and creating strategy, rating, scholarships for AI learning that's tailored to women.

That is why m um, is here. It's to help women, specifically men too, but women specifically not get left behind. Um, because we have, just in general, we have stayed away from stem, we have stayed away from technology.

It's just that that's always been the case. Representation matters now more than ever. It matters for those building the technology, and it matters for those using need the technology.

So the stats are still not awesome. You mentioned you're in different countries, I imagine women in, in different countries face different barriers and challenges in ai. I, I've been following your travels.

You literally go all over the world. Can you share some of those experiences and what it's like? I had the honor of being a part of the NDPs only Global, global Leadership Academy for women over in Doha recently.

And it was amazing. I think there was over 22 countries represented these women from Parliament and Human Wright. And then we got to talk about the skills of the future with them.

But you know, in some regions there's, there's limited access to education and there's societal expectations that discourage women from pursuing careers in tech or, or pursuing technology. Now, conversely, there are incredible countries that are investing in STEM and tech programs for women like Rwanda, where literally 50% of STEM students are women. And this is creating a wave of change.

And honestly, the most important thing here is gonna be global collaboration. By sharing resources and success story, we can inspire more women to see AI as a pathway to empowerment. We're coming to the end of this year and into 2025.

Where do you see this movement going for women for AI and sustainability? I am positive and I am optimistic that, that there is so much work to be done for good. I know that there are, there are bad actors.

I know that there are a lot of challenges. Um, but I am optimistic that AI can be a tool and a helper for good in all of these areas. If we get the right training.

And if we have conversations like, like ViiV, I think that the more diverse voices that we have on a global scale, the better AI can be for humanity. But overall, I, I am hopeful that AI is not, it's not gonna replace your job. I think it's going to be a tool to help you with your job.

It's gonna be a tool to help you succeed better. And hopefully my dream and and what we talk about at MIA is that it it's gonna allow, it's gonna take some of these mundane tasks off your plate to allow you to innovate, to allow you to strategize, to allow you to really, really envision things that we, that normally we can't do when we don't have the, the, the freedom to think. I'm hopeful that our world is gonna be more sustainable.

And I, I hope we're more loving and compassionate through these tools and through these kind of conversations. What would you advise for women that are thinking about using ai, but not quite sure? Oh my goodness.

I would say, first of all, go for it. Do not be afraid. AI is, is a tool.

It's a friend. It's not the scary monster in the closet. Start anywhere, start now.

The two takeaways I would say is one, the AI train has left, has left it, it's on the tracks and it's not coming back. Don't be brain and then just keep, like, keep a continuous learning mindset and the sky is the limit. So I would say go for it.

And if Mia can serve it, anyway, let us support you on this. We need go lock hands with a lot of other organization and companies around the world, and we will do it. I believe you, you have such great enthusiasm.

It's so great to have you on. Melissa Steyers, head of Global Growth and expansion for MIA ai. Um, check her out on LinkedIn and of course, check out Mia.

I think they're gonna be doing some really Exciting things next Year. Thank you, Melissa. Thank you so much.

I appreciate being here. So Melissa has some huge goals, um, with what they're trying to accomplish with MIA ai, but they're off to a great start. Uh, they're partnering with all sorts of corporations and organizations to get the message out that, as, as Melissa said in the interview, no, there's no dumb questions.

Anybody that has any interest in learning about ai, whether it's for productivity or for a job search or even on the job, um, MIA AI can help women at all stages and, and really just make AI accessible and more represented by women. So, uh, it's kind of an exciting movement. I had the chance to meet, um, Melissa at the Sustainable IT Award.

She was actually one of the people who won an award. So, um, very, very interesting to see what's gonna happen in 2025, because they do have some big goals. Fantastic.

I look, I look, I wish we would get to the point, especially in technology where we don't have to specifically call out women in AI or African Americans in AI or people from the, uh, you know, uh, LGBQ community in AI or any, we're all just humans in ai. Mm-hmm. But unfortunately, we don't live in that greater world.

And, and so in the meantime, I think it is important that we see people like Melissa and her story and what they're doing as well as others, right? Um, it it's an important, uh, thing that we need to be doing. Guys, I don't, you know, we've all, again, all been through this before ai, right?

Insecurity, Mitchell, you and I have fought these wars for years. Um, I'm, I'm gonna weigh into the land of gross over a simplification here. So, you know, don't kill me for this, but, um, I think there's a difference when it comes to AI in terms of being a techie hardcore programming.

And we have all the standard issues with women in that space that we will continue to have. But I think this whole area becomes more accessible to women as we shift to, as we said in the last segment. Um, as we focus more on the processes and the outcomes, I think we'll see more women involved in ai, um, in a lot of cases.

They just have a better mindset for that than a lot of guys. That's my gross oversimplification. But, um, I think that, um, this space is gonna be more inviting to women in general.

And I think, you know, we'll be pleasantly surprised how much progress we might make over the next couple of years. Steven, I with that you agree? Steven?

I know you have views on this. Yeah, well, well, first off, um, any field that doesn't, uh, proactively include, uh, women or minorities of any type in it is really missing out. Um, if we exclude women from the AI community, well then we've excluded 51% of the workforce globally.

That's not gonna be good for anyone. So we need to make sure that we're proactively including everyone everywhere we, we, we go. And, um, I think that there is an opportunity, uh, as, uh, your guests said in the interview, um, it, it's a, it's a wide open space.

There's, uh, opportunity for people to stake out new claims and to find new spaces that they feel more comfortable in. Uh, but of course, we have to be aware that many of the things that we do can exclude people and can keep people out. Uh, a couple of points, uh, you know, to to, to Mike's point, um, one of the things that I am actually pretty optimistic about is the collision between data and analytics and ai.

The data world is much more inclusive, uh, generally than the rest of it. If you, uh, you know, I was recently at the click connect show, for example, and, uh, it was much, much less dominated by people that look like me than most shows that I go to. Uh, which it was a very, very pleasant surprise to be involved in that, uh, data space and, and, and people in the data and analytics in a field feel, I think a lot more open to people that aren't traditional IT nerds.

Uh, and that's good. Uh, and, and I think that that's gonna, IM, IM impact ai. I think also the fact that, um, many of the spaces, as your guest said, are new and wide open, that means that people can come in, they can plant their flag, they can say, you know, I'm an expert in this.

If they feel empowered to do that, or if controversially or conversely, they don't feel disempowered to do that. So it's important for us to, to keep from pushing people out. And then the other thing that, that, the point from your interview, Bonnie, that I think was really, really incredible was this whole international aspect of it.

And the fact that, you know, we talked about this on, on the utilizing tech podcast a few years ago. I was interviewing some folks who were talking about the different ways that impact is, is impacting, uh, different parts of the world. And absolutely, there's, there's a lot being done in, you know, traditionally underprivileged company or, uh, countries and spaces and, and people.

And they're getting involved and, and, and it's gonna be exciting to see what ideas they come up with. So I do think that there's reason to be optimistic about this, but also, you know, back to the Randstad study that we talked about in the first block here, there's a 42%, uh, gender gap in, um, AI training between men and women. You know, I, I agree.

I was gonna bring up the same point about the data community, uh, being more, much more represented by women, women. Not saying still isn't a male dominated white male world, but it, that's my experience just in the workforce. Uh, I think a couple things is when we say AI skills, the, the thing we're at a different place than Alan and I were when we got into security in the early two thousands, late nineties, right?

Very different world in terms of men and women and, uh, e you know, equity and inclusion, things like that in the workforce. Same thing, even with cloud, you know, that 15 years ago is a different environment. So hopefully some of that, some of that progress, uh, of, of being under way understanding that this is important and things you can do to help move that along.

I think the other part of it is though, that AI skills fall into a lot of areas. You mentioned data and data analytics, Steven, there is user experience design. There is, um, as well as user experience testing.

There is analytics in, in algorithms, not just in data, uh, to get into, there are new applications. So how do you design this into either user experience or into a workflow, uh, in ai? Um, how, how do you direct the outcomes to be more useful in ai, right?

Where someone may be just real focused on getting an outcome. Well, how do you make sure it's one that's actually gonna be, move the ball down the field or be useful to the business, to the end user there? There's a lot of aspects of the, of AI because it like security, it touches everything or can touch everything.

Mm-hmm. So the good news is you don't have to start over and learn ai. You can take what you do in your special specialty areas are, and add AI to that.

And maybe a hard, you know, hard move into ai, but bringing those skills with you versus, I've gotta start over in something else. I'll take it a step further. To your point about women and data.

We all know there's a bunch of IT guys out there that are, you know, they're biased because, you know, they grew up in this math culture. I think they're all about to get their ass kicked. 'cause they're gonna all wind up working for women who have the data expertise or the business outcome expertise.

And if they can't figure out how to get along, they're gonna be out. 'cause people aren't gonna put up with that. They're gonna be looking for it.

People who can cross pollinate into different disciplines and understand how it works and drives different things. But the days when, you know, you could sit around and go, well, I just work in it and it's a guy shop, come to an end. So I think, I think you better wake up and smell the coffee real quick.

So to paraphrase, that'll be, that'll be those women's jobs. Um, okay, look, as I said in the outset, I, I pine for the day where we don't have to have these distinctions and, and we have to keep score on a scorecard of how many men, how many women, how many black, how many white, how many this or that. It's just humans.

So let's take a break here on Textron Gang, and we're gonna come back and talk about a platform mindset. I have some views on that, but I think Steven does too. You are watching Textron Gang Discover Textron Group, the epicenter of tech innovation.

We are your go-to for reaching IT leaders and practitioners worldwide. Our secret impactful content that sparks awareness, engagement, and top quality leads with us. You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more.

Join our satisfied clients. Let's revolutionize your tech journey. Contact us today and tell your story to the world in the most powerful way with Textron Group.

All right, folks, we're back. com and we've been inviting people to contribute content. And we got a post from the folks over at Capital One talking about what it takes to have a platform mindset.

And this was great 'cause Capital One is also one of the leaders of the DevOps movement, and they still make massive investments in that space, and we also encourage other folks to make similar contributions. But, um, Steven, I know you read this article, we've been talking about platforms for as long as I can remember in it. So when you hear the phrase platform mindset, what does that mean to you and what is changing here?

Well, I I will say from the get go that sometimes it scares me a little bit because, uh, there's a negative to a platform mindset, isn't there? I think, I think in many cases when people hear about, uh, the development of a platform or they'd hear about platform engineering, they're a little nervous that we're going back to the olden days when it was its own little world and it said, this is the box you must fit your applications in, and there's no flexibility. And this is what we use and this is our standard, and we made it the standard and we don't care.

And that's the it that I came up in when I was a, you know, uh, working in, in industry, it was very much separated and distinct from developers and platform or, and, um, applications. We didn't know what applications were running our platforms. We didn't care.

This is our platform and you'll run it on that, or you won't run it. Well, that's not gonna work in this modern DevOps world. And how do you avoid that?

So I, I wanna start off by saying that, um, the linked article in here, the Capital One, uh, presentation, I love that they talk about Conway's Law, which is that an organization that designs a system will produce a design that's with a structure that copies the organization's structure. In other words, you're gonna get the platform that you ask for, make sure you ask for the right one. And their way of, of dealing with this is to use a, a, a concept that I think is, is a very smart one.

It's one that I was exposed to actually, um, through my, uh, work in it previously, uh, called a, uh, community of practice. And the idea is that instead of saying, you know, you three go over there and make us a platform, would you, it, it's a very different app approach. Instead, you have essentially a, a more, a more open and inclusive group that comes together from a variety of perspectives that defines the problem, that figures out what needs to be done that tries, you know, proactively to work together to build something that works, something that makes sense in this particular environment, instead of having it be very, very strictly structured.

Because as soon as you say, you know, Hey, Mike and Mitch, why don't you guys get together and build us our IT platform? We are gonna get one that's gonna be very top down and structured and just what Mike and Mitch want, right? But if you say, Hey, where are the stakeholders who wants to come together?

Let's come together. Let's figure out what the problem is. Let's figure out how we should build this.

You're gonna get a much better, a much more appropriate platform coming out of that. And if you think that can't work, well Plat, uh, capital One would like a word with you because as they talked about in their, uh, platform con presentation, they were able to get it to work. So, Steven, I, I don't disagree.

I, and by the way, I think in modern platform engineering, there is still a little bit of that old Conway's law platform, but it's much more of the new way that you described, which is more inclusive, more input that helps us build this platform. com, we've also launched a new video series slash podcast called the Platform Engineering Show. I am one of the cohosts of it.

My other cohost is Luca gte. org, uh, community. It's two to 300,000 strong.

We recorded our first show, I think it's out this week. And the second show is being recorded, I think this week to be out. And we will be doing them every other week, and we'll have live webinars and very much like we do, for instance, Mitchell and I do on DevOps.

I'm bound. I think when we look at the platform mindset in today's idiom, right, in today's kind of world, what we're looking at is a post DevOps platform mindset, which is for a lot of DevOps. One of the evolutionary branches on DevOps was shift left where, Hey man, let's just shift it left.

And not just shift it left meaning earlier in the timeline, but let's throw it on the developer. So builder, developer, builder, you had to build your own platform to develop on 'cause you wanted that. You, the developer is the Alpha Predator gets to pick, or she, unfortunately, all too often he, but they get to pick what platform they're building on.

They get to build the platform even. And that, that may work when you got five developers. It doesn't work at scale.

And I think right there, that's the platform mindset. We can't shift it left onto the developer or the builder to build this their own platform. The platform mindset is we need to build a platform that the developer can go faster on that the tester and the security person and the SRE and all these people, the ITSM folks, even all these people who make up the employees of the modern software factory Can work on.

And in order for that to work, we do need to be inclusive. We do need to hear from all, you know, participants to build the best platform we can. But right there, that's the platform mindset.

And I think it's in our growth to shift left. I'm sorry, go ahead, Mitch. I, There's a lot of wisdom in this article, and you can tell it comes from experience in that you could probably replace platform engineering with DevOps, with Call anything you want, Whatever, right?

It, it's about bringing people together, getting buy-in, getting alignment, you know, and solving. But it's about solving real problems. And I think that's, you know, that's where the why questions comes in.

Why are we doing platform engineering? Why do we want to do this? Is it because we're supposed to have one?

Because that's what they talk about on Textron tv or is it some other reason? Right? And, and the, the good news is there are some very tangible problems that you can measure success from.

Yes, it can be developer productivity, it can be complexity. We have too many things that takes too long to, uh, release software or to update configurations, uh, or onboard people, things like that. Um, it can be, it can be, uh, making it easier to find the tools and information with portals.

So there, there's a lot of kind of dimensions that we've, what what we put around the platform engineering box. And I think like DevOps, that makes it pretty flexible. You can, you can adapt platform engineering to what your organization's needs.

Hopefully not in the Conway's Law way, but in a way that, you know, our, our problem isn't complexity. Our problem is this, right? Let's go work on this.

And we didn't know if we, if we've got that outcome. So I think you're focused on the why, why you're doing it, and how are you gonna measure that success. And is success done defined by Mike and Mitch that went in often and designed our own platform?

Well sure we know how to do that. No, it's the developers, it's the people who are doing SRE. It's the end customer.

It's whatever, whatever problem we're, uh, searching to solve. We wanna make a difference for the business and for our people that we work with. Yeah, I was just in here minding my own business.

What platform are we building? You and me, Mike. We gotta sign to rebuild all the platforms for Textron.

Yeah, We've got a new content platform using ai. Mm-hmm. Well that's the future, right?

We'll just have AI design the platform for us. We'll just talk to it and it'll come up. I'm sure it'll be great.

I'm sure it'll be just fine And it won't have any opinions whatsoever. Right? And if there's anything wrong, it'll fix it.

Well, you know, the answer to this problem in the past was, you know, if you didn't like the platform, we just built you another one, right? And so then we wound up with multiple platforms. So how many platforms can there possibly Be?

But that, that is one of the things about that platform mindset though, is that it's about scale. And you, in order to scale, you can't have infinite platforms. 'cause having infinite platforms is having no platform, right?

You need a platform, not infinite, but, But flexibility is the kind of key to it. 'cause to Steven's point, everybody embraced DevOps in the first place to get out from underneath the thumb of centralized it. And to a lot of folks, this smells like, you know, the, the same as the old boss, as the new boss.

So are you building an experimentation? Are you building a way of new things to come in versus just dictating the dogma of what the standard is? Right?

That's what, that's what we all hated or hated About. Sure. That in there.

But, and that's why, and that's why this has to platform engineering has to be inclusive, not it has to bring in pe, bring in the developers, bring in IT ops, bring in everyone. You know, I'm one of those people who feels like shift left. It was a, it was a smart idea because we had shifted way too far to the right.

We had shift, we had put too much power in the hands of IT. Ops and developers were really dissatisfied and it was holding them back from doing their job. But unfortunately, developers don't really want to own and manage production platforms.

They don't wanna be specifying platforms. I, I think ultimately they would like to have a flexible solution where they can just say like, these are the resources I need. Uh, don't tell me I can't have it, but I don't really wanna get into the nuances of how this is delivered.

So let's do this. Uh, that's why they need to have a seat at the table too, so that we can shift center instead of, you know, we don't want, we don't wanna shift back, right? We don't wanna back go back to the world of top down it.

But we also don't want to have developers have to worry about, you know, maintaining and operating databases and servers and storage. I think you're spot on. I think though that the leader of the platform team has to be somebody who has some app dev experience, right?

It can't just be, you know, the former IT ops team showing up saying, yeah, we suddenly get it. To have credibility with developers who's ever leading that team has to have some actual coding expertise. But they also have to have some actual operations expertise because they have to build something that is sustainable and, and, and can be operated Reliable, be inclusive.

It's gotta be inclusive actually saying, We're embracing distributed computing as long as it's in the data center, we're good. com is a great place for you to be frequently. We've got great content like this.

Also the platform engineering show, if you're in a video and podcasts, we'll be exploring a lot of topics around this on there and, uh, stay tuned for that. But guys, hey, we've been on a while. We gotta bang off of here.

We've got a ton of texture on gang material coming up. We've got a full week this week, our last week of the year. Just so you know, we are not doing Text on Gang the week of, between Christmas and New Year's.

So basically it's this week and we'll be doing some yearend wrap up stuff I guess towards the end of the week. I think, Steven, you're going to give us a yearend wrap up of what you thought were the big stories of, uh, tech Field Day this year. Yeah, absolutely.

Um, we're gonna be, well for what it's worth, we're gonna be doing a, um, a wrap up of the, um, um, infrastructure it infrastructure news on Gestalt it on Wednesday, which will be airing on Techstrong tv. Uh, we'll also be talking about what, uh, what the major trends are that we saw at Tech Field Day this year. And I'm, uh, really looking forward to talking about that.

Very cool. Until then, though, on behalf of the Textron Gang, this is Alan Shimel. We're outta here.

Have a great day.