What if you had a CEO Companion AI Agent?

**[00:00:00]** Hello and welcome to the web talk show. My name is Armando Piscareno and today we are talking about a companion AI agent for CEOs. So stick around if you have a small business or a business in general. I think this will be a lot of fun. Welcome everyone. If you have any comments, questions, suggestions, please drop them in the chat in the different platforms and I'll make sure to be looking at them and be able to respond live. So, we are currently live on YouTube X, LinkedIn, and Instagram. Welcome again, and let's get on with it. So, there's all this talk about AI and AI agents, and there's voice agents and there's all these different tools, but how can they really help me in my business? And what I'm going to talk about today, it's not exclusive to business. You can take this also for just anything in reality if you want to have some sort of assistant. I'll explain a little more in a second, but it does have very very big benefits in business and that's what we're going to talk about today. So, I was talking with a client yesterday about the potential of having an agent that helps you keep track as the business owner or leadership of what's going on within the business in general. Of course, you can have your dashboards, you can have your Tableau, you can have your PowerBI, you can have anything you want for business intelligence, but that typically means you have to either open it in an app and drill down and your people have to make your reports and you have to then find the reports and open them, etc. And if you're walking around the plant or you're out in the field, **[00:02:00]** then it becomes a little more cumbersome. Especially, let's give you the example of you're walking with a client and you're you're going through the grounds and you're just chatting and they ask you something and you're like, "Oh yes, let me pull it up." And you have to open your phone and even if you have the dashboard, you have to open the app and then go in, find the dashboard, wait for it to load, and then once you have it, read it. And then you can tell them, "Yes, we saw these numbers, etc., etc." So you there we what was that like four different steps we had to do just to get started. So, is there a better way? And that's what I'm here to try to expose today. What can we do to make it a livelier experience, a more human, funnily enough, experience getting your data? Because remember, all these systems, all these computers, they're just mechanisms that have helped us make things faster, better, have more efficiency, etc. We can process more data than ever before. We can have the power of computing in our hands, in our glasses now even. But the interface has been evolving and it used to all be typing and then there was touch and now there's some visual aspects and auditory aspects. And where we're headed is a fully natural languagebased interaction. And that's where it really gets interesting because far from your home Alexa or your Google Home or your Apple Siri, you can have something that actually helps you do things and achieve your job in a better way. So, we're going to talk about some examples of how this can go about. And going back to my conversation yesterday, if you had something that **[00:04:00]** interacted with your different systems, with your different data systems and your databases and everything, you could then have an agent, a voice agent for yourself, for your team that you can carry around with you. So it could be on your phone, it could be in the car, it could be in your computer, it could be anywhere. But basically, you just talk to it and then it could talk back and give you the data back just talking with it. or it could text you back if you want to get some uh actual data that you want to view. So again, far from drilling down into the reports and doing it all manually, even if you have those reporting capabilities, having it immediately, it makes such a difference. And I'll give you some example just to it's so it's so different when we actually think about what the use case might be. So, imagine you're in a plant. You go to one of your plants and you're walking around and everything seems to be working. Everyone's working hard, but you are curious and you're thinking, well, the numbers didn't add up as much as I expected yesterday that I was looking at the numbers. Now that I'm hearing the plants, I want to know more. So, if you pull out your agent and you might, again, you might have it in your phone right here. So you could just basically open it up and say, "How is our production today here at the Selma plant?" And the agent could then go into your different systems, figure out what the production values are, what they should be, how far off they are, and give you an actual real response. It could actually analyze some of that data for **[00:06:00]** you, and give you a relevant response based on you walking around the plant and asking it. So, it's no longer about just having data, which again, this is something I've discussed a lot this week with different clients. Sometimes you have a lot of data scattered everywhere. You just keep gathering data through the years, but you don't really use it as much because putting it in different systems, creating reports, it does add a lot of overhead. So when you can put something that can understand your queries in natural language, just talking to it like a human being like you would ask one of your directors and it processing it with the power of computing and bringing it back and then digesting it and giving it to you in a way that you a CEO or CFO might need. It's super valuable. So, I'm going to give you some examples again just just for fun. And by the way, this what I'm going to show now is funnily enough I and I was talking about this the other day as well created a little dashboard in real time right now just for the show um to help out with the live stream. And this is this is actually well let me go let me show you what I mean. This is so fun. This is the type of thing that you can do to make it clear that it's it's ridiculous what we're able to do nowadays. So, what I'm going to do here, I'm going to share my screen and let me see what uh I think I will go into Claude here and I'm going to share it. Yes. So, you see what I have here on the screen where it says manufacturing and production. **[00:08:00]** So, I was going to go on this live stream and talk about some of these examples, right? So, here's an example. I'll get into this in a second. So, let's say you're walking around and you say precisely what I said before. How's our production quality doing in the SEAL plant? And then the thing could reply to you in real time in voice. It could just say, the sema plant is performing well today. Current quality metrics show a 98.4 pass rate, which is above our 97% target. We've had 23 defects out of 1,4337 units produced so far. And then it could show you today's stats like in actual in a table or something like that as of blah blah, right? Or it could go in and you can maybe you're walking around again and you're like, "Oh, hey, what's the current output of line three?" You're just you're just asking it right on your phone. It could just come back to you with a very success answer. Line 3 is currently running 847 units per hour, which is 96% of our target rate. We're slightly behind today due to a 22minut changeover this morning. So as you can see this can be very very helpful for a CEO again as you're walking the plant as you're doing anything like that. Now, for those of you paying attention, if you're watching versus listening, you'll notice that on screen right now, I have this table that shows manufacturing and production. And then it shows some checkboxes and some messages and something that I'm reading from. Well, do you think I spent a lot of time preparing this this morning? No, I did not. I told the AI, hey, I want to talk about these specific points in different **[00:10:00]** parts of the business. And I'm going to go through this, of course, and I want a way to visually go through these questions and be able to say for my examples. Again, this is very meta, but I'm explaining what I literally did this morning. What are today's numbers? And then I want to be able to click it and see an example in real time. Hey, we're having a strong day. The sales are tracking 18% ahead with a target of 127,000 uh in revenue so far. Here's the sales as of you're seeing the tool that I'm using right now to keep track of what I wanted to talk about. I literally built this quick dashboard this morning. Well, AI built it for me where I have the questions and the topics that I want to talk about and it put it with little checkboxes that I can mark as I'm talking about things. And if I click it, it actually gives me the response for that question. If if I didn't want to invent it, I wanted to read it, I could just grab it from here. And then at the top, it's telling me how many topics have I covered from the topics that I wanted to cover today. So again, not related to the agent that we're talking about today, but interesting nonetheless because as you can see, these tools can allow you to just make something, create something out of thin air that you can use and dispose of afterwards once you're done. Will I need this checklist of topics for a live stream later? No, I won't. I just it'll just stay there. If I want to make another one another day, I'll just make another one. So, it's so interesting what **[00:12:00]** we can do nowadays with all this that it's just it's just amazing. Anyway, let's get back to some of these. And again, you have any questions, please drop them in the chat. I will open the chat here and if I see any comments coming in, I will be happy to answer them as well. So, what are other things that you could ask it? So, let's say again, you're a CEO, you're walking around, you want to see what's going on. Maybe you'll do something like, "Hey, any equipments showing signs of failure?" And then it could say, "Yes, I'm seeing two pieces of equipment that need attention. One is urgent and the other is a heads up for next week." And it could actually give me a table and show me it text me a message with an image or a table and it could show me the little red lights and yellow lights and it could say urgent hydraulic press. Number two, vibration levels are 40% above normal. The oil temperature is elevated. It's 185° versus the normal 165° and it's been degrading for 3 days. The predicted failure is within 48 hours. The impact would halt line one and line two. We recommend scheduling maintenance today. Now, the other one, we're going to monitor the conveyor motor 5B. It has some noise in the bearings detected by the acoustic analysis. still within operational parameters. Predicted failure 7 to 10 days and the effect would slow down packaging by 30%. We recommend scheduling for a next week maintenance window. Again, imagine you're just literally just walking the plant or you're in your office, you're talking with someone, and you just ask it. You raise your phone, you ask it, and it'll go into your different systems, **[00:14:00]** get the data, and then give you something like this already summarized, ready for you to take action. So, it's not only about asking it for data. It's also about doing check-ins and doing proactive things. Obviously, there's the sales and revenue side of things as well. You can just say, hey, which products are trending up this month? And it could say three products are showing strong momentum this month with the premium bundle leading the way. The top trending products from October 1st to 17th are the premium bundle. Its growth is 47% versus last month. The revenue is 89,400 with a 28,000 increase. Drivers are the new marketing campaign plus the 50% promo. And the Model X Pro, the growth is 31% versus last month with revenue 156,000 with an increase of 37,000. And the drivers are positive reviews. So all this again as a business owner so helpful because it's no longer just seeing numbers and seeing tables. It's getting actionable data right there in your hand and not even in your hand. You might have just your AirPods and you're walking around and you suddenly think of something. You're like, "Show me our pipeline by region." And it could send you a message and show it to you, right? Or you could do something like what's our current inventory for skew 42891 because maybe you're you're in the stock area and you're looking at things and it seems like you have a lot of something or it seems like you have very little. You could ask it and it could say, "Oh, the widget pro assembly is in good shape. You have plenty of stock across all locations." Then it could go into more detail if you wanted about the stock and how it's been **[00:16:00]** moving. You could say things like, "Do we have any shipments delayed this week?" Yes, we have three shipments delayed this week. One is creating a customer issue. Now, this one is a very good example because you ask it, it tells you, "Yeah, yeah, we we actually do have and one is creating a customer issue." So that is very important for for you as a business owner because you will want to address that or you want your team to address that immediately as soon as possible. Make sure that the customer is happy because you always want happy customers and you might not see this because you have your ticketing system over there and you have your inventory system over here and your shipments and you have HR over here and you have your data stuff over here and there's a disconnect many times. So if you have something that can look in the different systems and do that relationship in real time and say okay there's three and now for those three I just have to query those three in the different systems to see if there's any correlation and then you'll see oh yes one of them is having a customer issue we should address that. The same goes for weather and other things because maybe you say, "How are shipments going?" And then it would say, "Well, we're down 11% from previous week." But then it could also check the weather and other things. It could say, "But there is a current weather issue. There's a hurricane in this zone and that's why all the trucks has stopped, etc." Right? So many different things that having an assistant like this can really help you with. Other things could be customer service. You could just **[00:18:00]** ask it, hey, how many open support tickets do we have? You currently have 47 open tickets. Volume is normal, but the response time is a bit higher today. Right? Again, very actionable. Then you can ask it something like, what's our customer satisfaction score this month? Customer satisfaction is strong this month and improving. were seeing particularly positive feedback on response times and and it could show you again it could send you a text message. This is the fun part. You can do multi- uh platform and it could show you the reviews, how many fivestar reviews, how how was the rating and things like that for different products and just super valuable data. Yes, you have to have access to data to do this stuff, but many businesses have the data. they just don't know they have it or they have it siloed. So you might have a lot of information. You have information in bank statements. You have information in spreadsheets. You have information in your CRM. You have information in your ERP if you have it or your Air Table or your notion or your Monday or your House Call Pro. All the different systems that you use to run your business have data and it's just a matter of connecting them. And there's many ways we do that all the time. We connect different systems that might look like they don't work together. We can actually get data from one, push it to the other, aggregate all the data, put it in different places, and then of course we can build something like this where you have an agent that is not overly complex. That's that's a very interesting part. Sometimes we think, well, this is Tony Stark level stuff, right? No, it's not. **[00:20:00]** We're yes, it's fun in the movies and all, but the having your own little Jarvis for your business isn't far off like it used to be because of how LLM's large language models evolved so fast and got so accessible that now it's relatively simple to create an agent that talks to different platforms because it's two different things. There's one part that talks to the platforms and then those tools you can then connect into the actual chatbot or agent or voice agent that then does the whole conversation part. So it doesn't have to be that the agent has to talk to your data. No, you separate. You have one part that abstracts all the data, gathers all the data, analyze, etc. And then it's just fed to the agent whenever it needs it. So very useful. Some other examples are in the financial part. So you could say what's our cash position? Say your cash position is healthy. you have strong liquidity though you have a big payroll coming up on Monday and then it could give you the actual numbers there right give you your cash position the bank accounts operating accounts of payroll accounts reserve accounts give you accounts receivable um and then just give you all the data that you need in that moment in time because again you might be having a conversation and if you're with another maybe you're with a business partner that's coming from out of town right you're having a meeting or you're walking around again or maybe you're even at lunch, questions might come up, right? Hey, how how are margins for for this widget? And so the thing could tell you, oh, your product margins are are solid. They're ranging from 32% to 49% and this **[00:22:00]** XYZ product is your profit leader. Um, here are the numbers. Right? Because calling someone on your team to figure this out would mean calling out, making sure they're there. they have to go to their computer, they have to generate the report, they have to give you the data. Whereas this would all be real time. It would just give you the information in that moment. So, it's really neat. I I get excited because I've seen it work in different businesses and just the fact that even if you have very basic data, and this is never considered less important data, it's just basic data. Like if you just have your appointment schedule for all the visits that you're going to do. You're an electrician, for example. You have all your visits. Well, it's still super useful that you could be driving and just ask it, "How many appointments do I have today?" And I could tell you, "Oh, okay. You're booked solid from there doesn't seem to be a lunch break. Would you like us to try to move one of your appointments half an hour back so that you have a lunch break?" Right. that could help you out with even small things like that that just make your life easier, more comfortable as well. You can also use it for things like HR and workforce. Things like who's out of the office today, what's our current overtime hours this pay period. And this also gets into it's not only for the CEO. Maybe HR has access to something like this and they could just check and see, oh, okay, well, overtime is elevated this period. Particularly in operations, it's costing about 18,000 more than we planned, right? So there's there's so much we could do. **[00:24:00]** Any open position is taking longer than usual to fill and then it could give you more information. Yes, you have two positions that are significantly delayed. The senior engineer role is becoming problematic. And then it could give you some more data like senior software engineer open since July 12. That's 97 days and the target time to hire is 45 days. So we've got a 52 day delay on it uh which is quite extensive, right? So then you can go in and try to figure out what's wrong. Why aren't they hiring that position? What's wrong? Is there nobody there in the market? Are we giving a less than ideal package? What's going on? So, if you're dropping just right now, just in the conversation at this moment, I'm talking about having a conversation agent for a CEO, right? Like a companion agent. So, there's all these tools out there. There's ERPs, there's business intelligence systems, there's a ton out there that allows you to get data and analyze data. But even if you have the most advanced tools, you typically also need to have someone on your team to actually build the reports that then you as a CEO or stakeholder or president or vice president needs in order to understand the data. And this is where it really gets interesting. If you have a voice like a CEO companion kind of voice agent, then all this could be worked in a way where it's dynamic and it talks to the data. So you no longer have to build out a big report. You can just tell it, hey, I need a report of this parameter with this other parameter crossed with this other thing and the weather and it could generate in real time, get **[00:26:00]** all the data, analyze it, then give it back to you in a digested form without being an enterprise level cost to build or enterprise level implementation time to build as well. It's relatively quick. It's even faster than building an app, right? So, this is what the availability of the LLMs and how far they've gone gives us. A lot of people just think LLMs and Chip. It's just a fun gimmick. Yeah, because I use it to ask for recipes or I use it to generate my emails or whatever. Yeah, you can use it for that, but they're really powerful for many other things. So again, far from oh yeah, I wanted to write a book and then sell the book. It's more of a it understands natural language so well that I can give it a knowledge base and ask it about that knowledge base. And that's another topic we can discuss. But just with that, it's so powerful because I can have a manual and then ask it for something and I can read from the manual and I can get the response immediately instead of having to go through the manual. But talking about business, it goes even further because you can have tools and these tools can talk to discrete systems. There's no hallucinations there because it's really actually asking it. So you ask it for let's say any unusual activity today and then it could say yes, I've detected three unusual patterns today in the warrant attention. The let me see we have a high priority website traffic. The pattern is down 42% versus normal. The root cause is the CDN is having performance issues. And it's not making this stuff up because it actually does a query. And for those of **[00:28:00]** you who don't know how all this works behind the scenes, when the LLM does a tool call, that's what it's called. It actually runs a web service call. So it behind the scenes, it's like if you called an API. Again, it's a little technical, but the system calls an API endpoint. The API endpoint does whatever its magic has to do, brings the data back in a structured way in JSON most of the time. And now that means that the LLM can understand exactly what came back. So these are the items, these are the parameters, these are all the variables, etc., etc. And then it could just read that and then summarize, analyze, give you the data right there. But it's looking at data at that moment. And that's where it's really powerful because again it's not like whether it's going to be valid data or not valid data. It's actually querying that data and then using it just in the most basic part of their functionality which is understanding language and summarizing and all that sort of stuff. So really nice and what I've been doing is just going over examples of of things that you could use it for. So I'm just going to go over a few more. Another example would be you can ask, hey, what's different from last Monday? And you could answer, comparing today to last Monday, October 10th, there are several notable differences, mostly positive. Today versus last Monday, the sales today are 127,400. As of 11:45 a.m. Last month, they were 98,200 at the same time. So, there's a 30% variance. And the driver is the premium bundle promo that was launched. The production today is 4,238 units and last Monday was 4,180 units. The variance is 1.4% **[00:30:00]** essentially flat. Then for support tickets, we have 47 today. Last Monday we had 62 open tickets and then the variance is negative 24% or went down. then that's probably driven by the app butt fix that we have the we had a bug in the app and once that fixed the lower it lowered the volume of website traffic now there's also the 3,847 visitors and last Monday it was 6,000 620 visitors there's a variance of 42% and this is concerning so see anomalies and then it could go into different things so imagine having all this information at your disposal anytime, right? You could ask it for hey just give me all the system health indicators and it could tell you most systems are healthy but you have one degraded service affecting the website and it could explain all the infrastructure and every there you could ask it things like are there any KPIs trending in the wrong direction and it could tell you I've identified three KPI showing concerning downward trends customer acquisition cost needs immediate attention and then it can explain the CAC is currently 247 per customer But last month it was 189 per customer. So this is a 31% increase. It says wrong direction. Yeah. And so it's rising steadily for 6 weeks. And that's where it's really interesting again because remember it could fetch the data from your different systems. And then once it has the data, it can very quickly with that data that's already condensed, it could just say, "Oh, wow. Okay, so I have this month's data, last month's data, maybe the last six months data, and then I can very quickly understand how the trend is looking, where where we're coming from without having to again generate a **[00:32:00]** big report and decide which columns will go and which tables match with these other tables, etc., etc." you obviously lay the groundwork when you're building something like this, but then the the whole point of it is that it's dynamic enough that you can just go in and ask it things and that's where it's really fun. So, going back to manufacturing, I mean, just imagine going down the line, you just go, "Hey, show me the status of all the sensors on the packaging line." And then I would say, "Here's the current sensor status for the packaging line. Everything looks good except for temperature sensor 4, which is running slightly warm." And then it could give you either tell you the details or send you an image via text. And it could say something like, "Hey, here's a temperature sensor one is green, 72 32 degrees Fahrenheit. We've got temperature two in the zone B. That one's also green. That's 70 degrees Fahrenheit. And the target is between 68 and 75. So, it could I'm seeing visually here, but you could have it actually send you a text message with an image of the little lights, green, yellow, red of all the plants. Um, so it's there. We haven't even scratched the surface of what's possible with these tools. If you just think outside the box a little bit and consider what would I like to know in this time of day when I'm doing this and if you just think like that and then write it down and think back and say what would need to happen for me to get that answer well I would have to call finance I would have to call HR I have to call uh inventory and get certain data **[00:34:00]** from them and then I would need to have someone in uh data analysis team create a report get that data together and then send it over and I would receive the PDF and then I would look at it right so maybe those are the steps so once you look at all that and then say wow that data would be very helpful I just don't get it because it takes a long time or it takes a lot of resources to get it well if you had an AI companion agent you could just do it directly just ask it and again another example is hey how downtime did we have this week? Oh, we've had 4.2 hours of downtime this week across all production lines. That's actually 15% better than last week. So again, if you just have the downtime, just have one data point, it might feel like a lot, but once it tells you it's actually better than last week, then oh well, that's that's interesting. I I I didn't expect it to be the case, right? or hey, show me our pipeline by region or show me inventory turnover for 10 products. And it could go through 10 of the products and explain the different product turnover. Um, you could ask it about just a specific question again. Hey, what items are we about to run out of? And it could say, you have five items approaching low stock levels. Two need immediate actions. And then it could again send you a message or just tell it to you. And it could say critical this particular skew which is a bearing assembly we only have 47 units and we should reorder 200. The daily usage is 23 units. The stock will be gone in **[00:36:00]** two days and the lead time is 5 days. So we have to do a rush order today if you want to make it. The control board has also 156 units and the daily usage is 31 units and the stockout is going to be in five days. So this sort of information aggregated put together for you in real time so that you can make a decision is super useful because it just allows you to work better, be more efficient, focus on more important things like actually making business decisions and being creative and doing new initiatives. And it's all possible today. like you could build these types of agents today. It's not science fiction. We've literally built tools like this already. It is just a matter of connecting different tools to your different data sources or gluing your different data sources with something like NA10, things like that. Then putting it together, then creating an API for it and connecting your agent to it. And then you could just talk to it. We were experimenting with something last night for a client and it it was so fun to use it for the first time for some things because again it's just the difference of seeing the data versus talking to the data. You don't really see it until you actually experience it. So if you haven't tried this, open Chat GPT or whatever tool of choice, but Chad GPT has a good voice one. If you open the voice, there's a little voice icon or mic. It's not a micro, it's a you'll find it, but there's there's a little part in chip that allows you to talk to it and it'll have someone. It's an AI agent, right? But it's sounds like a person and **[00:38:00]** you can just chat with it and ask it things and it'll know everything that a Chibt knows, right? So, you can ask it about different things. Um, but just the experience of talking to it and how fast it responds and how accurate the responses can be. Of course, it can hallucinate and do all sorts of things, but just in general terms having the experience of talking to it might help you see, wow, okay, so this this is actually a thing like this this is a possibility. Now imagine having that power connected to being able to gather data from your system privately of course and being able to ask it things and it give you results in real time that are already digested for you. And this is not limited to you the CEO. It could then be adjusted so that it works in a different way for your VPs and your directors and your staff and the different workers in the factories. You never know. It it could be expanded in a way where obviously with permissions and roles and all that fun stuff, but considering everyone is authenticated properly, it it's just so helpful because it has access to the same data. So maybe there's a worker on the line and there they had an issue. What happens? Well, they have an issue. They complain about it or something happens. Oh, this part is loose. And then, oh, I got distracted. I went to lunch, whatever. They won't put it into the system until then that evening or maybe tomorrow or maybe they'll forget and then nobody will know and then that bearing will pop and then there will be an issue. But what if at that moment when the worker saw the issue **[00:40:00]** with the bearing vibrating too much could just really quickly open the thing and say, "Hey, um, there's it's there's there seems to be a weird vibration on this bearing on line B. I just wanted to report it, right?" And then the system can analyze that, check how important this should be, escalate it to the right department, actually create a ticket, send it to maintenance, and then maintenance can see it in their dashboard or even get it in their ear, right? And say, "Oh, there's there's an issue with bearing line B. Go check it out." Then maintenance can go and proactively check it. Oh, it needs oil. Oh, it's too hot. We need to change it, etc. before something actually happens and breaks the line. So there's so much space to work with, not only with LLMs and chatbots and agents and all that. It's just it's it's also the understanding the power of natural language processing. the fact that it's no longer just what you've experienced with Siri and Google Home and Alexa, which is not good if we are honest. It's going to get better as soon as they're they're working on it. But with the old style of home assistance, it was I mean, it was a joke. Now with what's possible nowadays, it is really going to change things and it's already changing things, but things are going to be quite different. Uh you can trust me on that. And yeah, it's it's it's going to be an a voice world. So remember, there's the glasses, there's the meta glasses, and now they have the AR version as well. Um, but we're used to using this, but we used to be used to using the computer mostly. Some people haven't ever had **[00:42:00]** a computer. Think about that. There's there's kids who then grew up and they went to college and they never ever used a computer. They just use their phone and that's what they use. And some people don't even understand the keyboard. They're like, "What's this? I'm used to just typing on my with my fingers." And the same will happen. So yes, the phones are great because they help us with a lot of things. They also distract us, but then it's it's moving towards voice. So now you're going to have your glasses, you're going to have a hearing thing, you're going to have a pin or something and you'll you'll just interact with the machine with the computers just with voice and ask it for things because again typing is just a way it's just a human machine interface that had to be built so that we could touch talk with mach with machines and so that evolves so that then you could just be walking around your house and ask it for things or you're like oh seems a little dry turn on the sprinklers, please. And it'll just turn on the sprinklers, right? Um, so it feels a little hot. Hey, um, feels a little hot. And then it should know, okay, where you are in the house, and then it should turn up that part of the house's AC. And yes, of course, you could do it with Google Homeland, but it's like, Alexa, can you turn up the It's the difference of having those passphrase keywordbased interactions versus natural interactions with something that's that's there uh listening to you is is quite quite different, especially again in the business context. So anyway, just like to share some examples of what you could do **[00:44:00]** if you created something like this, having just like a companion agent. It could be chat, it could be voice, it could be whatever, but think of the possibilities as a CEO, having a tool like this in your arsenal to help you be more efficient. And of course, if you are interested in something like that, please reach out. You can send me a DM on the different platforms. I'll be here to answer any questions about like is this possible for a business for what type of business? Do I have the right data? Do I have the right tools? There's there's all those discussions and I'm happy to have those discussions with with anyone who's interested because there's a lot we don't know that we don't know. So, so I'm always happy to help show or or explain what is out there because it might not make sense for your business, but it might make sense for your business. So again, don't hesitate. Please just send me a message. You can find me on all the different platforms. And if you search online, like if you don't know, you follow me or whatever. You just pop this thing up in your feed, you can search who is Armando Prescareno on wherever, Google Chad GBT perplexity uh Gemini, and it'll tell you and you'll probably find a way to reach me as well. So thank you for joining. Happy to be here on the live stream and