Modernizing Applications and Preparing for GenAI with ClearScale’s Kevin Epstein

This is Textron Tv. Hey everyone. Welcome back here to techron TV at Talent Shimel.

And we are happy to have a first time guest here on Textron TV with us. He is Kevin Epstein. Kevin is the Director of Customer Solutions at ClearScale, and we're gonna learn all about ClearScale and generative AI data prep in the cloud.

And maybe a little bit about Kevin. Kevin, welcome to Text Drunk tv. Welcome.

Thank you. Great to have you on here. Um, Kevin, as I mentioned, you're director of Customer Solutions over at ClearScale, but give us a little bit of your background, your history.

Sure. So, uh, just to level set, uh, director of customer solutions, what I do is, uh, chat with customers all day long, understand what they're trying to do and how they're, what they're trying to achieve and, and help them map out and chart out a path, get there. My path to where I am today is I started out as an infrastructure engineer, uh, back in the day on big Iron Unix systems, um, working through the data center and going through the whole VMware virtualization craze, and then obviously eventually HMR wagon to, uh, the public cloud, which I've been doing since around 2009.

First as initially as a customer, and then, uh, from 2012 onwards, uh, as a consultant in the field. That's when they were perpetual license, not their new subscription licenses out there. Huh?

It was a different world. It was a different World. Yes, it was, it was a very different world.

I don't know if you saw it today, it was an article on one of our sites at t is suing Broadcom over those VMware licenses, so it's a little crazy. Um, it's interesting background, Kevin, and congratulations to you. How long have you been with ClearScale?

I've been with ClearScale just under, um, two years now. And, uh, prior to this I actually spent a stint at, uh, Amazon Web Services, the, the, so-called mothership, where, where, where the, the stuff gets made, um, spent my time over there supporting, uh, state and local government and, uh, all through my career being, uh, supporting primarily me and large enterprises have businesses. Excellent.

Alright, let's, um, let's move over to ClearScale, if you don't mind, Kevin. I, I mean, it's a company. I, I've known of ClearScale for some time, but I'm not sure if everyone out here has, why don't you, if, if you don't, you know, if you can give us a little kind of elevator pitch on who's ClearScale, what you guys do.

Sure. So ClearScale is, uh, a AWS only cloud consulting partner. So we go deep, uh, on the platform and also go very wide in the platform.

So as some people may or may not know, AWS has a lot of services. Um, there's some work they might consider core services like, uh, EC2 and IDS for compute and databases. But then they also have, um, the breadth which goes to things like SageMaker and Bedrock for AI and ML solutions that, uh, customers are trying to build on the platform.

ClearScale actually started around about 13 years ago, uh, primarily in migrating customers to the cloud and also doing app modernization. So our premise was going to the cloud, simply exchanging your virtual machine for a cloud virtual machine isn't necessarily the best solution forward. And, um, what we actually wanted to do was propose to customers that if you, uh, if you modernize your application and take advantage of the elasticity that the cloud offers, that will be where you'll find your savings and your efficiency.

And that, that really is our focus. That's, that's how we take customers to the car. Excellent.

com, is that the right website? com. Yep.

And look, and I know this, and I think most of our audience knows this, right? There is a, a large ecosystem of AWS partners who are pretty much exclusive to AWS that are really helping with, you know, digital transformations, migrations, and, and all the rest. And of course, uh, AWS re event will be here, I think it's December 2nd to the sixth.

This year was kind of ringing a bell in, in my head, and we'll be there doing videos as usual. So very excited with that. Um, Kevin, if you don't mind, I think one of the, the, the big things that are is on everyone's mind of course, is ai, right?

Gen AI is kind of changing, a lot of, is changing a lot of what we do, but the potential for it to change even more is, is sky high? Um, and that is no, that is just as true when we talk about moving data to the cloud, preparing data, you know, in how, how we're going to use it in the cloud, where we're gonna store it, how we interact with it, and you know, that the whole idea of moving data around is always a scary prop for a lot of companies, right? It, the idea of taking data from here to there or, or what have you, is, is fraught with risk gen AI has a lot of promise to make this almost like data prep, uh, as code, if you will, right.

Automate it and so forth. What do you, what's your thoughts on it? What's ClearScale doing around this?

Um, so you're right, this is on everyone's mind, even if they're not entirely verbalizing it that way. Everyone is talking about gen ai. Um, if they're not saying gen ai, they're seeing things like chat GPT.

So it's on everyone's mind. Um, to your point, the thing that most folks are gonna struggle with, um, and, and I'm not talking about the folks who are gonna use chat GPT and type in a question and get an answer, but the, the, the businesses that are actually going to use generative AI in a meaningful way to change their business, those folks have a problem that may seem new to them, but actually isn't all that new. Uh, we've been talking about big data since the, the early, uh, uh, 20 teens, I guess.

And, uh, you know, there was always the big data conversation. How do we wrangle big data? How do we store it?

How do we report it? How do we, uh, clean it? And quite honestly, when I talk to customers today, this, these are the same questions we're getting.

So really what's happened is, uh, the folks who didn't round that bull the first time round are like, Hey, we have to do this. We still have to do this. Uh, some of them are struggling to actually maybe understand that that's what they have to do, but, um, data prep is the number one thing you can do to be successful in a generative AI pursuit.

Absolutely. And, but what, what's kind of the current state, if you will, of, of generative ai, you know, helping in here and, and really becoming a force? Yeah, I mean, I think, uh, it, it should be unsurprising that most folks start out with a, um, you know, a chat bot style solution, kinda get the bot to chat and, and, and seem, uh, seem logical and sensical and things like that.

And then slowly, uh, customers are saying, well, what, what can I do with it next? Because actually the chat bot isn't the thing that's gonna change anyone's world. Um, and so we start to talk about things like agent tech, uh, um, ai, and we give, uh, we give the AI agency and, uh, the ability actually to do something.

And that's where it will change people's worlds because it, it's quote unquote autonomous. It can make a decision based on data it received, and we've given it parameters, and these are the things you can do with that data, make the best decision. And, uh, when, when we get that right for customers, that's where they see the biggest difference.

Agreed. Agreed. Um, how I, I, so is this, is this whole, I mean, this is happening in real time, right?

Has ClearScale sort of productized this already or it's service sized it where, you know, the, the technology is mature enough, the problem that we're solving is understood enough where you can offer a viable solution, you know, at scale to, to customers? Sure. So we, we actually, um, we have a solution that we make available to customers is really kind of like a starter kit for generative ai.

We call it AppLink. And the reason being is we have a framework of generative AI solution that can be very easily integrated into a customer, uh, uh, environment and platform. And the idea here is to essentially link it to their data and start to give them interesting outputs.

And so AppLink today focuses primarily on, uh, rad, which is a retrieval, augmented, augmented, uh, generation. Um, and that just really helps customers use the data they have and the technology behind AI to actually, uh, answer questions. They have to find documents, uh, to surface documents that they don't know they have.

But the generative AI can find it just based on, on the way we storage for using vector databases behind the scenes. Excellent, excellent. Um, Kevin, people out here say, sounds interesting.

I'd like to learn more. What, what's the best way for them to engage, like to, you know, dig in, whether they're looking to buy right now, or maybe they, they're just information gathering. What would you recommend to folks?

Um, I mean, honestly, I, I recommend, um, do a little bit reading, um, try to think about some of your use cases. That probably is the stumbling block most folks have is what are, what is my use case? Because if it's your chat bot, I mean, you may, you may be successful in build one, but it's not actually gonna be useful to you in the long run.

Um, so really think about what could we, you know, if I did this, what would change my world? And then start to list those things out and then, um, you know, reach out to folks like SKA and just, uh, chat with us and we can also help you, uh, on the ideation. We've seen what customers are doing with generative ai.

Um, we can tell them, yeah, people have done that. It hasn't worked spectacularly or, that's very interesting and you should pursue that. We, we can, just based on what we've seen other customers do, we can give some good input back to customers.

Yeah, I mean, look, I, I think the, uh, the tech industry always embraces the next great idea, right? And, and AI has certainly been a, a huge big idea. We, we can't lose sight of the fact though that it's still relatively new technology in the marketplace and we're still experimenting right in, in what are the best ways to use it, how to use it, what to do and, and so forth.

So I, I think, I think your advice is dead on, right? Sometimes the best thing is to learn from what other people have done and how other people are doing it, and you build on it, right? And, and these things build on it and build on it until you go from, like we used to call it, from emerging practices to best practices.

I don't think we've established best practices yet. I think we still see emerging practices coming out here, and it's a, it's an interesting time to be in the field. I'd agree.

I I think we, we have some best practices, um, on the data storage and preparation side of things. Yes. On how does Gen AI do that and what do we do with Gen ai.

I would agree that that's still emerging. I think every day is changing and it will continue to change as long as new models are coming out with new capabilities. This, this is not a settled thing.

This, there is no, well, that's what it is now. Agreed. Agreed.

Hey, Kevin, I want to thank you for coming on here today and telling us all about this. com. com.

People can catch it there. And if you're at re event, stop by, say hello. We'd love to see you.

Absolutely. It'd be great. Alrightyy.

Kevin Epstein, director of Customer, uh, solutions at ClearScale and AWS partner here on Tech tv. We're gonna take a break. We're gonna be back with lots more.

Stay tuned.