Transcript
So, hello, everyone, and welcome. So excited to have you with us today. We have a great speaker and a great topic today. My name is Muna. I'm head of product marketing here at Eterra. And just before we get started, I am seeing more and more people joining us, so let's just wait another minute and see and allow people to join. And as we wait for more people to come in, I'd love for you to test out the chat box. Let's just make sure that our chat is working properly. So let us know where you're joining us from today. Great. Thank you, Chris, for being the first one. Chris is joining us from Dallas. Welcome. Wonderful. And that ensures that Chris can hear us well. Chris, just let me know as well if you can see our slides. Hello, Brian, welcome. Welcome, everyone. Thank you for joining. Wonderful. Thank you for confirming there that you can see and hear us okay. Wonderful. Great. So we'll just give another minute for more people to join us this morning or this afternoon. Depends from where you're coming in. And we'll get started. Wonderful. So the chat's working. We have Sid with us this morning. Hello. Good morning, Sid. Where are you joining us from today? I am in Lufkin, Texas. Wow. Okay. Is that... Tell us exactly geographically, where is that in Texas? So if you trace a line between Houston and Dallas, go about halfway between and then cut to the east to almost Louisiana, you'll see one of the largest lakes in Texas. And then it's next to a big lake that is part of the border between Texas and Louisiana. We are just to the west of that large lake. Sam, right where we're looking. Wow. Lovely. Okay. So we'll be hearing more about DETCOG today. And I think that we can get started. I see more people joining, so I think we can get right into it. So again, everyone, hello and welcome. Thank you for joining us. Welcome to a conversation I've really been looking forward to. I know that local government IT doesn't get talked about enough on stage, on stages like ours, and it should, because the pressure is real. There's a lot of small teams, sprawling jurisdictions, aging infrastructure, and employees across multiple counties who really need to get their job done. And Sid's going to talk to us about that. And I'm really excited to welcome our speaker today. I'm joined by Sid Munlin, Director of IT and not Information Technology, but rather Innovation and Technology at the Deep East Texas Council of Governments, or DETCOG for short. And today we're going to spend time on an unfiltered look at what it actually took to move from reactive IT to autonomous IT with Robin Bayatera. And Sid is going to help us understand what his team was drowning in, why they chose the path of agentic AI and autonomous IT with Robin. We'll talk about what they automated first, and then we'll go deep into what are the use cases and what are the outcomes that they are seeing. But before we do that, let's introduce our speaker. So Sid, again, welcome. Tell us about yourself and your role at DETCOG. Well, my name's Sid Munlin. I've been in IT for quite a while. Before I took the job with DETCOG, I was the CIO for a mid-sized municipality here in Texas. We were an integrated IT shop there, so we handled, well, everything. Everything from police cars, dispatch systems, fire, emergency ambulances. We had an interior jail for the city. We also had the judicial system. So, you know, not to mention we had all your typical city departments, you know, public parks that you have of a typical city infrastructure. And we, in an effort to increase efficiency, we centralized all of the IT work, even though some of it was fairly specialized, but we managed to do it. It worked well. The pattern still continues. I know a lot of other municipalities do that as well. But we had particularly good luck with ours and created a very robust environment for our users. I also, during that time period, was introduced... Sid, your sound... Sid, I'm sorry to interrupt you. I don't know if it's just on my end, but something about the sound is not coming across really well. You're cutting off for us for some reason. Okay. Is that any better? Let's try again. You sounded perfectly fine. Yes. I think now it's a little better. Let's try. Apologies for that. No worries. My apologies. Wonderful. So, we... Anyway, so we had the integrated IT department. We took care of everything. We were on 24-7, 365-day call, just like any other emergency services. And along with emergency services, one of the first tasks that I was given when I became the IT director was, at that time, we had the Spatial Columbia break apart, and it fell all over the region of East Texas. And so, they centralized in the city I was in. And so, I acted as the federal liaison from the city to the federal agencies, such as NASA, FBI, et cetera, DOD. They were all there to help coordinate the cleanup and recovery of all the Columbia pieces. From there, I also maintained my credentials in emergency management. And so, I was the nighttime or the chief emergency management coordinator for the evening shift and worked a number of disasters along those lines, from hurricanes to tornadoes to floods. So, it was kind of a multivariate role for me. So, again, I did emergency management, worked closely with the police forces, worked closely with ambulance services and fire, and just had a real public safety-oriented environment that we managed, but certainly didn't forget all of the rest of the city departments and took care of them as well. Wow, amazing. What a substantial resume. And from that, taking everything that you've done before and now bringing it into the realm of AI, and we'd love to hear about that in a minute and how that has transitioned you into this role. But just to ensure our audience are aware, so this is what we're going to cover today. We'll talk a little bit about what is DETCOG, what do you cover, what does IT look like at a company like DETCOG. We'll focus into, we're here to talk about Robin and your implementation of Robin. So, we'll cover what is Robin. Then we'll talk about your starting point, how you got to Robin, what were you trying to resolve, some of the first use cases, and then most importantly, what is the impact that you're seeing today and how are you measuring that success. So, let's get straight into it. I'd love for you to paint a picture of DETCOG, your IT environment, what does that look like, who do you support? So, maybe you can give us a little bit of an overview of what is the Deep East Texas Council of Governments and what does IT look like? Okay. So, Deep East Texas Council of Governments, a council of governments is kind of a strange political beast that sits at a level above counties and below the state. So, you have obviously federal, the U.S. government, then we have the state government, Texas, and then we exist as a resource to help coordinate the counties and what they get and what they're after and help them however we can. They're not over us, but they are our members. We're not over them, but again, it's a member situation. So, we exist as a resource for them. And many times the state will give us grants to administer on behalf of the region, so the groups of counties that are members of us. So, you know, and then from there, of course, we have our counties, our member counties, and then below that we have municipalities and so forth. So, it's a strange level. I don't think a lot of states necessarily, I think several of them have something similar. They don't necessarily call them the same thing, but it's just we act as that stop gap between the state level and multiple counties. Currently, DETCOG has 11 counties in it. We cover a rather large sloth by other people's standards. Texas is large, no doubt about it. And to us, we don't have that big an area, but to someone not from Texas or not from the U.S., it probably seems large to them. I think we have roughly 9,500 square miles of territory that we cover. So, something just shy of the state of, I believe, South Dakota is what my territory is. We have roughly 100 employees that help our different counties. And we are unique in that because our counties are members and we serve as a resource for them, we also service them and help them as well. So, they aren't part of our user base per se, but yet we also do help them with ticketing and we have branch offices kind of spread out throughout the area and that sort of thing. So, everything is made very, very more complex simply because of geographical diversity. If I have, for instance, a technician and he has to run up to help someone with something on the north side of our territory, well, that's an hour and a half drive windshield time one direction. So, theoretically, if then he gets something at the very southern end of our territory, well, that's another hour and a half drive from the central point. So, you can quickly figure out that three hours up there and back and then three hours down there and back, well, that's six hours windshield time out of an eight-hour day. That doesn't make for a whole lot of productivity, to be honest. Absolutely, and I guess this gets me to the next question of really at what point and what was the trigger that got you searching or curious about Atera in this case and then specifically into an AI tool and then get us into Robin. What were you looking to solve? Mostly, I was looking for a way to become more efficient, more able to help and assist our users and our members when we could and do it from a centralized location so that that way we had the ability to reach out, effect a solution for them without having to necessarily climb in a car and drive. And so, Atera first interested me because of when I came to this role at DETCOG, they did saddle me with a slightly different title, which forced me to have a different perspective than what I had before. Again, they made me the director of IT, but it didn't stand for information technology. It stands for innovation in technology. So, part of my job description right in the very title is that I have to find ways to be innovative here, that I have to find better ways to do things, ways to streamline the workflow, streamline the business processes just to make us more efficient and productive. So, one of the things that we were lacking when I took over the reins here was there was no cohesive help desk. It was really just kind of something that was being handled through just kind of a something that was being handled through email tickets and people getting up and trying to go find someone if you were in the building and that kind of thing. And real quick that you know threw a lot of blocks in my way. I was looking for something that I could do asset management through so that I could track troublesome systems, something I could look at repeat problems for, preferably something that had some kind of remote desktop system attached to it. I wanted it to be integrated with our antivirus agent and I began looking at those and it narrowed down the categories of what I was looking at. And Eteric hopped into there and I started looking at it very closely. And along those lines as I was looking at it and it checked off numbers, a number of boxes for me, quite a few. I began to get you know I was impressed by talking with the salespeople, talking with the engineers that were there and they had this AI piece. And so I started digging at that and talking with their engineers about what it could and couldn't do and what it would do and wouldn't do. And those thoughts you know eventually led me to you know sign on with Eterra. And again I was primarily convinced to sign on because it had all the check marks that I needed for what I considered a more efficient help desk. But it did have this AI piece that I could see would be had the potential to be incredibly useful if I could figure it out and get it into play properly. We worked really hard on setting it up. We spent a lot of time painting over it and trying to make sure that we had it as much information as we possibly could before we turned it loose. In hindsight, we had we spent too long trying to refine it and should have just cut loose with it earlier. It would have saved us a great deal of headaches because it does work in a way that lets it allow itself to gather knowledge as you go. And we were too busy trying to shove everything into its head on our processes and procedures beforehand and should have just you know kind of got enough in there and opened the doors and let it go in hindsight. But it has worked out really wonderfully. There's some very funny stories I have about it but one of the things was when we first started rolling it out there really wasn't a set name for Robin. You're kind of allowed to call it whatever. As part of our rollout, we chose the name Hal. Hal was our AI desk attendant. We started with a steering group, a small group, and introduced some of our more savvy knowledge users to it. Then in the second phase of the steering group, we introduced some of our more problematic users. Some of the people that had problems more often and were a little more technology challenged and that kind of thing. And in hindsight, we probably didn't explain it well. I had at one point two different users want to actually come and meet the guy they've been chatting with named Hal. I had to explain to them that Hal, as helpful as he was, wasn't actually a human. And they were a little... I don't know. I'm not entirely sure they believed the office but it was a very interesting reaction nonetheless. And that was during our training period, kind of rolling it out. And then I threw everyone for a loop because I announced that Hal was now changing to Robin. That generated some conversations that were actually pretty good to have because people thought we had fired Hal, a few of them. And then I was like, no, Hal's just changing names. It's the same. Anyway, they were curious if the personality was going to change. I was like, no, it's personality is going to be the same. It's just a new name. And so we had a lot of roundabout discussions about that. And for us, it may sound funny, but raising the level of awareness about what the product can do for people and how they can interact with it, especially once we introduced it into the Teams channels, where they could, as they work through their day and interact with other Teams channels and members on Teams, to have that channel where they can hop over and discuss the computer issue with Hal or ask what the best way to lay out a file structure is or any number of questions that they've ended up using it as a resource for. Well, that heightens awareness. And that awareness, the more aware they are of it, the more they use it, the more they use it, the better it becomes. So it's actually a pretty clean dynamic and works very well. Amazing. And thank you for that. And we will get into more depth on the different use cases and where you actually started off. I know you touched on it briefly on trying to fill it with all of the information. But it is funny that when I talk to some of our clients and especially the end users, when they want to politely say, Robin has really made IT more human. And I'm sure they don't mean it in a sense that IT-er. But yes, you come in angry and you're asking these questions. And I think this is part of what you're saying. It has a personality. It learns based on your organizational mentality and how you respond. And it starts to respond in a certain way that some people find humorous. And it keeps IT down to earth. So it's lovely to see that it's end users and IT are seeing a similar perspective. For the people on the line that are not familiar with Robin, I'll just do a short introduction. Robin is actually Atera's patented AI technician. We like to call it the IT technician that solves end-to-end tier one and tier two tickets. And it does this by taking real actions on devices, on the network, and in the cloud, often without needing a technician in the loop. And as just you mentioned right now, Robin can be available either through the help desk, Robin is on Teams, it's on Slack, WhatsApp for employees that are on the move, response through email, through a help desk desktop widget. There's different ways to apply Robin to your organization and make it work where the employees are actually working in their day-to-day. And just briefly, Robin is not an AI chatbot, right? So where organizations go in and they've applied assistive AI where a technician is still needed to work the ticket, or where the AI is suggesting and pulling a knowledge-based article and then routing it or deflecting the ticket, Robin actually works the ticket end-to-end. It starts from the detection of the issue, whether it's proactively or through a user engaging, it then diagnoses the issue on the device itself, whether it's the Windows, the Mac, the server, it then executes the fix, and it then verifies with the end user to ensure that the actual, that the issue is actually fixed. And we'll talk about some of those examples. So back to you, Sid, I really want to talk about the use cases, right? So, you know, AI, you said you started off with a pilot group, you said specifically, so walk me through how did you prioritize those specific use cases that you wanted to maybe get off your plate? Talk us through the process. Well, what we were primarily looking to do was cut down on the hands-on time for simple, repetitive tasks, and that's all we wanted in the beginning. Just those things that are time sinks, and we all, you know, everyone has them, you know, whether it's, for instance, one of the things for us was a lot of printers, different types of printers, you know, we have a plotter, we have all these different elements, and every profile requires their own printer setup. So, you know, we spent a lot of time, especially in a shared desktop environment, going by and setting up printers for so-and-so, or they're working out of this desk today, so we'd have to go set their printers up over there, and there was just, you know, wasn't a real good way of doing it, and we hadn't invested in looking at it through group policy or anything like that. Again, you know, when I took over the reins here at NEPCOG, there were a lot of challenges. Let's just, we'll call it that. They barely had Active Directory implemented. They did not have very good, you know, antivirus. There was a lot of challenges, so we, in addition to handling all the normal help desk priorities, we were also rebuilding and restructuring the infrastructure here, you know, especially looking at NIST compliance and, you know, Zero Trust and some of the other things. So, our hands were very, very full. We also were migrating them at that time from Google Works or Google Apps to the Microsoft 365 environment, and so there were a number of challenging large-scale projects going on at the same time, and certainly, you know, having one of my few technicians spend the greater part of a day because two or three people had desks around or were working in new spots, just, you know, assigning printers and that kind of thing, that was very counterproductive for what we were trying to do and very painful. So, to eliminate that was one of our primary keys. You know, I wanted to get off of those guys, the typical, well, it's doing what, you know, hey, reboot and call me back kind of questions and the, you know, again, simple things like printer setup, email setup if you're in a new profile or that kind of thing, and so, you know, that's where we really started, and again, because we were learning as we were going, and we were also constantly changing the environment in which it operated as we moved from Google to Microsoft 365, we were very selective in how we cranked up. We, again, I wanted simple tasks. I wanted to get those things off that were easy to do, that made a difference in time, and that, you know, the user could immediately tell did it work or not work, and so, we also built in, you know, some safeguards that if they hit a problem that, you know, that Hal couldn't figure out or Hal couldn't solve to, you know, create a ticket for it to create the ticket and kick it forward so that a technician could be involved, a human technician, and then move along those lines, but that was really the criteria in the beginning was just getting some of those simple tasks, the repetitive tasks, the time-consuming tasks, getting those off our technicians' plate so that other things could be put on there that, you know, were not a better use of their time because in the end, you know, our customers are our customers, and we want to make sure they're efficient and productive, and that's what we're hired for, but if we can find a way to do it and free us for the harder tasks or the more complicated tasks, then those are dollars better served, spent, than the other, so it becomes a, you know, it's a fairly easy alignment with the business process, and it's fiscally responsible, so those Those are some of the engines that drove that simple decision that, yeah, let's get the simple stuff off the plate. Let's let it take care of that and see how it does. That was our beginning pilot groups. Yes, ma'am. No, sorry. I interrupted you. Yeah, you're fine. Do you have a question? Absolutely. Yes, I was going to say, and as you talk about those specific pilot groups, is it because you said maybe people that are more technology aware, more tolerant, from your perspective, is it tolerant to AI? With all of the hype that we hear, AI in terms of environmental friendly, AI is something that is going to take someone's job. I don't want to talk to an AI chatbot. As part of that selection, was it more of people that you can get quick feedback from, or did it also have an element of how receptive they would be to AI? Was that even anything that you had to consider? It absolutely was. One, we wanted folks that were a little bit more technologically savvy. I wanted folks that were receptive to it, and I wanted folks that were administrative in nature, so department directors, our CEO, our vice president. These are the kind of people that I wanted on it because by giving them an initial take and giving them an advanced view of it, where they could see how it was beneficial by not having to have us in their office when they hit across a problem, and then being a little more technologically savvy so that they weren't as easily frustrated was a very key element. I put a lot of thought into that for the simple fact, they gave me buy-in for management. It proved that it was a productive decision to bring in the AI, and just how useful it was. That was part of my initial pilot group. Once they were on board and enjoying using it or maybe not enjoying, but certainly accustomed to using it, now we went to that next level and I started finding why more technologically challenged individuals or the ones that were more of a problem work case because they had a lot of the same routine tickets and that kind of thing, and just weren't as comfortable with technology in general. Those I brought on board, how do I put this? I went and I found my problem cases. I went those and found those that I thought would be resistant to it, so that in our pilot steering group meetings, I was able to combine those folks that were having issues or would be resistant to it with the folks that were pro it, and let them discuss it, let them talk about it, let them figure out between them what was going on, and in a lot of cases, I was able to just listen because there were some great conversations going on, where the challenged users would bring forth issues, and my pro users would offer them solutions to it without IT having to be involved at all. Amazing. That's where I wanted it. That is exactly where I wanted it because now you have non-IT folks coaching those that were resistant to it on how to better use it, and that was key. Once I had the two ends of the spectrum, it was very easy for the entire rest of the flow to happen. In other words, I was able to wrap in the whole rest of our agency in one fell swoop because I'd already handled the problematic or resistant potentially users, and I'd already handled and I'd gotten buy-in from those that were pro-AI or pro-technology, or in many cases, some of them didn't care about technology at all, but they were pro-productive, and so they were able to see the efficiency and the productivity enhancements that it brought to the IT department, and that's what pushed them to accept it so readily, and that was a major win in our case. Amazing. Thank you for that. I do want to invite our audience, if you do have questions, please feel free to drop them in the Q&A and we will address them. This has been a really interesting conversation and we've got a lot of additional questions to continue, but if you do have a pressing question for Sid, please go ahead, put it in the Q&A and I'll make sure that we ask it. I want to move on to another question that comes up very often, responsibility and guardrails. Bringing in autonomous resolution into, I'm guessing, a government IT environment isn't just a technology decision, it's a real responsibility on your shoulder. I do want to ask, what kind of guardrails did you put in place and how did Robin help you ensure that you have those guardrails, visibility, and trust? Texas is not necessarily unique, it may not be unique now, but Texas recently passed legislation on how you may work with AI in a government role. We were utilizing Robin slash HAL before that. I am pleased to say that, again, being a little nervous about bringing AI and knowing that there were potential naysayers in the organization, one of the things that we really did was look at compliance heavily. We put some very, very strict guardrails on it about what data it could access, what data it couldn't access, what file systems were off-limits to it. So I am pleased that in our testing, all of that function falls. I can't tell you that occasionally a user didn't cut, copy, and paste something into it, it probably shouldn't have. But the good news is we own all of that space. So it's not like that information can get out anywhere. Even within a Terra, it's an owned space and we put a strong enough guardrail in there that if it does encounter that kind of proprietary data, it alerts us but also wipes it from its banks and that kind of thing. But at first, we have since loosened the rails to be in accordance to not be as strict. We're still well within our legislative guidance, which was actually wider than what we had narrowed it down to begin with. But that was a challenge. We met with legal several times, we met with HR several times, we have a little bit of HIPAA compliance. So we brought in legal on that. It was very much a matter of making sure that the one, that we drew those rails tight and two, that that data was owned by us, that that workspace, that environment was ours and it couldn't be cross-populated. Even sharing within the organization was a no-go. So we were very careful about what it was able to retain for utilization in other workflows or other runbooks, as well as to what information it had access to. So you did your deep homework and a Terra was here to support, obviously, with all of our security and we're on Microsoft Azure, we follow the AI principles of Microsoft. We are ISO certified. We're probably one of the first companies to get the ISO 42001 AI certification. So definitely, we work very closely with partners and clients like yourselves to ensure and make sure that you can get this through our security and your security. We talked about the role. Go ahead. I was just going to say the engineers were invaluable because we'd come back from legal and we're like, okay, because it wasn't always very clear-cut. So we shoot off an e-mail and they come back with suggestions, and between the suggestions, we tweak it and then we'd turn what became the rules back into an executive summary or a case presentation. We present that back to legal, legal would look over it, give it the okay. So without the deep knowledge and the insight that a Terra's engineers brought to the table, it would have been a real struggle, but that was a huge time saving for us as well, almost being able to throw that out to y'all and then a Terra giving us back answers and guidance and suggestions. Y'all were really great on that. So it really worked out. Thank you. There is a question here from Robert about whether there is an on-premise hardware. The answer is no. It's an AI SaaS service, but a Terra does have an agent that sits on the end device, which Robin uses on the device to diagnose and capture telemetry, but you do not have an on-premise of a Terra as such. It's a SaaS application. Moving on to the next question. Oh, sorry. Maybe I missed this one. Oops. How long before you saw measurable impact, right? So you did mention that you wanted to apply everything to AI before you had the confidence or before you would actually test it. In hindsight, you would have probably started earlier, but how long before you actually saw measurable impact and how are you measuring the impact? I would say we took probably a good 45 to 60 days doing the initial setup and working with our first pilot group. And again, that was far too long. It really didn't need that level of detail and scrutiny that we put it under, but we felt better safe than sorry. So 60 days there, pilot group was using it, getting buy-in from them. Yeah, I'd say easy within 60 days. We brought the next group on. They ran about two weeks, so 45 days there-ish. I would say we've noticed the real decline. Okay, well, so there's two things that created a huge decline. So right around the three-month mark, we had it rolled out to the general populace. It was on every desktop. It's every client of ours was using it, and we noticed, obviously, a productivity increase. We noticed it was being effective by those that would use it, but by the same token, we had a lot of people that just wouldn't bother, and they would go ahead and just put in a ticket. Yeah, and so I was talking about it with the Atara engineers, and I happened to say something along the lines of, man, I wish I could not give them a choice. I wish they had to go through Hal slash Robin to even set up a ticket. And Atara's engineer said, yeah, we can do that. I said, what? And they said, yeah, we can do that. I said, okay, well, let's flip the switch. So we went in and we forced all of the users to now start with Robin and work the case with Robin, and then if it was too complex for Robin to solve, then they could, you know, essentially Robin would put the ticket in for them. And was there grumbling? Yes. Were there some nasty stares over the lunchroom tables? Yes. Did I see about a 25% increase in productivity? Yes. Immediately it jumped up because now people were forced to give Robin the chance Robin a chance to do for them what we wanted it to do. to do for them what we wanted it to do. Took a little getting used to on their part. Some more than others were a little recalcitrant and trying it out. But yes, at that point when I turned on that feature, it skyrocketed. We had seen hints of it, hints of what it could be. We had a glimpse of the promise, but when we began forcing everyone to it, yeah, it really took off. Not that I'd recommend that for everyone. I was often worried that I'd come out to four slit tires, but luckily my people are nice and of course we had done that. But yeah, there were some grumblings. But in the end, everyone got used to it very quickly. Everyone began to appreciate it. I would say if I have anyone recalcitrant nowadays, it's typically a new hire. Often that's handled within the department by them just explaining, no, this is good, you should try this. One of the things that was a big win for us was we went from being a, which was unusual for me, for 21 years I dealt with being on-call 365, 24 hours a day, seven days a week, 365 a year to being in more or less a 7.30 AM to 5.30 PM shop with no after hours call. That was very unusual to me at that time when I first took over here. But one of the things that HAL did that our users really did appreciate was we have a number of them that work from remote offices or from home. The fact that they could seek out IT help with issues 24-7 now, and not just between 7.30 and 5.30, that became a big draw as well. That was a win. That was definitely a win on our part. Absolutely. I'm a great believer that technology alone doesn't drive transformation. It's people like you that do, and then obviously, you need to apply to a change management, right? Like how do you anticipate the resistance? How do you build in that change management even if it's forced? And at some point, maybe you do need to flip the switch at once because that's the only way you can get people to operate. Key question from my side is, you know, we talked about the end users, and obviously, Robin is key to the end users, but there's also the technician side, right? Your team, how was their receptiveness or perception when you said, I'm bringing in an extension to the team, I'm bringing in a tier one AI technician that's going to handle, you know, those repetitive? What was their first reaction to this? I was very careful in how I presented that because at the time that we came on board, there was still a lot of talk about people losing jobs, especially in the IT industry, to AI. And one of the things that sits our region apart and one of the reasons I needed Atera's type of tier one, you know, agentic agent to help us is because we're very, very rural and agrarian in this area. And IT talent tends to congregate in large cities. That's where the jobs are, that's where the pay is. So finding skilled individuals with a solid IT background in our area is challenging to begin with. And the more remote you get, the greater the difficulty is. So knowing that, you know, I certainly didn't want to present to my talent any kind of issue. I needed them firmly on board with this. Once I sat down with them and kind of explained my vision, my goal, how I wanted it to fit into their workload and their work life, it was pretty much instant reception. I mean, they were really on board at that point because they saw it as them being cut loose from routine work and able to do more productive and challenging things, which is what they wanted to be doing. And, you know, consequently, it's what I want them doing, too. So that alone got me the buy-in that I needed. So I didn't have to run that gamut of the whole, you know, does this mean AI is taking my job kind of question mark. It acts as a force multiplier. It's not a force replacement. It does a great job of truly allowing for those things that are challenging issues to be what's presented to your technicians, making them far more effective than they were before. Again, force multiplier. I can't express that enough. It really has made a difference in how we handle things and the amount of work we're able to get done in a day. Amazing. So bringing them on board with all of the value and what's in it for them and then how they can probably start their Mondays not with a full backlog, but rather with that, as you said, the first 25% already that happened. And if they are diagnosing tickets, those are already coming in since you're forcing users forcing. You're encouraging users to start with Robin. It's already collected all of the information and they don't need to do that back and forth. And of course, the reality, a lot of your employees are out there in the field. You don't want to be doing site visits, as you said. So anything that you can take off their plate so that they can do that high value strategic innovation work that you were brought in is key. I think what better to end from my side and of course, again, to our audience, if you have questions, this is your opportunity as we're coming to the end of the session. What advice would you give to decision makers on this call? Words of wisdom, especially maybe those in public sector, what do they need to think about before they start? And then what opportunities might they be underestimating? Hard question. Yeah, it is a bit because I certainly saw the value in it very early on because I have very few technicians, I have a lot of users and I have a lot of mileage to cover. Again, our territory is just slightly smaller than the state of South Dakota and that's pretty big. So for me, it was a good choice because of our rural nature, the size of our territory, the fact that AI talent is hard to come by, those all factored in. And part of that was, I want to keep my people engaged, I wanna keep them active, I don't want them getting bored by the same old trouble tickets and that kind of thing, I want them to actively be improving themselves and their skills on the job. For me, Robin was the stopgap that automated that, that let them move off of the typical routine tickets and move forward and try more challenging things and try stretching their legs a bit more. Not only is Robin a great resource for our clients, customers, but Robin does a great job of helping technicians diagnose issues that it couldn't fix on its own. Again, because all of the information is carried forward in the ticket, the technician can interact with it and help narrow down things and help look at other things. All of the information that Robin gathers about the PC, the situation, the application issue, all of that's right there at the technician's fingertips. Same thing, I've got one guy who, his favorite thing is teaching Robin how to do more things. He loves putting together run books for Robin to learn how to do new tasks and see problems that we only see rarely, but are they a problem? Yeah, but he likes putting together those run books. So, hey, more power to him. I would have to say that if you're a decision maker, one, you have an environment where you would actively benefit and having that kind of force multiplier in your possession. And I think, I can't think many environments that could, but buy-in, you first need to get administrative buy-in about it. You need to do your homework, make sure you can explain why it's safe, why your corporate data, whether it be HIPAA data or CEGIS data or personal information, any of that, how is it going to be safeguarded? Because that'll be some of their first questions and you need to be able to explain that. Then I would say, show them the value of it, let them experience it for themselves. Make sure that they see that it could act as a force multiplier for you. And if you've got that administrative buy-in at that point, there's really not much left to do except implement. So, that's kind of the advice I would give, is make sure you have the right answers before you walk through the door, because their concerns are gonna be cost, security, and what's it gonna do for us? How well will it do for us? What kind of force multiplier are we talking about here? I think in my environment, it's somewhere between a fourth and a third of our tickets that come through, which is good. And it's hedging its bets as we speak, because it's learning more about our particular environment and our solutions for certain criteria problems. And yeah, I would, like I said, the main thing, buy-in, but answer the security questions first. That's your homework, if you're gonna want to implement it, make sure you have those answers. Amazing, Sid, I wanna say a big thank you. Really what you've shared today is exactly the kind of real account that this conversation needed. A look at what it takes to move from reactive IT to autonomous IT in an environment where resources are stretched, and obviously the work never stops. And for me, the theme that I'm taking away is that this isn't about replacing your team. Definitely a force multipliers, you said it's augmenting them, and it's about finally giving a small team the leverage of a much, much bigger one. So I really, really appreciate you sharing with us and opening the doors of DETCOG and what you've implemented. I do see a final question, even though we're just about to end, and I'm not sure, maybe Sid, you can answer this one. From your perspective, was there a bandwidth increase requirement in order to allow your end users to engage with the AI? No, I mean, now, we have a fairly strong bandwidth. We have a fairly big pipe. I believe we have, I don't wanna get this wrong now. We have two gigs up and down, and for us, it didn't really register. I mean, we do monitor that, and I haven't seen any huge increase in our internet utilization since we implemented it. I'm sure there's some, but whatever it was really didn't tip the scale here. And so, and that was a concern going into it. I won't lie, especially, because I really, at the time we were doing it, really, at the time we were doing it, the metrics on it were kind of hard to come by. I no. But what I can tell you is it's fairly negligible from what I've seen. So there you go. Wonderful. Thank you for that. And with that, I want to say, first of all, a big thank you again to you, Sid, for taking the time to join me today and discuss how you've implemented Robin at your organization. I want to say a big thank you to our attendees today. This session is being recorded, so if you've got, you know, other teammates that you'd like to share with, we'll be sharing the recording in the next 24 hours. I invite you all, you know, tune in to our webinar page. We do a lot of ongoing webinars with clients. And if you do want an in-depth demo, we didn't use this opportunity to zoom in on what Robin is, you can click that request a demo button, and our team would be more than happy to give you a tailored demo and walk through your unique use cases. So once again, thanks, everyone, and hope to see you on our next webinars. Thank you. Goodbye, everyone. Thank you.
