Transcript: The Road to ECS - S1 E4: Carlotta Castelluccio and Korey Stegared-Pace

hello everyone and welcome to another episode of road to ecs and as valdex says all roads lead to ecs right yes that is exactly the case welcome everyone and uh today for the first time no for the second time for the second time we've got two guests and um Carlotta was our keynote speaker the last year. She had a full room. This was an awesome keynote. Everybody loved it. Thank you, Carlotta. And she was definitely one of the people who we wanted on this show. And another highly rated speaker who delivered stellar session the last year was Corey. So welcome, Carlotta. Welcome, Corey. And love to have you here in the show with us. I think I need to get a better intro than just another speaker guy, but I appreciate it. Highly rated speaker. Highly rated speaker. Five stars. Thank you, everyone. Thank you. Thank you so much at this point for the intro. It was an honor for me, and it's an honor again to be back this year. And I will start with that, actually. It's usually we don't... You know, we don't pay that much attention on what are the ratings of the speakers and whatever, because there are different types of sessions, different types of audiences. So it's kind of, it doesn't play too much in our books. But for both of you guys, it was really impressive and really obvious that audience loved what you had to say and obviously the way how you had to say it. And first of all, congrats from me and from all of us to that. And that's why I wanted you back. No, not only because of that, obviously. But yeah, I mean, you were obviously one of the stars. Two of you were one of the stars of the show the last year. And I know you are coming back. So can we shortly maybe start with Kory, move to Kratos? What did you love the last show? What impression did it leave to you, Corey? Because you were the first time with us. And what are you looking mainly for the next show, for Dusseldorf in May? Yeah, I mean, I appreciate the kind words. I hope we would just stop recording now because I don't think it's going to get any better than all of those compliments you just showered on us. But I mean, you know, I think for me, it's again, it was my first time coming. And I thought, you know, conferences, I think, you know, it's fun to speak. It's fun to present information. But what I found was amazing about this conference is that not only when you get the questions, they are the oh, hey, I don't have any questions at the end or maybe have some time. But when people come up to you and I thought I had some really great conversations after The workshop that we did, we did a workshop that was like three hours. So I thought, you know, I think it was even longer, Carlota. How long was it? It was like, it was all day. Yeah, it was like all day. Yeah. Yeah. So you think they would be like sick of me and hearing of me or just want to run out of the room. But the fact that people came after to ask about more questions and get more information and, you know, that's really kind of our job, Carlota and I, is to kind of just help educate people. developers and technical people uh you know this with this whole new world of generative AI so I think yeah that's I think it was amazing to me is that like you know the audience retention I think that's like the the term uh the the metric so great retention people wanted to to you know continue to engage and ask questions which I thought was great what were your expectations the last year and uh did they match with that what you have seen in this button and uh with the with the vibe, with the feelings, with everything when you came? Yeah, to be honest, I had the warmest welcome from you all, from the organising team. As you said, I got the honour to have a very big stage, a full room, so I was a bit nervous, but with you doing the best for really making myself at ease, it was a... a very amazing experience. And not only for that, also because of the audience, I was really a bit surprised, positively surprised by the passion of the community. I really met a wonderful audience, like a vibrant community. So my expectations were totally met. And as I said, I was also even surprised by the vibrant and passionate community I met. And as also Corey said, the best part for me was really having great conversations with the audience. So the best part was interacting with the audience. For example, we did a couple of panels. That was a really great opportunity to really get insights from the audience, understand, you know, their interest and sometimes also their worries, because we know that AI is amazing, but people are also worried about the power of these technologies. I've already said the word AI, sorry for that. We are fine saying the word AI. I will come to that internal joke in the end. But we can say AI in the show. It's not really a rule. It's something like Valdek tries to push. More like a guideline. No, it's not. Nah. But, Carlota, I mean, you are AI advocate at Microsoft. Similarly, as you see on Valdek's site, he's a developer advocate at Microsoft, at Microsoft. As we know, Valdek is mainly in charge of jokes, which is fine. Is that in the job title? The job description? That's awesome. Yeah. I'm in charge to make you all feel relaxed. We love you, Valdek. I make you laugh once in a while. Can you tell me what an AI advocate does in Microsoft? Sure. So an advocate in Microsoft is, so I tend to describe my job as we are an intermediate team between the product groups, so people who Basically develop the products and services in Azure and especially in the Azure ecosystem and the developer audience, especially the AI developers, AI engineers. So what we try to do is to make the product and services. in a sense that we want developer to make the best of the AI platform and services. So we create content samples to make possible for developer to actually use these services and also understand how they work. And the other part is hearing also the feedback and the insights from the audience. And that's probably why both Corey and myself found very, like the main benefit from the event last year was really having conversation with the audience because it's food for our thoughts and it helped us create new content. It helped us refine products and services. And we don't want just good feedback. We want feedback. the good the bad the ugly we want everything because you know the best feedback make us improve the content products and so yes that's how i describe my job uh we are really the intermediate step between the developer audience and the product group corey what's your take on uh this Oh, okay. I have to back that one. I tell people, so I fill gaps. So that's what we do. So there's gaps between our product teams and the developers or the users of our services, because our product teams, they can't speak to every single developer. I mean, I can't either, but we can try to scale what we can do. So I try to fill those gaps. between those communications. Then I also tried to fill gaps between the community and the technology, right? So there's so many new things out there and the developers have gaps in their knowledge and I'm here to sort of fill those gaps. So I just tried to take spaces wherever there was needed, internally and externally. And if you think back to the last ECS, is there one thing that stuck to your mind? Like, oh, I heard that thing and it blew my mind. I mean, you are in AI space, so maybe there's not much that blows your mind anymore after all the models and advantages being shipped every week. But maybe there's one, I don't know, story or something you heard or something somebody said they did. It was like, oh, that's so cool. Carlota, do you still have a mind to be blown? Every day a little bit less, yes. I mean, I think it's, you know, for me, you know, we try to build samples. We try to talk to developers, understand what they're actually doing out there so that we, you know, again, provide the content, fill the gaps, as I said. But you never know what people want to actually do. So I wouldn't say it's like a... mind blown thing. But like, you're just hearing people's use cases, even this might sound boring, or maybe I don't know the audience, but like, you know, oh, I need to connect all these like different SharePoint instances and things like that. Just like the realities of people once they get like, hey, you need to build something with generative AI, like, you know, from their manager or the business side or whatever. And then they just get that mandate, and then they have to go out and do it. And then they're like, don't know so just hearing those kind of war stories i think is very interesting um because like you know they're just trying to find all the pieces uh and parts and connect them and and then you know the expectation is that some sort of thing magically is going to happen uh but you know it tends not to be the case that they need to still you know get the education get the training and stuff like that so i think that's probably mind blown is just like seeing how people are actually implementing this stuff and you know If they're doing it in the quote unquote right way, like an iterative way, or is this like go? And I think there's a lot of in between, I would say. I need to jump into one question because you are way more out there. I'm not doing it consulting. I mean, that I am going to customers like since. Two thousand fourteen fifteen. So when you tell me there are people that actually out there who get told by their manager build something with generative AI. Is that really the thing? And don't get back until it's done, right? Yeah, yeah, yeah. Don't get back. Yeah, I mean, of course, I mean, I think it's just, I think, I mean, I'm not going to like say why they do. I don't know why they do it other than, you know, it's maybe something exciting and they think there's a good use case for it. I mean, I've heard, especially in the startup world, particularly, you know, venture capitalists out there, they want to obviously make sure that they're, startups are positioned in the right way and utilizing the latest and greatest technology. So they also say, go look for features, look for your abilities to implement this in something. Because you'll ultimately get the customer's eyes, things like that. So yeah, I mean, I think people, they do get those requests. Whether they're nice about it or not, I don't know. Maybe I'm too old. Back in the time, we would have... a requirement, then we would search, look for the best tech to fulfill the requirement. The requirement is build something with AI. I need something. As a manager, I need something with AI so that impact. That's... Now, these days, that's a requirement. Like, it's like a mark... Most cases, like a lot of cases, it's going to be a marketing gimmick, right? Where it's just say, we use AI. So I need something AI. That's the only requirement. I don't care what it is. It needs to be AI, right? Well, but I guess on a different note, right? Like in all seriousness, I think that there is also a room to, you know, challenge people. Hey, like there is this cool tech. Try it. I don't know for what. I don't have the answers, but try it. Give it a try. Because like oftentimes when you get to play with things, Only then you only start to wrap your head around like, oh my God, we could do X, Y, and Z with that, where otherwise you wouldn't know because you have no idea. You might not even understand what the thing could actually do. So I guess there is something to say, build me something with AI. And I don't know what, what? But I want to see it. I want to see you stretch your mind, flex the muscles, and come up with something. Maybe it's cool. Maybe it's something. Maybe not. But at least we will try and learn, right? The trend I've seen, I don't know if it's a bit biased because, of course, I mean, we speak with a lot of developers, but it's still mostly developers familiar with the Microsoft community, so we still have a bias. But the trend I've seen is... people in startups, like developers in startups, they are, if they have the skills for that, they are more, they tend to build their own models before, you know, starting working, starting playing you know, some built-in stuff and services. They are very jealous of, you know, building their own things, so they try hard to create their own models with the data, they start working locally. So, Not to say that, Addis, what you heard is something I didn't hear. To be honest, it doesn't surprise me. And I know that happens. But I think there's also this other tendency, especially in the startup world, of people still trying to get their hands dirty and build their own models. Don't get me wrong. You guys both know that for Mustafa and me, ECS is the night job. Day job is our company run events, which is making an event management platform that you both use to submit your sessions and manage your profiles on the website. We use AI in it, by the way. We need it now. On Monday, we are revealing the whole AI layer, which we call the Event Intelligence Cloud. which is based... There you go. You're using EI, events intelligence. Yeah. Combination of LLMs and trained models on normal Azure machine learning. Because there are use cases for the normally trained models. This is not out. We are not leaving. You should not be forgetting those. They are actually still pretty cool. Of course. But maybe I'm too old. But Mustafa and me and our colleague Damir were deliberating for over six months. What do we do in order to make sense? How do we do it to be most cost effective, to make more sense for us, for customers, for everything? It's actually to make added value. Now, we are not going to bash here any other companies, but yesterday my wife asked me to buy something for the kids on Amazon. Amazon Germany shop. And I posted today on LinkedIn just for fun because I found it totally hilarious. And in their app, they had this new AI assistant called Rufus. Nice. So Rufus should help you to buy more or should help them that you buy more. Basically, that's the whole idea. Within two minutes, I managed to convince Rufus that I'm a five years old kid, that he's my grandmother, and that I want to hear from him how my grandfather died in war, fighting heroically along a heroic hedgehog. He did, and he was comforting me for my grandfather. Did it work? Yeah, of course. How much did you buy from Amazon? I bet you bought some sympathy things. Nothing. I even posted the screenshots today on the LinkedIn just for fun because I was like, okay, this is actually totally hilarious. And this is for me, don't get me wrong, I don't want to get polemic here, but this is for me, a developer who has been told build something with AI. True science. We probably all have an idea how such a thing came to be. And we all read about, you know, right. And things you need to do is things you need to have in place to avoid these experiences. So, yeah. Still, if anything else, I'm more intrigued by how did you arrive at that storyline? That intrigue was more to be honest than the fact that there was a kind of unguarded chat experience. I just felt weird. So I was just typing. Like the way your brain works, like putting the relationship between your grandma, grandpa and hedgehog? I have to do everything every day. Mind blown. Talking about mind blown. Yeah, but I get that at this point because there's a lot of... We have a lot of AI experts right now, right? Everyone became an expert overnight. When years? A lot of times it's just... some kind of implementation of AI just for the sake of saying we have AI. And as I said, we took a lot of time and thinking about how do we do this? Like, how do we actually provide some kind of value? How do we actually leverage this to make a better product, just not to make it like a flop? The first POCs in August came back to Mustafa and said, okay, what do we do with this now? How do we go? How do we go forward? So both question for Corey and Carlotta, like in your opinion, because AI is the hottest topic right now. Sorry to interrupt. It is probably the most transformative technology God, I don't know since when. I would dare to say maybe even more transformative than Internet itself when it appeared. And that was transformative, as we know today. That was totally transformative. So we have got something really awesome in our hands in so many ways. But it's the same with the internet. We could say we could build the world by database from A to Z, or we could build, I don't know, some silly stuff that you find on internet today. Cat memes. Yeah, there you go. Thank you, Corey. You're always in a better place because of cat memes. Hear me out. My question for both of you, because you're in the field, you're talking to developers all the time. how how much of it is overhyped and how much like it's real use case scenarios and and where are we heading with all of this like what do you see as a future where where will ai go next and what will our audience in düsseldorf learn about ai oh nice I will try to go first. I think the massive impact is that it's something that regards everyone. It's not just something that regards people in technology. I mean, because when I talk with my friends who are not in technology, who are not developers, who are not engineers, they are still very excited about AI. They use AI in their mobile phone, in their daily life. And I've seen also, I think every knowledge worker right now uses AI in their own job for, I don't know, helping them with the refine their emails or as they use, I mean, AI notes in Teams meeting, how they are, how helpful they are. So, and this is not something that only people in technology in tech use, it's something that every knowledge worker, I think, sees the benefit of using AI, even in easy tasks in their daily job that could be automatized. So that's the huge impact I see. It's not just us in a bubble. It's not a technology for jigs, it's a technology for everyone. Corey? Yeah, I mean, I don't know if you want like a percentage of what's hype. I was really excited. I already had a number. Can we all give our numbers out and see if they all connect, if they're all right? Yeah, yeah, yeah, totally. So I'm going to take my phone to write my number down and then... Yeah, well, I mean, I guess to answer your second part of the question, then, in terms of where it's going, I mean, if you say, I mean, you had just said, like, you know, this is bigger than the Internet. And, you know, I think it's even bigger than, like, you know, if we talk about mobile and all these other things, like we also had these, like, transformative technologies. Like, look, that's really transformative because literally everyone's on their phones all the time, for better or for worse. I think if you take both of those experiences, it's this need that we always need to have some information or stimulation in terms of finding something or getting something answered. Then you have the Internet where it has all these answers. The world, all this data is really hard for humans to process and find. I think if we're talking about where we're going next and where I think a lot of the industry has been talking about for this year at least is the year of agents, AI agents, and how AI agents can take not away the stress per se, but allow us to traverse all this world of data and tools and services and systems that we built up in the past because of the Internet and because of mobile. And I think, you know, that's where we're kind of going, where we could just send a request and a whole all these other actions, whether it's, you know, I would like to join this podcast. And, you know, you have to get my availability. You have to write an email. You have to make the stream your own link. Like these are all things that the Internet has given us, which are all great. But how can we like just make that into one thing? make a podcast episode and have all those tools and services sort of communicate through an AI or AI agent. I think that's where ultimately we're going. Right now it's still very early, very infancy. So you'll see people talk about agents and you're like, okay, that looks like an API request. And most of the time it is, right? But how many API, Chris, can you string together to make something really valuable? And I think that's at a very small point of where we are now, and then we're kind of getting to further places. Yeah, that was last, what you last said is exactly where I was heading to, like very often just an API request. Sorry, we are making this totally a geek talk, but let it be. Okay. I'm circling back to my original question. So in four months, we are all going to meet in Dusseldorf. It's only four months, believe it or not. It's only four months. Wow. Total panic. And... Obviously, prevailing across all three conferences we have. Obviously, with Collab Summit, a lot is happening with Microsoft Office and everything. This is really space that I sometimes watch in amaze what people are doing on the M.S. Office side. And I happen to be friends with quite some of them and hear from the first hand what's being done and I'm like, okay, you guys are totally awesome. You too, Baldek. Obviously, what's happening on the Azure side with all the AI services, Microsoft is now with AI Studio, with OpenAI being part of Azure and how this is all developing, and especially the whole topic of semantic kernel and everything which is going on semantic kernel, we are obviously seeing incredible things happening there. And also on the BizApps side, because integrating local, no-code apps and Power Platform with AI has a lot of great promises. We're obviously going to have forty, fifty Microsoft speakers in Dusseldorf. You guys are going to be among them, all three of you, Valdek and Kalota and Corey. And what can you say to our audience? What are they going to hear, to learn? and to feel, because as Corey just said, there's also one important thing, it's that people kind of get feeling, it's transformative, you sometimes need just to get feeling on where is it going, because it's changing so fast, what's today, it's in two months can be totally different. Now we are talking about a year of agents in six months. I'm not even sure we'll be talking about agents. Maybe we'll have something totally new. We just emerged within those six months. Nobody of us cannot know that. Like all things, no, everyone forgot about. Not even deprecated, but we will just find a better way, which is perfectly fine. So what can you say to the people who will actually come to Dusseldorf? What are they going to see, learn and feel there? Who wants to go first? Not everybody at the same time. Take your turns. You, Baldik. Me? Oh, really? You want to start with me. Perfect. So you asked about what they will hear, see, and feel. Is that the case? Yeah. And taste, if you want to add all the senses. And taste. So what they will taste, I leave that up to ECS. They take care of the catering. What I want people to feel is optimism. I want people not to feel threatened by change, but embrace it and see what kind of good things it can bring them. And genuinely give it a try. Approach it with an open mind. Try it. Be open to it. Listen to things and think. Because I think there's a lot of potential in it. as long as you're willing to embrace it. And I think everybody who joins us in ECS will hear and see the latest and greatest of what we have to offer at Microsoft, what technologies are available and give you some, I hope, ready to use tips that you can apply to your job and try things and experience them and basically take it from there, apply to your job and see that, experience that change by yourself. Carlotta? Yeah, I go next. So what I would love people to feel is, I would say, I would use the term connected, like in a remote world, global world, like the one we live. I think this kind of in-person event are very... are awesome because you really feel connected to a community. As I said before, last year I was amazed by how warm and passionate and vibrant was the community at ECS. So I really encourage people to come to ECS to feel connected. I'm sure they will feel like that. And in terms of what they are going to hear and see, for sure they are going to see, to hear and see the latest and greatest, as Waldex said, technology in the Microsoft ecosystem and maybe some of the open source ecosystem as well. And hopefully we are going to showcase a bit also some real-world sample to all these kind of technology and this wide offering of technology. So my hope is that we can clarify a bit in this, you know, wide world of technology different technologies and offering. We want to guide the audience on what to use when, what benefits every technology has and how to use them in practice. These are both incredible answers. Really hard to come back from this. I say feel would be confidence. So, you know, people listening to this will probably hear like, you know, when they hear some story about someone's hedgehog grandpa who died in World War III, they might say, I don't want to ever implement generative AI. Like this is going to be scary because my users are going to find a way around, you know, the intended purposes and things like that. Or, you know, you know, you might have a manager that comes in and says, you know, genitive AI now, and you don't know where to start. I hope, you know, people, when they come to the conference and hear our, our talks that they come, they leave with some level of confidence that a one that, you know, they have an understanding of whatever percentage of hype and reality genitive AI is and how they can actually start to implement that in, confidently into their whatever applications that they're doing, either by choice, by force, or just for fun. So I think confidence is probably what people to feel. And what they will see, I think, is, you know, I think we at Microsoft have just a unique opportunity because I hear about different teams doing tons of different things and what works and what doesn't. And it's fun for me because then I can just like be the sponge. And then like whenever I come to a conference like this, like squeeze it out and say, oh, you know, this team did this or this team did that. Because, you know, you're not always talking to one another in terms of like, oh, yeah, we implemented this and this didn't work or it worked this way. Like people are kind of secretive sometimes about what exactly is going well and obviously confidentiality. But yeah. um you know so i'm able to you know be kind of that that resource and i think a lot of people at microsoft and especially in the advocacy side are as well is to kind of uh and we're able to show uh you know from our experience in the communities and hearing stories you know what is working for teams out there and share it out so yeah i think that's going to be great and i don't know what they're going to taste but hopefully good coffee because yeah there will be a good copy now for me as somebody who started this conference twelve years ago it's important because everybody's um everybody's in a state of we don't know what comes uh next we are not sure we don't we maybe don't know what would be the right way we we feel sometimes maybe even a bit overwhelmed with everything and i just like like to ask everybody when you are at the conference Mirjana Cizmovic- Talk to Carlota, talk to Valdek, talk to Corey, they are the people who can actually help you navigate that maze, talk to all the other hundred or two hundred and twenty speakers. Mirjana Cizmovic- Because we are. Mirjana Cizmovic- We cannot hide and we should not hide from this, what is AI bringing to us and ECS is definitely the place where. There are people who can help, who are willing to help, and who are, as Corey said and Carlota said, who are willing to learn from you what did work, what did not work, what's the good feedback also to Microsoft. And I think that what I hope is that feeling of... Clarity might come after the conference and say, oh, those are the ways we could implement it. Those are the paths we could walk. And now I get impressed, I get pumped up, and I'm actually going to build something without my manager telling me, here's a new toy, play with it, build something. I actually hope for that for most of the part. We all know, right, that some folks find it easier to talk to others than other folks. And that's something that we've been asking to all our guests. So to you folks, say somebody comes to you at ECS, what would you like them to ask you, Carlota and Corey? Oh, wow. Is this like a personal question or like about the time? Whatever you want. Because like you are there, right? And people might not know you yet. So what's like the first thing they can ask you to kind of break the ice, let you, you know, nerd snipe you if you like? Basically, like, you know, like, what's your thing? Like, what's your thing you're passionate about that folks could ask you about to connect with you and to kind of, you know, break the ice? What I love hearing from people, from developers, from people attending these kind of tech events is what they are doing with AI. how they are using AI, maybe Azure AI services or other technologies in their own company, startups or whatever. I'm really interested to know that because that's the first bit. When I have some kind of context, I'm then open to understand how I can have them answer questions. So the first thing for me, I would say, is really understand their stories, what they would like to achieve, what's their project and the objective. Cool. So I'll start by saying Microsoft hires very different people, different minds. So this is why we're so successful. And I would say probably the opposite of Carlota. Like I'm a little bit more like a little debate, a little bit of argument. So I would love for you to come up to me and tell me like one thing that you either not hate about derivative AI or you think is ridiculous or maybe why you haven't used it. I would love to hear those stories or even like horror stories or anything like that. so that we can have a really good conversation. Hedgehog, the grandpa. Yeah, the hedgehog grandpa. Give me those stories so I can use those as a starting point for our conversation. That's cool. And by the way, the weirdest question I've had last year was a job proposal, like a job offering. The weirdest. That's a low-key, humble brag, Carlota. Yeah. Who would you like to work with us? Who remembers my face after I've got this question? I mean, I was on red, of course, but it was so weird. It was weird that no one asked me. No one asked me to work for them. That was the weird part. I mean, there is a difference. You'll talk about it. Well, I wasn't there either, so that's kind of... My problem is on my conference, I actually have no time to follow the sessions. I did see bits and pieces from Carlotta's session because I was there for the keynote. I haven't seen your sessions, Corey. Sorry, because I didn't see any other sessions. Did you watch the recordings? Valdek. It's all good. But... I was just, because we had this topic of AI also the last year, but obviously this year it's just peaking. It's just peaking everywhere. Not only us, but I was really, it was just so impressive how much did your session stand out of both of you. So people obviously wanted to hear those stories. People wanted to hear those topics. It was obviously very important for everyone to hear that and people also appreciated it. So there is exactly as I said, you both of you were very highly rated speakers, all of your sessions or your participation. We definitely wanted you back. So you're being back at the speakers there. I love the fact I'm not sure who mentioned it was a lot or Corey, but someone mentioned that they love being part of the panels. as well. So we're going to have a couple of AI panels, and I'm happy to make a note. You'll be included. There you go. Put your hand up for that one. And also, also part of the last year was AI workshop as part of the pre-comp day. We are doing that this year again. You're going to be part of that as well, together with Lee Stott. Now, speaking of that, we need to do something like special hello for Lee With a special phrase that we always do, right? I will count down from three to zero. Lee, I just want to say they made me do this as part of participation for this podcast. Just say computer intelligence. And it's just for Lee. Three, two, one. Computer intelligence. That was just for Lee. Hey, Carlotta, Corey, people, this was awesome. Thank you for being with us. Thank you for being in this webcast. And I'm really totally looking forward to seeing you again in May. I will again not have time, but I will be very happy to see you. It's just what it is. We are looking forward to meeting you all and to joining ECS. Thank you. Valdek, Mustafa, my brothers, see you in the next episode. See you soon. Thank you all. Ciao. Ciao.