Having AI FOMO? Practical Tips for Logging Real AI Wins in Your Business for 2025

Timestamps by PodSqueezeIntroduction and Guest Introductions (00:00:00)
Julian introduces the topic, guests, and sets the context for the discussion on AI adoption.

Sean’s AI Adoption Journey Begins (00:01:05)
Sean shares how and why his company started experimenting with AI in early 2023.

Early Realizations: Company-Wide Involvement & Security (00:03:37)
Sean discusses the need for broad adoption and secure AI platforms within the company.

Kevin’s Background & State of AI Adoption (00:05:25)
Kevin introduces his experience and comments on how early Sean’s company was in adopting AI.

AI Literacy and Experimentation (00:07:16)
Kevin explains the importance of hands-on experimentation and literacy before large-scale AI projects.

Industry Maturity and Training Needs (00:08:31)
Kevin describes the immaturity of the industry and the need for training, literacy, and security.

Accelerated AI Adoption Today (00:09:26)
Discussion on how organizations can now adopt AI faster due to improved tools and resources.

Evolution of AI Models and Hallucinations (00:11:56)
The group discusses improvements in AI models, reduction of hallucinations, and the creative aspects of AI.

Championing AI: Leadership’s Role (00:14:08)
Sean emphasizes the CEO’s role in championing AI and the importance of persistent communication.

Operationalizing AI: OKRs and Forcing Mechanisms (00:15:55)
Sean describes using OKRs to drive AI adoption and operational use across departments.

Hitting a Plateau and Bringing in External Help (00:17:02)
Sean explains how their AI adoption stalled and how bringing in Kevin for training reignited progress.

Training and Middle Management Engagement (00:17:59)
Details on the workshop with Kevin, involving middle management and influencers to spread AI skills.

Bridging the Skills Gap in the Workforce (00:19:10)
Kevin highlights the gap between leadership and workforce AI skills and the need for deeper training.

Impact of Training: Renewed Momentum (00:21:05)
Sean shares the impact of Kevin’s workshop: opening eyes to possibilities and building practical skills.

Practical Example: AI for Resume Screening in HR (00:22:16)
Sean gives a real-world example of using AI to screen resumes and the process of upskilling HR staff.

Organic, Bottom-Up AI Adoption (00:24:16)
Sean describes how AI adoption became organic, with employees initiating their own AI projects.

Culture of Experimentation and Mandates (00:25:06)
Kevin and Sean discuss the importance of a culture that encourages experimentation and mandates AI adoption.

Rethinking Processes, Not Just Automating (00:26:12)
Kevin warns against simply automating old processes and stresses the need to rethink workflows.

AI as a General Purpose Technology (00:28:56)
Kevin compares AI to electricity, emphasizing its broad, transformative potential and the current supercycle.

Addressing Employee Fears and Upskilling (00:32:49)
Sean discusses overcoming job loss fears by positioning AI mastery as future-proofing careers.

Top-Down Enablement and Support (00:34:13)
Julian summarizes the need for leadership to prioritize, enable, and support AI adoption.

Bottom-Up Enablement with Guardrails (00:35:01)
Kevin stresses the importance of AI policies, security, and practical training for safe, effective adoption.

The Trough of Disillusionment and Persistence (00:37:47)
Discussion on the hype cycle, frustration with tools, and the need to persist through early challenges.

Keeping Training Up-to-Date (00:39:19)
Kevin explains the challenge of keeping training current and the importance of practical, hands-on learning.

The Pyramid of Suck: Where to Start with AI (00:40:14)
Kevin introduces the “pyramid of suck” framework: start by automating tasks people dislike.

Personal Productivity as the First Step (00:43:47)
Julian and Sean agree that focusing on personal productivity is the best starting point for AI adoption.

2025: The Year of the Agent (00:45:42)
Kevin discusses the rise of AI agents and the growing ease of building complex, automated workflows.

Embedding AI in Client-Facing Products (00:47:23)
Sean explains how his company built a dedicated AI team to embed AI into their product suite.

Scaling AI: 2000 Custom Agents (00:51:04)
Sean describes creating 2000 custom AI agents for clients and the process of scaling this capability.

Rethinking Workflows and New Capabilities (00:52:00)
Sean shares how AI enabled new workflows, such as SMS response agents, that weren’t possible before.

Common AI Adoption Mistakes in Startups (00:53:00)
Kevin lists common mistakes: expecting 100% solutions and building tools unnecessarily instead of buying.

Keeping Teams Upskilled Amid Rapid Change (00:54:53)
Kevin outlines approaches to ongoing training, including workshops, annual programs, and bootcamps.

Final Takeaways and Closing (00:57:10)
Sean and Julian summarize the key message: AI adoption is approachable, requires leadership, and is achievable for all organizations.

Speaker 1 00:00:00 Okay, we’re ready to go. Yep. Ready to rock and roll. Okay. Well, good morning everybody. Thank you for joining us on this Growth Elevated webinar. Today we’re going to be talking about getting started with practical steps in AI and logging wins in 2025. My name is Julian Castelli. I’m the host of the Growth Elevated Leadership Podcast and super excited to be joined today by Kevin Williams and Sean Miley. they Kevin is an AI expert and adoption consultant. And Sean is the CEO of my advice. And they have been doing some work in AI that I’m super eager to learn from. And I hope you all, you all are as well. Welcome. Kevin and Sean.

Speaker 2 00:00:47 Thanks for having us. Thank you.

Speaker 3 00:00:49 Yeah. Thank you. Julian.

Speaker 1 00:00:51 Hey. So, guys, we talked before, and I’m super excited about about, sharing your story with the community. Sean. You know, where were you when? When you were starting this journey? I don’t know, a year, year and a half ago, you know.

Speaker 1 00:01:05 How were you thinking about AI and what caused you to take a step forward? Tell us. Walk us through the little. A little bit of the story. Sure. I’ll do that. By the way, tell me one you’d like me to to cue the slides.

Speaker 3 00:01:17 I will do that. So I think it might be helpful for the audience to understand a little bit about, our company and what we do and what size we are, because I understand there are different size companies listening to this podcast and different sea level executives. But basically, for context, you know, we have marketing software for small businesses. We also adjunct the software with some services, and we’re about 175 employees. I have a CEO. I have seven direct reports, of which five of those I would consider to be senior executives. And one of those senior executives, a guy named Brian Burdett, who is our VP of engineering is also important to this story. So Julian, it’s interesting. You know, when I started to sit down and, you know, get my thoughts organized for this podcast this morning, I realized that we’ve actually been at this for a little over two years.

Speaker 3 00:02:12 I think it’s a 12 months. I think it’s 11 months or 18 months. And then all of a sudden, I realize we actually started this journey in April or May of 2023, and it’s now June 2025. So it’s it’s really interesting.

Speaker 1 00:02:26 That’s that’s earlier than most, I think. When, when did ChatGPT come out? Kevin, do you know?

Speaker 2 00:02:31 Yeah. Right at the end of 2022. So, April is very, very early.

Speaker 1 00:02:36 Wow. So so what what what what put you on the front foot, Sean?

Speaker 3 00:02:40 I think it was really the fact that Brian Burdett, our again, our VP of engineering, who I’m going to refer to as Bebe, I think he and I recognized pretty early that this is probably the most significant technological development of our lifetimes, and that probably made it the biggest opportunity of our lifetimes. And I think that now, looking back on things, I think we actually underestimated how big, big it was. It’s probably the biggest development in the last two centuries, maybe more.

Speaker 3 00:03:09 It’s just, you know, the possibilities here are amazing. So, Julian, we started really simply. We actually started with a very small team playing with individual subscriptions. And that team included me. I am not a technologist by training. I come from a financial background.

Speaker 1 00:03:28 You put your hands on catchy beat. Okay. So it was right. It was right there you were the early days when people couldn’t even spell it. You were. You were in there testing it.

Speaker 3 00:03:37 Yeah, we were and we were using we started with the free version and then we, you know, upgraded to the $20 a month personal version. But we really quickly realized two things. The first one in both of these things are really critically important. right? But the first thing is, we realized that it wasn’t enough to have a small group of people at the company trying to manage this, that this had to be a whole company effort, and it really has to be a whole company effort, because at 175 people, nobody on the senior team is close enough to the actual day to day work getting done.

Speaker 3 00:04:15 To understand all the potential uses of AI at the actual employee level, the employee who’s doing the work. So it’s really important that we try to distribute the technology throughout the organization. That was the first thing. The second thing we realized very quickly was that we needed a secure platform. And what I really mean by that is that if you’re using the $20 a month personal subscription or the free version, you’re actually helping to train the public model. We have confidential business information and confidential information on our clients that we weren’t comfortable Will allowing the public model to learn from. So we wanted our own firewalled version and we went out and set that up. So we have a private firewall version. It’s the team’s version, and we made it available.

Speaker 1 00:05:03 To signing up for teams. Or you have to do anything special for that. Sean.

Speaker 3 00:05:07 No, it’s really not nothing special. You just sign up for teams and add people as you need them added. And the way we’ve got it configured, you know, anybody can go in and make a GPT and make it available to the whole company, make it available to a subset of the company.

Speaker 3 00:05:22 It’s really powerful and it’s well structured.

Speaker 1 00:05:25 Awesome. Sean, I want to come right back to you. But Kevin, talk a little bit about sending AI labs and you know, your background for the audience. And then I want want to ask you to put in context based on, you know, Sean, starting two years ago, how early is that compared to, you know, the norm that you see across your clients?

Speaker 2 00:05:44 Well, the norm is still getting started. So it’s a way, way, way ahead of the curve. I’ll say that that certain industries. Saw the existential crisis coming and sitting in a marketing based technology organization. Shawn realized pretty quickly, I’m not going to want to speak for him, but pretty quickly that the nature of that industry was going to change. So he leaned into it really fast. If you’re an industrial company, maybe less so. Right. Sundial Labs. so my own background is in high velocity digital marketing. So I used to be a direct consumer brand owner, and I built a few brands, into decent, global scale.

Speaker 2 00:06:26 And I was spending, you know, eight figures on digital marketing and as digital attribution, the way that digital marketers tracked people across the internet became more and more complex. I had to learn, lean more and more into machine learning based approaches to make sense of all of that data. So I was basically doing AI before I was Cool. I had the fortune to exit those companies. What feels like five minutes before, Mr. Altman dropped all of this on us at the end of 2022. And I was definitely paying a lot more attention than most. But I found myself in the right place, right time, with the right amount of time and the right expertise. And I just dove in to, being the AI guy. And, I started off trying to solve everybody’s problems.

Speaker 1 00:07:16 I’m glad you said that. That’s what. That’s what I describe you about sometimes. Kevin and I always feel like, you know, guilty about just being so colloquial with it. But you’re my guy.

Speaker 2 00:07:27 The guy? Yeah, I could do anything.

Speaker 2 00:07:28 I’m a Swiss army knife.

Speaker 3 00:07:30 he’s our guy, too.

Speaker 1 00:07:32 Yes.

Speaker 2 00:07:33 So the the industry is still super immature, right? So when I dove in and I was trying to solve people’s problems, they weren’t really ready for their problems to be solved because they didn’t really understand what I was talking about. So I’ll hit this many, many times. But really, Sean’s diving in and experimenting with GPT early. That was a form of literacy, like experiencing the limitations of the technology, the frustrations of the technology, and some of the wins such that he could iterate through that if instead he had jumped right to some sort of giant product development change or product roadmap change, it probably wouldn’t have worked out because internally his team wouldn’t have had the the.

Speaker 1 00:08:17 They wouldn’t even know what you’re talking about, right?

Speaker 2 00:08:18 Yeah.

Speaker 1 00:08:19 Yeah, exactly. It’s all changing so fast. I feel like I’m on the literacy path now for two years because it’s every week it changes.

Speaker 2 00:08:27 Yeah. Everything changes. Just keeping up with it is frankly, exhausting.

Speaker 2 00:08:31 But what? Ascend? Ascend has basically two wings of the business. One is training and literacy and increasingly a focus on AI security, risk and governance. So we run training programs, workshops, courses, things like that that really bring people up to speed. And then on the other side, we also have a consulting and implementation business that attempts to identify work, work, work flow issues and provide guidance on how to solve them or to how to how to solve them ourselves.

Speaker 3 00:09:04 Hey Kevin, I have a hypothesis here that I would like to run by you because I think it would be beneficial for the audience. You know, we’ve been at this for a little over two years, but I think if someone is just starting now and they decide that they are jumping in with both feet and they’re going hard, I think they can do it faster now than we did it. What do you think?

Speaker 2 00:09:26 Oh well, 100%. There is a there is something that has happened even in just the last four months, which has been really eye opening.

Speaker 2 00:09:34 So when I speak about AI, one of the things we talk about is the training window. And I’m not going to get technical on it, but essentially the I didn’t understand itself or LMS didn’t understand themselves because their training window lagged really far behind the actual information that was out there. So a new feature might launch on GPT, and you can be you could say, hey, GPT, tell me about deep research and tell me how to use it. And it would basically toss up its digital hands and say that it has no idea what you’re talking about, because that information wasn’t in its training data. What’s happened more recently is the addition to of search, which we’ve actually had searched in the LMS for a while, but search has been incorporated into contemporaneous training nodes not to make it sound like super technical, but basically relevance of contemporaneous search is closer to the training patterns of before. So you can now, the reason that this is so much easier now is because you can stumble into, say, an agent platform like n a n file that away size n a n is an awesome agent builder platform is fantastic, but it was really hard to use six months ago because it was still really wonky.

Speaker 2 00:10:50 But now, since it has a GPT enabled assistant that has contemporaneous information, it can actually stay up to date with the latest and greatest tactics and trends and guide you on a much more granular level. Whereas before you were sort of left in YouTube land. And I will admit that that that I have a lot of young staff and I laugh that that I’m the boss who, like, tells their young staff to go watch YouTube so that they can keep up to date on things, and they should tell their parents that all of those YouTube hours actually pay off at the end.

Speaker 1 00:11:22 Of the day. It is. It is amazing.

Speaker 2 00:11:24 But it is easier to round it out. It is easier now because the tools themselves can tell you how to use the tools. There’s still a frustrating a little bit of a gap. You still have to bring up your skill set, but you’re not flailing in the dark and hitting a wall. Now you can break through that wall and get practical solutions, which is going to be a huge accelerant for people who are getting into it now.

Speaker 3 00:11:47 For sure, there’s more general knowledge around it. There’s more understanding. You can come up to speed quicker. I think it can be done a lot faster now. Absolutely.

Speaker 1 00:11:56 And there’s less. The models are better, right? There’s less hallucinogenic hallucinations. I remember when, you know, when I dove into it 18 months, two years ago, I remember doing a few things. And I heard a topic the other day where someone said that AI has moved from being a clever parrot to actually being, you know, an intelligent, you know, a contributor. And I remember getting a little discouraged. Discouraged by the clever parrot at first. Yeah.

Speaker 2 00:12:23 yes and no. I mean, you need to remember that hallucinations are actually a feature and a bug in these systems, and that’s really important for people to recognize that if you were to, you can actually train hallucinations out of lmms. The problem is that it’s all based on statistics, basically. And if you were to train the hallucination out, it would snap to the highest probability result.

Speaker 2 00:12:48 So if you’re on iMessage and it’s suggesting the next word it it would do that. It would just snap to that most likely word, which takes away that feeling of creativity and depth that it has. Now what’s happened is the models are now layering on top of each other, and they’re adding what’s called inference. So they’re taking more time to consider their results. So it spits out that suggestion and then it runs. It basically assesses that suggestion and sort of gut checks it the way a human would likely do if they were reviewing somebody’s output. And that’s help help limit hallucinations. There are also a lot of other pieces of technology that are out there, such as Rag, that help control hallucinations as well. Do walk carefully because hallucinations are not going away. But guess what? All three of us hallucinate all the time too. Especially as we get older.

Speaker 1 00:13:41 So when we get older. Right?

Speaker 2 00:13:42 You know, it is sort of.

Speaker 3 00:13:43 Like.

Speaker 2 00:13:44 Let’s make errors.

Speaker 3 00:13:45 They’re called fantasies.

Speaker 4 00:13:47 They’re called fantasies.

Speaker 1 00:13:49 All right. Well, look, I’m going to I’m going to do the tldr. The tldr is because Sean started early and because he made it a company priority. That’s why we’re featuring him on this podcast. And so, you know, Sean, there’s people here listening. They want to understand what they can do and learn from your example. So please walk walk us through kind of, you know, your journey here.

Speaker 3 00:14:08 Sure. Absolutely. So we started out with the very early version about two years ago, and how we very quickly realized that everybody needed a subscription, that the whole company had to be bought in, and that we needed to secure a platform. But giving everybody a subscription in a secure platform, while it’s a necessary condition, it’s not sufficient. Right. Just because you give somebody a tool doesn’t mean they’re actually going to use it. And so really that’s where the role of championing comes in. And this to me is probably the absolute most important piece of it is we as senior executives and especially as the CEO, because the CEO has a voice like nobody else in the organization.

Speaker 3 00:14:50 You have to be the champion behind us. I am not from a technological background. I’m from a financial background. Yet I dove in and learned what I could so that I could be an effective champion. And that CEO’s voice is critical. You know, I had the privilege of hearing a guy named Ron Williams speak in 2019. So Ron Williams had recently retired as the CEO of Aetna. And Aetna was a fortune 500 company. Still is. And under his tenure, it had done extremely well. But one of the things that he said in this presentation that has stuck with me forever is that as a CEO, you have an obligation to repeat the same thing Over and over and over again until you are sick of repeating it. And then when you’re sick of repeating it, you have to repeat it over and over and over again. Because when you say something, only about 20% of the people in the organization actually hear what you want them to hear. And of the 20% that hear that, most of them forget it almost right away because they’ve got their jobs, they’ve got their family problems, they’ve got all this other stuff.

Speaker 3 00:15:55 And since I heard him say this, I’ve looked into it and there’s a lot of research to support that. So when you’re championing something, you just have to press over and over and over again. So we got the rest of the senior executive team B, B and I on board. It took it took a long time. And then they started championing. And then the people, the middle managers that reported to them started championing. And pretty soon you start to build momentum, right. And championing alone is another necessary but not sufficient condition. Conditions. So the other thing we did was we actually we use OKRs internally. And at one point I came out at the beginning of 2024 and I said, all right, every single department has to have an OKR that helps them to develop two uses of AI in that department. Operational real operational uses in this OKR period. Right. And so that’s kind of a forcing mechanism. So we gave them the tools we started championing. We put in a forcing mechanism and it really started to take hold.

Speaker 3 00:17:02 But Julian it stalled. You know people got to a point in this journey where they did everything they could do. It’s a new technology. Nobody really knew how to use it. And we started getting some good stuff in process in our operational work environment. And the journey stalled. People were just tapped out. And you don’t know what you don’t know. So at that point, you know, I had been really fortunate that I had gone to Growth Elevated in January 2024. I saw Kevin speak. Right. Kevin and I didn’t know each other before then. We’ve since become good friends, because we have a lot of the same interests and live in the same place. But anyway, I called up Kevin and I said, hey, Kevin, you know, we need some help. Can you come in and help us to jumpstart this effort again? It’s stalled. And so we brought Kevin in for he and I workshopped what that would look like, and we ended up coming up with a full day of training.

Speaker 3 00:17:59 And I literally flew people in from all over the country. We’re a remote organization. We had there are seven people that report to me and Kevin. I think there were 18 or 20 in that workshop that you did for us, 18 or 20 people. So we went down into the middle management layer. We even, you know, brought in a couple of people who were below middle management who were strongly interested in AI and were influencers within the organization. And having that one day workshop, having Kevin opened people’s eyes to what was possible and teach them skills that they didn’t have. Got us going again. And we’ve been going like gangbusters ever since then. There’s more to the story, but Kevin wants to say something.

Speaker 2 00:18:38 Yeah, I have a quick add to that. And it’s it’s that, people like me are really good at getting people like Sean really, like, jazzed about all the things that you can do. And now there are a lot of me that are running around doing this. And what happens is, if you if you don’t push it farther into the organization, all of the ideas go to not because the types of development and workflow optimization that you’re trying to do with, with LLM type technology are just different than what people, what organizations are used to.

Speaker 2 00:19:10 And even dev focused organizations that have some, some capability forget that their dev team doesn’t know anything more than anybody else does about this. This is all very, very new in general and pushing it farther down into the organization, which is something coming a bit of a specialty for us, to be honest, when we’re trying to build that capacity. And a band in the middle of the organization allows those smart ideas to hit the ground and be mvpd a heck of a lot faster because people deeper in the organization understand what the heck you’re talking about. It is shocking. Absolutely shocking. And I this is probably the fourth time this week I’ve spoken publicly that how behind most of the workforce is leadership is starting to get with it. Any conference you go to is going to have AI sections and panels and whatnot, but nobody’s paying attention to the rest of the workforce. So there’s a stat I think I have it later that that only 15% of employees feel like they’re getting adequate training around AI, yet leadership’s hammering at them to use AI.

Speaker 2 00:20:15 Use AI.

Speaker 4 00:20:16 Yeah, yeah.

Speaker 3 00:20:17 So there’s a skills gap there for sure. People can be given a tool, but if you don’t teach them how to use it, if you don’t give them the proper training, they’re not going to be able to use it the way you want them to use it.

Speaker 1 00:20:29 So, Shawn, you started really early because I remember when Kevin gave that talk that was January 24th. So that was 18 months ago. So you had already bought the tools, got your executive team together, put in OKRs, and yet it still wasn’t enough. But then, you know, you made this a priority. You put budget against it. You flew people in. You hired a third party. What were the things that that that that, you know, working with Kevin and bringing all those people together. How did things change? You had the OKRs, but then after that, what was the difference that that allowed you to really take the steps forward?

Speaker 4 00:21:05 So it was really.

Speaker 3 00:21:06 There were a whole bunch of little things.

Speaker 3 00:21:07 But the two main things that happened for the people that came in to see Kevin is they number one. He opened their eyes to what was possible. Sometimes you just don’t know what is possible. And with AI, there’s so much that’s possible. Especially if you look. You know, what was that 13, 14 months ago when Kevin did this for us? People didn’t really know what it could do. And so he opened their eyes and that was really critical. The other thing that his team did is they came in and they taught skills. They taught the team how to make a simple GPT. And people then realized how easy it was, and they started making GPT to help with their everyday life, right in their everyday work environment. It was those were the two biggest things that we got out of that day was the eye opening and the skills development.

Speaker 1 00:21:56 So I’m looking at your bullet points here. Maybe we could go to an example like your HR director said. Gosh, it would really be valuable to help screen resumes because they do it all the time, right.

Speaker 1 00:22:05 What what did you do to support that? Because I guess I could see some of our rank and field people saying, you know, with the encouragement from leadership, here’s an idea, but I don’t know what to do next. Is that the kind of the situation there.

Speaker 3 00:22:16 So so that is actually a recent one. You know, there are areas of our organization that are further ahead in this journey than others, and HR is one that hadn’t done much with AI yet because our HR, we’re only 175 people. We have two people in HR, right? It’s not that that much. So so we had this issue where we’re trying to hire and we’ve got over 1000 resumes for this position, and the head of the department was complaining to me is like, the HR is not getting through these resumes fast enough for me. I’m jumping in to help and I don’t have a lot of time. So I went to the head of HR and I said, hey, you know, have you thought about using AI to screen the resumes? And she was like, whoa, we can do that.

Speaker 3 00:22:56 And I said, absolutely. And so I sat her down. I literally said, you can build a custom GPT in ChatGPT, right? And I said, how come the HR software, which we use jazz HR to help with recruiting and resume organization and interviews and all of that, I said, why isn’t it in their platform yet? And she said, well, you know, they’ve had some issues, blah, blah, blah. I’m like, you need to think about changing platforms first of all. Right. Because if they’re not putting AI in their platform, then we need a different platform. But second of all, you can probably just do this with ChatGPT. And she came back to me and said, you know, Sean, there’s actually a course that we can purchase that I can take that teaches you how to do this with ChatGPT. Can I have budget for it? And I said, absolutely.

Speaker 1 00:23:45 That’s awesome. And did they build a custom GPT? What was the what was the the final answer there and how was the impact?

Speaker 3 00:23:53 So Sariah is a non-technical person.

Speaker 3 00:23:56 She is in HR and she went out and she took that course. And yes, she learned how to build a custom GPT. You know, she comes and asks questions of our engineering team when she needs to, but for the most part, it’s, it’s pretty straightforward and pretty easy to do. Once you understand the basics.

Speaker 1 00:24:16 And I’m going to jump a little bit to the end here, Sean, because you told me in our prep, you you’ve fostered an organic adoption such that that things are happening across the organization. I think you said you’re not you’re not even aware of all the places where it’s happening. Is that true? That was.

Speaker 4 00:24:32 Absolutely.

Speaker 1 00:24:33 True. That’s pretty cool.

Speaker 4 00:24:34 Yeah.

Speaker 3 00:24:35 It is cool. And that, I think, is really the biggest point, Julian, is if you really want AI to permeate what you’re doing, it’s not enough to have it start at the executive team. It really has to come from the bottom up. You have to create this culture of people that want to do it, that recognize that this is going to make them super productive at their jobs, and to embrace it and start to give the organization suggestions how to use it from the bottom up.

Speaker 3 00:25:02 That’s the culture you’re trying to foster.

Speaker 1 00:25:06 Absolutely. Kevin, is that is that consistent with what you see across different clients who have been successful?

Speaker 2 00:25:11 No.

Speaker 4 00:25:13 no.

Speaker 2 00:25:14 I wish I wish.

Speaker 4 00:25:15 For honesty.

Speaker 2 00:25:16 Like Sean. you know, one observation is you see a lot of really big players out there. CEO of zoom, Shopify, etc., that are putting out these AI manifestos, as it were, that we’re going to be AI first and yada, yada, yada. And to his credit, Sean was early to that table to as far as establishing a culture of experimentation and embracing the technology, before it hit an inflection point. So I’m not going to detail your mandate. It might be something that you might want to do a little bit, but it is similar in my understanding, to how these other companies have approached it, that if you are not going to be on board with this, then this may not be the environment for you. We’re not here just to preserve actions for actions sake.

Speaker 2 00:26:12 We’re here to make the organization better. And one of the things that I do see, unfortunately, way too often is there’s an approach where you look at human oriented processes and you just try and recreate all the human oriented processes. So if you have 72 different processes in the organization there, you can do that. You can have somebody like me come in and build agents that do process one through five and then six through ten and so on and so forth. And all you’re doing is replicating the way that humans do things. But culturally, what you really need to do is you need to look at the why you’re doing things and what bridging that gap and reassessing how all this came to be. Like, Sean’s organization’s been around for a while, and it’s always been a tech forward organization. But there are processes that exist based on human limitations on both the internal side and the client side, and natural timelines that result. And particularly in marketing. But this can be applied pretty much anywhere. We build these really long calendars around things, and they’re about meeting to meeting and then meeting to scope and then meeting to ideate and whatnot.

Speaker 2 00:27:28 And really, if you could compress all of that where everybody comes to the table, like loaded for bear and can really solve problems, the rest of that calendar isn’t that needed. There might be production pieces or scheduling pieces or whatever. It might be the mistake that that. So we’re tearing up, right? The first mistake is that that that that company is trying to solve human oriented problems. The next mistake is even if they recognize that they can change the way they do things, they don’t actually change their processes to capture that gain. And this is a leadership group. So it’s something that that that CEOs and leaders need to understand that if you take a six month process and make it one month worth of worth of work, but you don’t actually change that process in the organization, it’s still going to take six months. You’re not actually changing the output dynamics of whatever you’re doing. So you as a leader need to recalibrate and realize, okay, well, client onboarding now only takes about two hours because we’re doing X, Y and Z.

Speaker 2 00:28:33 But everybody’s workload is built on it taking two months. So how do we capture those gains as opposed to people sort of you know, let’s go skiing this afternoon, which don’t get me wrong. Like culturally that might be that might be something you want to do too and just make people’s lives better. But capturing the gain to something that that I see most often.

Speaker 4 00:28:53 So, Kevin, you.

Speaker 1 00:28:56 You brought up this topic about AI being like electricity. And I pulled up that slide because I think that that that dovetails really well with Sean’s story about saying, look, it’s about empowerment and it coming bottoms up. And that’s kind of like if you have electricity, you know, you could do some top down projects around how do we use electricity. But at the end of the day, it impacts everything. And I think that’s what I heard Sean saying. but but you you had some views on this in terms of super cycles.

Speaker 2 00:29:21 Yeah, I had some strong views on this, and it was sort of funny in our prep meetings because Sean and I were having Julian being being wanting to deliver very precise value for this community was sort of hammering on us about what what what tools, what are the tools, what are the things that you do? And it’s a really elusive answer because the answer is it depends.

Speaker 2 00:29:43 And the people too many people are looking at Lem technology as a tool set. It’s just a new wrench that you use to do things, as opposed to being a new approach to the way you kind of do everything. And electricity is the best example. Like trying to explain the tooling around electricity. Doesn’t really get anyone anywhere.

Speaker 1 00:30:07 It’s because there’s a million uses.

Speaker 2 00:30:09 There’s there’s there’s there’s absolutely endless uses. And framing this around a super cycle. I mean, there’s been a lot of really, really crazy news in the last couple of weeks with Dario Amodei is the CEO of anthropic came out. This seems like it’s maybe a little bit off on a PR perspective, but he came out and said that he thinks that 50% of knowledge work is going to be basically out of work in the next five years, and that we could have unemployment in the 10 to 20% range. Like, that is a jaw dropping, crazy thing for a frontier AI person to come out and say, and I can’t say that that that I necessarily disagree with that in the from what I’m seeing, but it’s it’s a sign of it, this being part of a super cycle Goal and super cycles have been around since the dawn of time.

Speaker 2 00:31:00 Super cycles are usually defined by general purpose technologies. GPT just have by by by accident that redefine the way we do anything, everything. take steam and electricity coming together to cause the industrial revolution that changed human patterns across the globe, shifted our economy from an agrarian economy to an urban economy, created all of these new business models, all of these different opportunities, all of these different jobs and careers and businesses that were never possible before. And then super cycles tend to build on one another. And this particular slide is showing the super cycles that everyone on this call have already lived through. And a lot of these felt like sort of boiling frog type deals because they took a long time. I’m still having conversations with people about cloud cloud deployment type stuff like cloud is not done. Cloud is still like happening. The difference with this particular supercycle is it is accelerating so rapidly that the new businesses, the new careers, the new jobs, the new approaches, the new, the new uses of electricity haven’t quite snapped in yet.

Speaker 2 00:32:17 And to be honest, even people like me who are really paying attention, it is really hard to say what those are going to look like yet. So here we all are. We’re all running our businesses, and there’s this question of whether or not the business dynamics that that you’re building on today are being built on sand because of the change in technology and really leaning into it and understanding that and rolling with it becomes really, really, really important.

Speaker 4 00:32:49 Hey, Kevin, I’ve.

Speaker 3 00:32:50 Got something to add there. If it’s okay, Julian.

Speaker 4 00:32:53 Of course, you know, one of the.

Speaker 3 00:32:54 One of the things that, we found as resistance at first was people were afraid they were going to lose their jobs, right? That I was going to take their jobs away. And that dovetails with the story that you just told about, you know, 10 to 15% unemployment in five years. But what we were able to do was help people to understand that if they mastered AI, then they were mastering their future.

Speaker 3 00:33:19 Because the future of the knowledge worker is the future of mastering. I think of themselves as pilots in a cockpit where they have all this advanced electronics around them, and they are flying the airplane, right? The AI is the technology that they’re using. And anybody who mastered that technology has a future. And that really resonated with a lot of people.

Speaker 1 00:33:41 I love that, Sean. And that gives people, you know, the hope and the motivation to do it. And you’re right. It’s this combination of fear and anxiety. You know, the the the the view I have on this slide, Kevin, is, you know, you didn’t have to you didn’t have to mandate from the top down for people how to use the iPhone for the mobile mobile group. Right. You didn’t have to mandate from the top down for people how to figure out how to use Google in the internet age, right. And so that’s kind of what I’m hearing from you guys, right. It’s like, okay, it’s going to happen.

Speaker 1 00:34:13 It’s going to happen sometimes on the personal level faster than it work. But but Sean, is is it fair to say as a CEO that’s kind of your point. You have to you have to make it a priority from the top down and then do the enablement and then do the education and support versus choose or decide where you know, where first. Is that a fair summary of what you were? You were trying to explain to me when I was trying to trying to get get more, you know, step by step instructions.

Speaker 3 00:34:39 That’s a beautiful summary, Gillian. I think you encapsulated it perfectly.

Speaker 1 00:34:44 Very cool. All right. Well, let’s let’s let’s continue here. so, Kevin, for the people out there say, okay, well, I want to I want to be like Sean, even if it’s to two years later, you know, what can I do to do bottom up enablement, enablement with guardrails?

Speaker 2 00:35:01 guardrails are important here because there’s the stat that’s sitting in the middle of that page that 75% of employees are already using AI tools.

Speaker 2 00:35:10 And guess what? If only 15% of people actually feel literate about the tools? Most of those people are using those tools at best ineffectively, and at worst, they’re using them in a way that can compromise organizational security. there’s a marketing truism that if a product is free, you’re the product, right? So any employee who’s sitting there on the sly, sort of phoning in their work through GPT is not only not doing their job, but they could be exposing some of your corporate data to risk. So if you don’t have an AI policy, you need to have an AI policy that defines what tools that people are allowed to use, which tools they’re not allowed to use. defines where it’s appropriate to use this. It drives me crazy when I see policies that are like, there’s no AI anywhere. You’re cheating on your job. I don’t care how you get the job done, like get the job done to the quality that we need the job done to, and use tools effectively, but hold people accountable as well.

Speaker 2 00:36:17 So I like to see aggressive policies that still have accountability that are built into them at the end of the day. The excuse, oh, GPT came up with the wrong case or whatever it is, that is not an excuse. You are responsible for the outputs from whatever tool you’re using, and you should be held accountable in terms of continued employment. you have to lean into literacy once you have some sort of a policy. it is totally crazy. Most, most organizations that run G Suite or Google Workspace just have Gemini enabled right now. So it’s just there. It’s just a star that’s sitting in the corner of everybody’s screen and nobody really knows how to use it. And worse, again, they don’t know if they’re using it correctly or data’s leaking into it or whatever, and it is just complicated enough that people need significant time with training and applied bits and pieces to get them up and going effectively, because what happens is they’ll get really frustrated with it. I’m sure there are people on Microsoft Teams that are on this call.

Speaker 2 00:37:25 the copilot in Microsoft is just absolutely horrible to work with right now. I promise someday it’s going to get better. I see signs of light, but people get enormously frustrated with it, and they’re like, they toss up their hands and they move on. They never push through that sort of that sort of uncomfortable, period of disillusionment. And you do have.

Speaker 1 00:37:47 To the trough of disillusionment. Yes. We’re seeing right now. Many times.

Speaker 2 00:37:51 Exactly. Hype cycle. Right. And everybody on this call is in their own place on the hype cycle. you know that people have fallen off and they’re in that trough of disillusionment. Either a lot of people there, a lot of executives who are like, oh, we tried it in early 2024, didn’t go anywhere. And they haven’t kept at it like Sean’s team has. So they haven’t pushed through. And I can promise you, you pour a few billion dollars into Silicon Valley and good stuff occasionally pops out the other side. A lot of these tools are getting a lot better now.

Speaker 2 00:38:24 So people who have sort of thrown up their hands and walked away are missing an opportunity. but people need to understand the actual tools they’re using. It might be just GPT. It might be copilot. It might be Gemini. maybe in the organization. It’s a creative organization. And they use the Adobe AI suite, but they have to be encouraged to find training or collaborate on training because even as a training provider, I can tell you this stuff moves so fast that this is why I have to direct my team to like YouTube. And I say just ignore anything that’s older than, say, six weeks because it’s going to be obsolete. So you have to find the most up to date, reasonably reliable training. But it should be practical, practical, based on practical applications so that people are actually getting their hands dirty and they’re trying to build projects.

Speaker 4 00:39:19 Kevin, I.

Speaker 1 00:39:20 Think I share a story. I think, I have some inside baseball, but just just keeping a list of the most updated YouTube, resources that to get you up to speed seems to be like an unending task for you.

Speaker 2 00:39:32 Oh, it’s unending. It’s absolutely unending. And I’m trying to. I loathe social media, so, like, having to hang out on YouTube makes me die inside. but, there, there there are great people to follow out there. but building MVP’s is next. Like solving real problems. Like finding little things in people’s in people’s jobs that they can deal with, like the, the HR, resume vetting type thing. That does not solve the entire problem, but it solves part of the problem that was taking up a bunch of that person’s time. And that’s a win for most people. And actually go on to the next slide, because that’s.

Speaker 1 00:40:10 What I was gonna say. We’re just where do we start?

Speaker 4 00:40:13 So where do we start?

Speaker 2 00:40:14 So, our own, growth elevated member, Kathy Lewinski first coined this, all the way back in 23, the gym days of 2023. She calls this the pyramid of suck. And it’s the idea that, people want to just change their business. I was saying this earlier, but they want to just revolutionize the way they do everything in their business.

Speaker 2 00:40:37 And you can get there from a product perspective. But the smart thing to do is to first identify that people, that things that people don’t want to do or don’t like to do because they stink. And the example I like to give are like stars. You get a great star who’s awesome on the phone and on email, and they can close deals and they can do all of these things. And then you put a hammer on them and tell them that they have to keep their CRM fully updated, because they do. There’s good reasons for that. But you’re taking this person’s who’s person whose personal genius is in sales, and you’re turning them into an admin and sucking away from the sales time that they should be doing.

Speaker 1 00:41:16 Have them research phone numbers, endless, you know.

Speaker 4 00:41:20 Web service, stuff like that. Yeah.

Speaker 2 00:41:22 As opposed to you want them on the point of the spear doing what they’re supposed to be doing. Right. And we all do that as leaders. You don’t want to, but it’s just like the nature of it, right? So if you liberalize that, that that person’s workflow.

Speaker 2 00:41:33 So they get to focus on the thing that they’re best at. Usually they like doing that thing, particularly if it makes some money, right? They lean into it. You have more sales, it’s a win. And then this next layer are things that we don’t do well. These are we call these the MIT AI Learning Lab calls, these clerks, colleagues and coaches. And this is a this is what I would call a colleague. So these are things that individuals don’t do particularly well. I’m a terrible visual marketer. So I have always been reliant on graphic designers and long turnaround cycles and things like that. And now, at least in basic presentations and things, I can use my AI tools to amplify my design skill set, to get to things that that work for, whatever presentation or whatever it is. So I’m amplifying myself. So translation data analysis or just some research or examples of these things where individuals can amplify themselves more culturally, this is going to be a problem like that? One is going to be a huge problem in larger organizations, because it’s great for entrepreneurs in this sort of tier to be like telling people to enable themselves and put on 15 different hats and do their thing.

Speaker 2 00:42:47 That’s fantastic. But if you have these large matrixed organizations and you’re encouraging people to expand their sphere of ability, they start falling into other verticals and it’s going to cause like corporate organizational chaos and change consultants are going to make a ton of money in the next few years, I suspect. But then once you have the your legs going, only then do you really start thinking about things that will actually reinvent your business. What does the what does your business do fundamentally? And can you change what that is using this technology? But if you start there, you’re going to kind of fall on your face.

Speaker 1 00:43:26 But I think in terms of like Sean’s framework of enabling from Bottoms Up, there’s a really high alignment with motivation. If you think about the pyramid of suck and things you really don’t like to do, and that’s that’s maybe a lesson that or a task, you can say, okay, everyone create a list, right? And that that could be a good way to start thinking about, okay, well then x percent of those can get automated.

Speaker 1 00:43:47 I’m going to move on to the next framework here, which we talked about. You know, that’s kind of the personal productivity, category here where, you know, in the need for structure. I asked these guys, can we create some some steps? And the first one that we’ve been talking about a lot is the personal productivity, making sure everyone’s enabled, figuring out what are the what’s the bottom layer of of of the pyramid of suck that, we hate to do, making sure that people are trained to do those type of things. if that’s one category, you know, is it fair to say, like, everyone on the, on the phone, if you do nothing else, focus on that and get to like a B+ level there to, to get your team working. Right. Because I think the next two get a little more intimidating. And I do want to save time, Shawn, because you’ve not only done it internally with some of your operating tools like you described with that recruiting, you’ve also embedded it into your product forward facing which which, you know when you start hearing about that.

Speaker 1 00:44:44 If you haven’t done the first, the jump to that feels very intimidating. But is it a fair, fair statement to say, look, almost everyone can start with the first column and make real gains? Is that a fair statement?

Speaker 3 00:44:56 100%, I believe so, and I actually think you can go from the first personal productivity to embedded AI functionality and skip the second for a period of time. I mean, that’s what we did. We’ve actually spent more money and more time in our organization on embedded AI functionality in our client facing products than we have in our internal stuff.

Speaker 1 00:45:17 And I want to I want to jump jump there in a minute. Kevin, any any you know, we talked about personal productivity. Do you have some high level things to talk about in terms of what’s possible in the middle column. And you know what? What advice you’d have to people. And then we’ll skip to because we only have ten minutes more. I want Sean to talk about how not only, he’s enabled his team, but he’s putting it into his product and making my advice more competitive in the marketplace.

Speaker 2 00:45:42 So 2025 is the year of the agent. it is. There has been a ton of hype around agents. Agents are basically single purpose, lem activities and that that just accomplish one thing. It might be lead enrichment or whatever it might be. And what we’re seeing is the chaining of them together and even adding some level of intelligence together. most, most, if not everybody on this call knows what Zapier is or what make is. those have existed for a long time, and they’re absolutely great. They’re basically digital duct tape, to connect different things and adding agent level, insights to the middle of those value chains and reducing the friction as far as connecting to other services. And there’s some wonky stuff going on with MCP and some other standards that are happening, but they’re all enabling a dramatic reduction in friction between all of these different platforms that allow mere mortals to put together very, very complex, AI enabled workflows that solve specific problems.

Speaker 1 00:46:48 And I’m going to recommend, Kevin, that we set up maybe a separate, separate activity sometime in the future where we go deeper there because that that could take forever.

Speaker 1 00:46:56 But let’s really just jump and let Sean show off some of the things that he’s done. And then meanwhile, I want to talk to the audience. If you guys have questions, I have a little tab here so I can see them. Please enter your questions in zoom and we will save some time to try to address those. But Sean, let’s jump jump forward here to, the things that you guys are doing with your product and just be available for people so you can see it. We’re not going to go through all the pages.

Speaker 3 00:47:23 So these are just some highlights. There’s actually more that we’ve done. But Julian, I think one thing that’s important to realize is that, you know, we recognized early on we had to embed AI in the products, and then we had another realization, which was that to really do that, we needed to allocate time and capital in a way that we didn’t have those capabilities internally. So as the CEO, you know, you have to make a capital allocation decision.

Speaker 3 00:47:49 And we realized we needed a dedicated AI product development team. And so we went out and I authorized him to go out and build a 15 person AI development team, 15 people just dedicated to building AI into our client facing product suite. When I say we’ve invested more time and money in our client products suite than we have internally, that’s really what I’m talking about. The other thing we realized is he was doing that is that, you know, Bebe was at that time managing all of our software engineering, and he was managing our platform delivery. And if we really wanted him to be able to focus on the AI piece, we had to give him more time and space to do that. So I actually took over platform delivery, and that reports to me now to give him more time and space to focus on embedding AI in our products and in our organization. And so those are really important considerations and decisions that you have to make in terms of allocating people skills and capital. But it’s been amazing. Like we have been the first to roll some of these things out for our segment of the market.

Speaker 3 00:48:57 And this is just a sampling. I mean, we’ve got a tech stack that runs websites, it runs reviews, it runs social media. It’s got all of this sort of peripheral things around that. It’s got a CRM, right. It has SMS messaging. And so we’ve been able to sort of one at a time kind of AI enable these features. So automated review responses, it is best practices. It’s best practice to respond to all reviews. Most of our clients don’t do it because they don’t have time. So we’ve now created a voice for them to be able to do that. It’s their voice. It’s a model that’s trained on their company. It speaks the way they want to be heard, and so it’ll respond for them automatically. Right. I mean, it’s awesome, right? It allows them to follow best practices. And since most of our clients are in medical, we also have a HIPAA compliant agent. So all of those responses are run through the HIPAA compliant agent to make sure that they are actually in compliance with HIPAA.

Speaker 3 00:49:57 When we when we were building this out, we analyzed a bunch of our clients review responses, and we realized that over 90% of them were not HIPAA compliant. So in addition to giving them the ability to respond without taking time, we’ve also made them legally compliant. It’s super powerful. The social media posting again. You know, there are best practices around the types of posts, how frequently you post, and a lot of our clients, because they’re small businesses, they just don’t have time to do it. They don’t have the staff. They don’t have the money to hire a social media specialist. And so now we can do that for them with AI in their voice on the schedule that they define. We tell them what best practices are and most of them follow that. website chat agent. Right. These are now starting to become more popular, but our website chat agent is actually trained on our client. We have an individual website chat agent for each one of our clients that knows what our client does.

Speaker 3 00:50:57 It’s been trained.

Speaker 1 00:50:58 How many clients do you have, Sean? Like, give me this size of that scope like that? Is that hundreds of different ones or dozens?

Speaker 3 00:51:04 It’s it’s almost 2000. Julian.

Speaker 1 00:51:07 You have 2000 different agents.

Speaker 4 00:51:10 Yes.

Speaker 1 00:51:11 Wow. And because you’re smiling, that’s. I imagine that’s easier than I’m imagining.

Speaker 3 00:51:18 It’s easy once you figure out how to do it at scale. Yes.

Speaker 4 00:51:22 Figure out how to do it in scale.

Speaker 1 00:51:24 2000 customized agents. But like you, it’s behind you. You’ve done it. That’s. That’s quite amazing.

Speaker 3 00:51:31 Yeah, it is amazing. And once again, once you figure out how to do it at scale, it’s really all about training the individual models. And once you figure out how to do that at scale, it’s actually quite simple. We can set one of those up very quickly now.

Speaker 1 00:51:45 That is.

Speaker 4 00:51:45 Great. Yeah. Yeah.

Speaker 3 00:51:48 But it allows you to go deeper. Like when when Kevin was talking about it’s a way to rethink what you do.

Speaker 3 00:51:54 Like don’t just use AI to do what you have been doing in the same way, just without people.

Speaker 4 00:52:00 Yeah.

Speaker 3 00:52:00 You have to completely rethink what is possible. And so for us, that means rethinking workflows like Kevin was talking about. So in terms of this SMS response agent, And that’s not necessarily something we were doing before, but because we have this AI response agent, we can now in just not just do it on chat on the on the website, we can actually do it over SMS. So we’ve built that capability into our CRM, our CRM, where our agent can correspond on behalf of our clients in the client’s voice to do things like provide information, set up appointments, all those kinds of things. It’s really it’s an extension that we probably wouldn’t have had without AI.

Speaker 1 00:52:47 That’s awesome. Hey, we’ve got a question from the audience. what blind spots do you see in how startups use AI? What are some of the mistakes or blind spots as asked here?

Speaker 2 00:53:00 I think one for me is definitely assuming 100% solution is is is possible.

Speaker 2 00:53:06 And if we’re not, we’re there in some in some limited cases. But this idea that 80% there isn’t good enough. Like, if you can free up 80% of somebody’s time to do something else, that’s a that’s a win. Like lean into those things. the other that I see is building stuff that they shouldn’t be building. really, the buy market has become much, much better out there, as I said earlier. And, you know, people get really excited about this stuff and they see something as saleable. And like Sean turning his air scanning tool, which, by the way, that’s very saleable. That’s not a really complicated project, but deciding to turn that into like a full blown development exercise, is a mistake. If it’s not, if it’s not a competitive advantage to you to build something and it’s not available on the market yet. Just wait. Chill out. Like there’s the time will go.

Speaker 1 00:54:01 On as soon as what you’re.

Speaker 4 00:54:02 Saying, it’ll.

Speaker 2 00:54:02 Come. There’s something. Who, out there who’s working on your problem right now? They just haven’t quite had gotten it to market.

Speaker 2 00:54:10 and you know, focus on there’s there’s there’s more low hanging fruit out there.

Speaker 1 00:54:15 Okay. I’ve got another question here from the audience saying, you know, how do you keep your teams upskilled if you commit to, you know, I think the message is invest in your team in terms of education and enablement. But how can you do that? I’m going to ask the two questions when it’s evolving so fast. And so the first question is going to be catering to to you a little bit, Kevin, you know, what’s the scope of a project like, like you did with Shawn. Like, you know, how much does it cost? How long does it take to kind of do do something like that? And that’d be the first question. And then second would be, then how do you stay updated? Because, you know, six weeks later, you know, 50, 50% of new new product has come out.

Speaker 2 00:54:53 So one of the approaches that we always take to a workshop is to is really to try and teach people how to fish that trying outside of the core technologies, like like using a chatbot and learning how to prompt a lot of it is framing how you need to approach these problems.

Speaker 2 00:55:10 And people can fish for a decent amount of time with with that sort of information such that they can then go chase more information that’s contemporaneous for whatever tool that they’re after. But, you know, we’ll we we do charge a decent amount for that. We’ll come in for 5000 for a half a day, or 10,000 for a full day, to work with larger teams to do that. Increasingly, we’re actually being engaged on an annual basis. So one of those is coming for right now where we’ll be doing a larger companies. So companies that have a couple thousand employees will do broad scale trainings that are applicable to everybody. How the heck you use copilot? and then sort of town halls to keep people updated with advancements in the tech stack that’s specific to the company. And then we do these deep dive programs where we’re enabling people within the organization. I mean, that’s that these are not small type projects, but for organizations that are really trying to lean in to build capacity, there are programs like that that are springing up on the technical side.

Speaker 2 00:56:17 There are semi technical side. There are a lot of cool boot camps and things like that that people are leaning into to, try and develop agent based development skills. And on the developer side, developers are very well aware that they don’t really know how to use a lot of this stuff. So you’ve got a lot of people who are taking like night programs, trying to upskill themselves such that they can run captive models and be useful with vector databases and things like that.

Speaker 1 00:56:45 Well that’s fantastic. you know, we’re we’re at time, Sean, you know, last comments. First of all, congratulations. And thank you for sharing with with everyone. I think you guys are a great example of someone who’s, ahead of the curve or at least on the front end of the curve. And, and, I think the message that this is approachable and doable through leadership is a great one, but I’ll give you the final word.

Speaker 3 00:57:10 Well, I think that the the important takeaway here, I am obviously a big proponent of AI, right? I love it, I believe it can impact everything we do and it’s not.

Speaker 3 00:57:20 It requires dedicated effort, but it’s approachable. It’s doable. Everybody who’s listening to this can do it. It’s just a matter of making it a priority, making it the priority. And you can do it. You can get it done. You can have an organization that buys into it and uses it extensively.

Speaker 1 00:57:37 That’s fantastic. Well thank you Kevin. Thank you Sean. To everyone listening out there, the whole presentation, we’re going to put it, on on the the portal for growth elevated. And you can obviously reach out to Kevin and Sean directly. we’ll we’ll share the information, all their contact info. Guys, thank you so much for your time today. I think this has been great.

Speaker 3 00:57:59 Thank you Julian. Thanks, Kevin. We’ll see you.

Speaker 4 00:58:01 Guys. Bye.