Unlocking AI Potential in the Enterprise | Utilizing AI Ep. 14

Enterprises are deploying AI technology across the compute stack from the cloud to the data center to the edge. On this episode of utilizing ai, Jeremy Foster, SVP and GM at Cisco Compute discusses the reality of AI applications at the edge. With Brad Shiman and me, AI applications and retail manufacturing, healthcare and other industries are increasing the demands for processing at the edge.

And companies like Cisco are leveraging the advances in GPU networking and storage to meet these needs. Although mega projects get all the attention, maybe the edge is where the real value comes from. Ai welcome to utilizing ai, the podcast focused on practical applications of artificial intelligence from the Futurum group.

Every Wednesday, we explore news and use cases in the ways in which AI is transforming enterprise IT and the industries it serves. I'm your host, Stephen Foskett, president of the Tech Field Day Business unit here at the Futurum Group. Before we dive into this discussion, let's meet who's on the panel today.

Hey, Stephen. I'm Jeremy Foster. Thanks for having me here today.

I'm the Senior Vice President for Cisco's Compute Platform And Stephen Brad Shiman here from Futurum as well. I'm the VP and practice lead for data intelligence, analytics and infrastructure. And as I mentioned, I am the, uh, head of the tech Field Day business unit now Tech Field Day and Cisco go back a long time.

And Jeremy, we have worked with you many times over the years at various aspects of Tech Field Day, including, uh, literally just last month we had Cisco present at our AI infrastructure Field Day event, but Cisco's a big company. Um, before we get started here, talking about how we're utilizing AI and how enterprises are gonna be using it, tell us a little bit more about the Cisco Compute Organization just so that we have some context for this conversation. Yeah, so within Cisco Compute, we provide a lot of the computing solutions for the AI use cases in the data center all the way out to the edge, as well as, uh, help work with the other business units at Cisco to provide the appliances and things that you run Cisco software on for something like Catalyst Center or an AP controller.

So we, we really have a broad focus, not only on the compute platforms themselves, but also on the management platform, which is really where we, uh, focus on differentiating the customer experience with Intersite and helping customers day-to-day lives be a lot easier running and managing their compute environments. And we had a whole, uh, season or two, uh, talking about AI at the edge. And I think that that's really kind of what we're gonna wanna dive into here is the ways that AI is going to be implemented.

Now, I think most of us realize at this point that AI is pretty much gonna be everywhere in the enterprise software stack, uh, from the infrastructure and operations side all the way out to the end, uh, applications and, you know, customer service and so on. Um, tell us a little bit more, give us a sense, how are Cisco's customers deploying AI based solutions in 2025 and 2026? Not how do they hope to, or, or what do you expect in the future?

How's it really happening now? Yeah, these are great questions, Steven. There's a couple of schools of thought, right?

Edge is not just about ai. So as you had this long, uh, podcast series, right, I'm sure part of it is all about, there's a tremendous amount of compute and storage and networking that gets done at the edge, whether that's a retail, healthcare, manufacturing. So what's really shifting in the AI world is what are those use cases that people are gonna go after and wanna deploy?

And that's largely driven by the software capabilities that are delivered from the ISVs, whether that's the realtime analytics for things like video, which obviously applies to a whole host of vertical use cases. Uh, better industrial automation for things like manufacturing and predictive maintenance type use cases. So those are the things that are driving, I think, a big shift towards what we're seeing customers going after.

And, you know, enterprises deploy applications after all kind of, that's the, their core tenant. And certainly within information technology and, and the, the lines of business owners are saying, Hey, we have new capabilities that we wanna bring online. A lot of those use cases are gonna play themselves out at the edge.

To your point, that's where a lot of the data itself is being generated. And so you tie those things together and you take what used to be an area people deployed infrastructure on, uh, because it was a necessity, and you started transitioning that over to being something that's actually like a big business driver for everybody in the industry. Yeah, it's interesting you say that, Jeremy, because we just finished publishing a forecast, um, for the data intelligence analytics infrastructure service that, that sort of asked the question, well, what are companies spending money on?

And, um, by and large, what we're seeing is an acceleration of investment, but it's not just in pie in the sky what ifs it's in, how do I actually optimize the spend that I have so that I can do more with that spend? And as you mentioned, and very rightfully so, that isn't happening just in the data center. It's happening at the edge on device across the entire spec compute spectrum here.

And so being able to right size and right locate, uh, your, you know, resources that you're using to build AI in, you know, solutions is critical this year. Yeah, I think it's, it's really different in terms of what we're seeing today versus what we saw, you know, even just a few years ago, right? And that, that ultimately goes back to any of these use cases are generating a tremendous amount more data.

So if you, depending on the various sources, you know, they, in the industry, they'll call out things like 75% of the enterprise data is gonna be generated at the edge in the next couple of years. You know, some people even quote that as that starts to shift, that means the underlying technology has to be updated refreshed because those edge locations most likely aren't ready to use the technology and that, uh, that that's gonna be required. So as AI inferencing starts to, uh, want to be put closer to those workloads, now you've gotta have edge AI ready hardware to go out there to be able to support that.

I think we've clearly closed that gap over the last couple years, and that enables these use cases. Um, and then that's where customers are saying, great, I've got edge infrastructure, it's critical to my business. I can now upgrade it, I can spend it more efficiently.

Because ultimately AI is the, an amazing way to drive efficiency into your business processes. And depending on which one of those use cases we talked about earlier, uh, is gonna dictate, you know, how and what our customers are delivering. Um, in terms of the actual use cases that we're seeing, I think it, a lot of them are right now realtime video analytics related.

Uh, that in healthcare is some amazing use cases. Like I was talking a meeting with a, a hospital group a few weeks ago, and they were telling me the new deployments that we're working on together can leverage eight megapixel cameras in the individual's rooms. And actually without tethering these devices to anything else, just the video analytics themselves, their software is able to understand when that nurse uses a syringe to pull out, uh, whatever medicine that she's going to give.

If it's supposed to be, you know, two milliliters and they pull out three, an alarm's gonna go off. If they pull out one, an alarm's gonna go off. And that's amazing.

Think about the applications of making sure that you're delivering the right outcome for that patient. And they had a whole host of those types of, uh, you know, use cases and, and you get into manufacturing, you know, realtime safety is one that's very, uh, I think commonly deployed right now and, and out at the edge on those IOT devices. And then retail is just completely reinventing their business.

So when you look at things that we're getting involved in now, not just on the networking side, which is historically where Cisco's played at the edge, but now across the security, uh, side of things, the compute side of things, the storage side of things with that unified edge platform that we, we launched, you know, we're getting to really play out all the different pieces of use cases and aligning to those ISVs and working with them to deliver those use cases is a pretty, uh, pretty fun place to be. Yeah, it's, it's interesting that you bring this up because, um, so, uh, season five, we talked, uh, back in 2023, we talked about Edge and one of the sort of, uh, overwhelming messages that we got was that the, uh, development of, um, machine learning technology would actually increase the collection of data and increase the capabilities in manufacturing and healthcare and retail and other edge locations. And you just said exactly what they predicted would happen.

And, and, and it's, it seems like a paradox that increasing processing of data at the edge will not reduce the amount of data, but will actually increase the amount of data collected. But that's exactly what was predicted and what we're seeing. And those of us who have studied this space, and even those of us who are just existing in this world have certainly experienced this.

I I went to the doctor recently, and of course now every time you go to the doctor, they have, uh, AI assisted note taking. Um, I went to a, uh, retail location that had, um, the sort of grab and go and walk out, uh, recently. I, um, you know, similarly have, um, worked with some folks, uh, professionally here at Futurum who are in the manufacturing space, and they are doing exactly the types of things that you're saying.

They're deploying higher and higher resolution sensors and cameras. They're processing video, they're doing things that they never thought they could do. Uh, one of our favorite, uh, one of our good friends, uh, works at a greenhouse and they're using, uh, cameras to literally watch all the tomatoes and strawberries as they're ripening, and they're able to use AI to really monitor that and improve crop yields.

This is the sort of thing that had been promised from ai, but people are, I think, distracted by the, the, the ever present chatbots that they don't realize that there's really practical stuff happening here, uh, at the edge. So I'm really glad that you're, that you're bringing that up. And of course, this demands a great deal of processing and storage, uh, horsepower.

But I think what you're saying is, yeah, we got that. And, um, and that's been my experience as well, talking to some of the other hardware vendors. Um, you know, what sort of hardware is being deployed out there in these types of locations.

Uh, you know, how much iron is there at the edge, Right? Uh, I would say there's a lot more today and certainly a lot more coming over the next few years. It's gonna depend greatly on the, the use case and the, and the situation, right?

If you're in retail, for example, uh, you, you can look at some publicly available information around like, yum. Brands talked about doing a big deployment, uh, of tens of thousands of locations. And you know, a lot of that type of work doesn't mean you necessarily have to have, uh, a GPU training grid every time you need to order a chicken sandwich, because it turns out if you just wanna use your voice to order a chicken sandwich, there's only so many ways that Stephen and Brad can order a chicken sandwich.

So a lot of that processing can be done with, uh, just regular CPU. And those are the types of, uh, important things that you can do to optimize inside of retail where you wanna get the right tool for the right job. Then you go to manufacturing.

Well, in manufacturing they may have a facility, a lot of times that's gonna be more like your data center, maybe not your enterprise data center, like in terms of size, but obviously there's a lot of technological equipment that go into a manufacturing facility. So you might have a little bit more iron there. And depending on if it's a robotics type use case, you might have a lot more iron there.

And then you get back into the core data center itself. Where I think the important thing about Edge is that, yes, we're making a lot of changes out to those architectures from hardware perspective, but to kind of play off the manufacturing thing we might be using on the manufacturing floor in real time doing analytics to understand whatever widget we're making, uh, what the quality of that is, and flagging the, the something that doesn't meet our standards and pulling it off the line, right? But at the same time, that company probably has 50 locations with 50 of those machines that all need to be able to talk back to what's happening, uh, and take that data that they're gathering from each one of those 50 machines, push that back to their data center that's maybe in the cloud or maybe on their enterprise data center floor, retrain that model to make it as efficient as possible and then redistribute it back to the edge so that those 50 machines are running better today than they were running three weeks ago.

And, you know, those different types of scenarios are gonna depend on what kind of iron it goes out there, but, um, I think there's a whole gambit of technological solutions and then architectures that are built to support them. Yeah, Jeremy, it's not just a case of plunking down a CPU and some storage and maybe a 50 90, you know, sitting on the side inside of a retail to say, I'm doing Edge. Is it, it it's a lot deeper than that in terms of the things you have to consider as a business that's trying to take advantage of that scaling out across all of these.

Pss tell us if you would, why you companies need to think about that hardware that they're putting down and why, for example, having the optionality to, to say swap out, you know, CPU for GPU, you know, right next to each other in the same box as critical. Yeah, I think it's the same reasons that optionality isn't critical, is for the same reasons that a lot of these AI projects ultimately end up failing, right? So the same things that apply to the edge that apply to these core use cases, whether that's, you know, I need to have data readiness, I need to make sure I have infrastructure that can scale, uh, I need to have in network, which at the edge is a lot different network concerns oftentimes than what's happening inside the data center.

But the network's gonna be critical in how I'm moving that data around. And then security and governments like those four things. You, you've gotta have it in the data center outside, uh, you know, the data center.

And I think too, it's also, um, when you look at the data center itself and in terms of how we treat the racks that we have and how we're always trying to optimize those by cramming as much into, uh, you know, a single unit as possible to, to optimize, you know, what those cycles are doing, instead of having dead time downtime to always be, you know, right at the limit, you can, you know, tolerate if you will. And I presume, sorry Jeremy, go ahead, man. Oh, I was the, the other point you asked about the flexibility in terms of the hardware, right?

A lot of customers dunno what they dunno, and that's okay because they wanna get started with their AI use case, but that iron, uh, to Steven's point may change pretty quickly. So like when we were building out the Cisco Unified Edge, for example, we pulled a lot of customers, how many people are gonna actually deploy GPUs out on the edge? Some people said, you know, 15% of people said, yeah, we're already doing it primarily in retail for those video type use cases.

And a lot of other people said, about 75% of 'em said, I'm not doing it today. But if you ask the question over the next five years, do you think you're gonna need the type of technology edge? The answer 90% of the time is yes.

So how do you help somebody refresh what they're doing today and then be in a position to upgrade easily later? Those are the types of problems we're trying to help tackle with, with some of the new technologies that we're deploying. And it's also to simplify as well, is, is it not?

Because if you have that optionality, uh, you know, in a unit you can bring in the workloads and that you need like, let's say security for example, right? Right next to inferencing. I think, you know, if you're managing separate devices, that that's a lot, a big, a much bigger load and a bigger cost factor.

Yeah, it absolutely is. And I think the other piece of simplification is the software, not only the software stack that they're gonna be deploying, but how are they operating and owning this equipment? It's one thing to deploy it, it's another thing to be able to make it easy to upgrade and operationally efficient out at the edge where it's oftentimes you don't have dedicated IT staff to be able to go into your retail site and, and perform these upgrades and those types of things.

Yeah, that's expensive. Yeah, We've heard, yeah, we've heard a lot about that, um, at the so Edgefield Day and the, you know, the edge, uh, series here. Um, and that's something that there's a bunch of companies that are doing some pretty cool things on.

Um, can you talk a little bit about how Cisco is working with partners in this? Because you know, I, I assume that you're not talking about coming in there and, and just sort of dropping some hardware 'cause that's not how you guys roll. And, and what we've heard from, um, customers, uh, as well is that they're really interested, especially when it comes to deploying ai, this is such a new space, they're really interested in working with a partner that works with other partners that can deliver a complete solution.

So talk a little bit more, if you can, about how you're working with other companies to solve some of these problems that they're facing. Yeah, Brad just said it, you know, at the edge that's expensive. So when you're doing something one time, it's expense can be expensive when you're talking about AI in the data center.

But when you're doing something across 10,000 locations, it can get really expensive if you make a mistake. And that's where Steven, to your point, people wanna work with partners, people wanna deploy tested architectures. So what we've been building is with the Cisco Unified Edge, putting together a platform, and if you listen to G two, who's our chief product officer, you'll always hear him talking about the platform effect.

And for many years we've obviously sold routing, we've been out the edge bringing security and the type of policy and governance out there. Um, but we haven't necessarily been in piece. So what we wanted to do is bring it all together where we could take compute storage, Cisco networking, and Cisco, you know, WAN optimizations and all those technologies that have been there, put it together not only for our customers, but for our partners and for those sis out there that are deploying these types of solutions for customers and make it simple for them to deploy and for our, as it relates to our channel partners and things make it simple for them to deploy for our customers and for them to operate a business on top of.

Because if it's easier for a customer to deploy, it ultimately is easier for a channel partner and si to be able to maintain that for a customer. Uh, because oftentimes to your point, even customers should say, Hey, I'm gonna pay somebody to make it work. So now our channel partners and those sis have a, a chance to run a more operationally efficient services business.

Yeah, I think efficiencies, when you're talking about maintainability, you know, are, are paramount and as you mentioned, Jeremy, if I'm rolling somebody, you know, in a car to a location, um, that is money out the door. And so having the ability to, to not just centralize in a flexible way, but to also enable companies to automate, um, that the management of that asset at the edge in the hospital at the point of sale is critical. Can you tell us a little bit about, you know, how Cisco views that?

I know that the company's made quite, quite a few investments in that area in particular. So I I think our viewers would love to hear a little bit about, you know, how you're using things like AI itself to help make those devices more manageable. Yeah, probably two different levels.

There's the Cisco level, which we can talk about with, uh, uh, the overall integrations that we're doing across the many different solutions that we have, right? You know, you have network solutions, compute solutions, storage solutions, you name it, we've probably got it in the portfolio. So the first thing we have to do is simplify and integrate those, uh, things into a common experience.

And that's what we're doing with things like AI Canvas, which you've seen recently that's actually leveraging ai, uh, yep. To, to bring together the Cisco cloud control and bring those individual element managers things like inner site on the compute piece, uh, or it might be catalyst Center on the, on the enterprise networking piece, and start to bring those things together. That's, that's really important because we don't wanna eliminate customers operational approaches to managing their existing infrastructure.

In other words, when we talk to a customer and say, well, do you want one tool that does everything? They typically say no, but they would like to have their networking team do what their networking team does, their compute team do, what their compute, storage and virtualization team does. Um, but at the same time, what can we do to use AI to make that overall experience faster and more intelligent, right?

That's what we're focused on there. And then when you double click into, okay, well how do we do this with Cisco Unified Edge, what were the things that we cared about? Uh, it's kind of funny into this job probably three years ago, but I've been Cisco for a spent sales organization crawl around customers, data centers, customers edge locations, and it always bothered me that when you looked at an edge location, there's wires everywhere, there's a router, there's switches, there's servers, there's computer, and basically, I would look at the type of thing and go, gosh, as an industry, we've really failed customers at the edge.

Like it's just an afterthought from our, you know, from our competitors, frankly, where they build crappy desktop computers and make 'em as cheap as they can. And they say, Hey, customers, why don't you figure out how to manage all this stuff? And it's just not good enough, especially as you start looking at this world of ai.

And so what I wanted to do is look at this and solve for a problem to build something at the edge. And if you needed to build a solution at the edge, you need cloud management. Well, guess what?

We have intersite, which has a million devices in cloud management. You'd also need really cool hardware that's likely modular, easy to service, all those types of things. We did not have that.

We'd never actually built hardware for the edge. So we set out to build and listen to customers about how would you build this? Where would you put it?

Uh, you know, how can we build hardware that can be mounted horizontally, vertically up against the wall in the corner? How do we build fans that when I'm talking to you, if it were in the room right now, you wouldn't hear it because we've acoustically optimized it to be quiet so it can be around people while it's operating. Uh, we wanted to put a network that was built into it so that you don't have those wires hanging off this thing everywhere.

And then we wanted to build, to be powerful enough to start with things like CPUs today and go to GPUs tomorrow if that's what customers wanna do. And that's exactly what we did from a hardware perspective. And then we wanna marry that upsight where we can manage it all from the cloud and automate it all from the cloud so that customers aren't worried about, well, what's the proper setting for my nick for this particular operating system that I'm running?

Because we know what that is, we've documented it as part of our validated designs, and they can build that into templates, policies and profiles, and they can push that out across five sites or 5,000 sites consistently and reliably. And those are the things that people, as they've been testing the platform, go, cool, this makes sense. Um, and then I think where Cisco can really, uh, continue to add value is like on top of that, taking those inner side pieces, combining with what we're doing with Cisco, uh, you know, catalyst and bringing that up into the overall architecture and leveraging things like AI canvas, because when you're troubleshooting now, I can use AI as a, as a tool, uh, to help me troubleshoot my, my challenges set of funding.

It's, it's very much, you know, the, the idea of, you know, having, uh, infrastructure as code and code as intent, and that that sort of, you know, progression is where I, I see the industry really barreling right now. And so you, you have to step away from the days when you just had a PBX that sat in a closet somewhere with a sticky on the door that said, do not open at peril of your own demise. You know, we have to treat it like it's in the data center.

We can't just treat the edge. Like, it's like you mentioned an afterthought. Yeah, it's, it's interesting to think that, um, you know, of all the companies out there, Cisco has a lot of experience in deploying hardware in really adverse conditions, um, as opposed to a lot of compute companies.

Now, certainly a lot of, uh, the compute providers have, um, been working on this for a while, and a lot of them do have experience. Um, but that being said, uh, you know, I mean, Cisco has had hardened iot, you know, all weather, all environment devices for a long, long time. And it is really neat to see, um, you know, from a hardware geek standpoint, it's really cool to see the ways that they're building, you know, that you build a server that can handle extreme harsh conditions, dust, grease, grime, uh, heat cold, et cetera, vibration.

Um, but let, let's turn a corner here in the conversation as we kind of wrap up. I wanna talk to you, Jeremy, about, um, maybe a little bit of a controversial opinion. And that is that despite the fact that if, if you ask the, the person on the street about ai, they're gonna talk about basically cloud and chatbots and mega data centers and so on, that's really where they think the center of gravity is for ai.

I wonder, and I suspect that you might agree with me on this, I wonder if perhaps Edge is the real center of gravity for AI deployment, and this is really where the rubber's gonna meet the road, not these, um, sort of mega projects, but literally these devices under the counter or in the closet, or, you know, on the manufacturing floor, maybe that's really the center of gravity for ai. What do, what do you, how do you respond to that? Yeah, I think I think about this way in this podcast, you kicked it off, Stephen, about, hey, let's talk about practical applications of ai.

I think the way that people are gonna experience the practical applications of AI are going to be from edge use cases, not necessarily from mega data centers that are doing effectively building out models or building out things that will be then used for things like Edge locations, right? And it makes a lot of sense because this is where customers are gonna experience the world. It's gonna be off their phone, it's gonna be off the things that they're manufacturing and building, and ultimately from an enterprise perspective is how they're gonna drive efficiencies into their business is gonna happen with how they operate their business.

And most customers' business is not building AI models or doing AI research. Yeah, it's, it's interesting, isn't it, how, um, we have learned over the last four years or so here, uh, to to sort of move both toward and away from that mega project that Steven mentioned with Frontier Scale models, that that can only be housed, you know, with a nuclear powered data center. And what people, I think, sadly fail to realize sometimes is that the real power and sometimes the, that long tail of value doesn't always live with the frontier model, but instead lives within the 70 billion, 20,000,000,008 billion parameter model that's doing something important at the edge.

Like watching your tomatoes and making sure they don't, your crop survives. That's critical. Absolutely.

And it, it's funny because those are the sort of tales, um, you know, when you talk to people about ai, um, you know, you get these sort of weird off the cuff anecdotal stories about AI is gonna bring about new advances in medicine and, and productivity and, you know, manufacturing and so on. But they, I think people just sort of toss those things off without really thinking about, um, too much about that, because I think that they're just so enamored with, is it Thinking. But, um, you know, sort of going forward here, Jeremy, um, where, where does Cisco go next, um, with AI at the Edge?

Uh, what should we be looking out for? Yeah, I think what we're looking to do is really kind of close down that trust deficit when it comes to building out AI and, and infrastructure at the edge, where how can we help customers not only refresh this infrastructure for those core computing capabilities that they need at the edge that we talked about, but also be ready to expand into those new use cases that their ISVs are gonna push on top of them in terms of, you know, hey, this is a better way to operate your business and you're gonna maybe need a, a, an ability to grow into that or to leverage some new technology. And those are the big pieces that we built there.

But most importantly, we're wrapping all those edge use cases in with security automation and those types of capabilities into these new platforms like, like Unified Edge, where, uh, you know, they're just getting started shipping. And I would say for us, that's gonna be a, a journey over the next, not only two years, but 10 years, which is what we built that platform to last. And we're really just at the front of it.

So on the edge front, like where we're going, I think we're just at the very beginning of it. And we're gonna continue to work with the industry folks that you would expect, like the NVIDIAs of the world to have more and more hardware components going to this platform, more and more software components to even improve the operation capabilities of IT together today. And more and more integrations with the things across the rest of Cisco's portfolio.

So improve on experience, whether you're using all Cisco type edge locations or even working with some of the third party, uh, solutions that you might have out the edges as well. You know, um, I really loved what you said there, Jeremy, about the trust deficit and the importance of that, uh, to me, that it's not just how you optimize the spend, but, but how you remove risk from that spend. And with ai, as we all know, we're, we're getting much, you know, better at it, but it's not a case of, you know, it's like a new relationship.

You don't step out of the gate after the first date and go, I would trust this person with my life. It's not like With, just to be fair, did you, it worked out really well. Well, that's, it's good.

The odds. Um, yeah, so, you know, when we do our research, we always ask, you know, practitioners, you know, and practitioners in what's the, with two, you know, the vast majority came back and said, the biggest problem is trust in the data, trust in what it's, I have the right data, I'm doing the right thing with it, and that my AI therefore is making the right decisions. Yeah, a hundred percent.

I think the security piece is one aspect of trust. The other side I would throw out there is, you know, just working on a lot of projects with customers is it's the actual trust that they can deliver the outcome. So that's trust in themselves and understanding that if I make this investment, I'll get something back for it.

Because, you know, a lot of times we talk about failing fast and failing in these AI projects, you know, 50%, it depends how you talk to, right? What number they wanna throw out there. But a large majority of the time these things fail, that's actually okay, because a lot of these things are, let's understand, let's fail fast and let's go onto the next thing.

But when you start about getting to these practical edge projects, you can't fail. So, you know, being able to pull that together and lean in and trust that they're gonna build that and then invest in that, I think is really the front end of where we are and where we're gonna be headed is improving that and see more and more of these practical use cases roll out for the next few years Here. Well, I think that's a great, uh, that's a great, uh, kind of reminder of the topic here of, uh, utilizing ai.

I mean, that's really what we're all about. And, um, you know, Brad, thank you so much. Uh, Jeremy, thank you so much for joining us.

Uh, before we go, I do wanna give you a chance to, uh, let us know where can we connect with you and continue this conversation, uh, Jeremy first. Yeah, absolutely. Very, uh, active on LinkedIn.

So you'll see more blogs related to Edge and then what we're doing from a Cisco compute perspective. Um, you'll see got a bunch of great industry events that we'll be at here in terms of GTC Compex, so, uh, we'll be, we'll be out there and we'll be looking forward to your feedback. Excellent, Brad?

com, but I also like Jeremy, like to wander the wasteland of, of LinkedIn, and you can find me there at Brad Shiman, all one word. Excellent. And you'll find me there too, uh, Stephen Foskett.

Uh, also you'll find the Tech Field Day videos, including, like I said, uh, some recent presentations from different parts of Cisco at our AI infrastructure, uh, field day event. com to learn a little bit more about that. So thank you both for joining us and thank you, uh, for listening as well to the utilizing AI podcast.

If you enjoyed this discussion, please do subscribe, uh, either on YouTube or in your favorite podcast application. And also, please consider giving us a rating or review. Uh, it really helps us.

This podcast was brought to you by the analysts and experts from the RUM Group where Insight meets ai. ai, the utilizing AI YouTube channel or to the Text TV app. Thanks for listening and catch you next week.