Transcription
Everyone thinks they need to be a prompt engineer or a programmer or be a genius. You don't need any of that. The only thing you need to do well is communicate clearly, meaning you need good context. Our guest today left a career in a multinational corporation, founded the largest AI automation school in Brazil, and has already trained thousands of students who earn multiple six, seven, and eight figures. Today, taking a product from Antropic, we have a chat, right, where you go, write, ask a question, do research. We have the coworking, which is for you to automate repetitive tasks, and we have the Cloud Code, which, besides doing all of that, also literally creates anything, from web applications to applications to connect these cameras here. So there are no limits today. The limit is here. >> I can say that we are living in the best moment for the intelligent lazy. >> Exactly, man. The idea to execution is minutes, hours. So imagine that in Cloud Code I have an idea, I start talking to the tool there, I put in some context, and that idea, in 20 or 30 minutes, I can create a minimum viable product. In the past, to build this took time, you had to hire a programmer. Not today. Today, the tool already does it. Today we will understand what Cloud Code is, why it has become the favorite tool of those who operate in digital and technology [music] and how you can start using it now. >> I am T Lorai and the Hotmartcast begins now. >> My dear Tali, you are very welcome. Man, I've wanted to have this conversation for a while. My job today is to hold back the two crazy guys. Why? Why? Because Ricardo, from far away at Hotmart, besides being from technology, is the one who has most put himself into this business. We even gave him a mission, but he embraced this mission in such a way and transformed it into his life's work. >> A life's proposal. Exactly. So, Ricardo is a guy who dove into this a lot, and Thesales is one of the main mentors, one of the people who teaches this the most. One of the first, one of the first to cross this bridge and go from one side to the other when nobody even knew what an LLM was yet. So, today, man, our idea, and if you don't know what an LLM is, we're giving you the news. Right? Today our idea is to focus a lot on cloud, because it's the trend of the moment, and we've been discussing off-air, it might be the trend for the future too. And Thales is one of the main references for this. However, before we get into the episode, Thales, we prepared an engagement action, and you were kinder than average, because here's the thing, traditionally what do we do? We choose the best comment to give something as a gift. You suggested we do something different. >> So we will have something for the best comment. Yes, but you will deliver it to 100% of the people who commented on the episode. What will you deliver? Tell the audience. Everyone who comments on this episode will get an entry to the Cloud Code operation, which is an 8-hour workshop where I show everything about Cloud Code, from the basics to creating skills so that you can automate some task with Cloud Code. >> So some people in this workshop manage to automate their things and hadn't even touched anything like it before. So it's really cool. This. I'll give a testimonial here, because it's not just talk, it's hands-on, in fact, in the physical immersion. >> It's operating. >> You leave there operating. Thank you very much for the gift, my friend. Of course. Together. >> Oh, Thales, let's start from the beginning just to set the standard, because our idea today is that the conversation should serve the crowd that is already playing, that is already operating, that has their hands on the work, but also those people who are still in doubt, a bit insecure. So, let's start from the basics, which is the following. What is the big difference of cloud code, not cloud code, but cloud, in general, in the evolution of tools in general, and why is it different today from ChatGPT, which is where things started, right? >> Okay, let's go. Ah, Antropic today, in my opinion, is one of the best artificial intelligence companies there is. Why? Because besides being very good, obviously it leads this market. I consider it to be like Hotmart in its niche. Perfect. >> Hotmart leads, it brings innovations. Antropic does the same thing. So what Antropic does, others implement later. Exactly. It brought some protocols there, everyone copied, and today it's a market standard. So, because of that, it's, let's say, the absolute leader, but not just because of that, right? While everyone was just fighting for the customer side, right? Like ChatGPT for you, Gemini for you, what did Antropic do? Oops, let me look at the company side too. So, where it grew the most was on the business side too, for companies. >> Yes. >> And at a certain point, ah, they had an employee there, Boris, he created a tool called Cloud Code for his personal use. But that worked so well that other developers at Antropic started using it. >> They thought: "Man, what is this? You're getting things done very quickly. [snoring] Wait a minute, let us use it." Then they started using it in his team. The thing worked so well that today Cloud Code alone is responsible for a slice, if not the largest slice, maybe the second largest slice of Antropic. And the most curious thing is that Cloud Code today creates Antropic's other tools. So Cloud Code created Cowork, for example, Cloud Code created I don't know what. If you look at the release notes of an Antropic product, man, >> it's surreal. Every day there's an update of something. Almost every day there's an update of Cloud Code. So, how are you going to compete with a company like that that has this product delivery speed? It's almost impossible. >> So they are very good at trends, at innovations, at driving things. They are also very good at execution, right? Now, one thing that perhaps those watching us might get very impressed by the interface or the use of GPT, [snoring] of Gemini, right, which is that thing where you write, write your command, and so on, what is the difference and eh from normal cloud >> to cloud code, so that people can understand? What is the main difference? Okay. Today, taking a product from Antropic, we could divide it into three. We have a chat, right, where you go, write, ask a question, do research, just like a ChatGPT, >> right? >> We have the coworking, which is for you to automate repetitive tasks and do some other things there. And now we have Cloud Code. Not now, it's been around for a while, right? And we have Cloud Code, which, besides doing all of that, also literally creates anything, from web applications, applications to connect these cameras here, Black Mask, so there are no limits, right? Today, the limit is here. So the big question you're talking about today is, since there are no more limits, now the difference between people is who knows how to be clear. The difference, the differentiator is no longer technical, the differentiator is clarity. So, the person who has clarity about what they want to build is now ahead of other people, because AI has leveled this, especially Cloud with Cloud Code, and it allows this to be done masterfully. I can say that we are living in the best moment for the intelligent lazy. >> Exactly. [snoring] >> Not only that. Not only that. >> We are living in this moment, and we are also living in the best moment for people with ADHD. >> Exactly, man. Boris again, the creator of Cloud Code, said this on ex, right? He said: "Man, we are living in the best moment for people with ADHD." Because it's literally for those whose minds are very agitated and want to do 10 things at once. In the past, this was a curse, right? Because of lack of focus, you couldn't do anything. But now, man, from idea to execution is minutes, hours, understand? So imagine that in Cloud Code I have an idea, I start talking to the tool there, I put in some context, and that idea, in 20 or 30 minutes, I can create a minimum viable product. Obviously, it won't be the final product, but you can already get an idea of what it will be, understand? So in the past, to build this took time, you had to hire a programmer. Not today. Today, the tool already does it. That's why whoever has the most clarity and can provide the most context about what to build will win this war. Do you continue testing other things too, like, to say, "Damn, I tested it and it's good or bad?" Or do you say, "Man, I'm not going to waste time on this anymore, let me focus here, because here I already know it works incredibly well." What have you been doing, >> man? Let's go. This is very difficult, right? Because I'm a guy who loves testing everything. Let me give a very quick context that everything is about context. Ah, 20 years ago, when I worked at a technology company, I was an IT manager, there was a website called Super Downloads. I remember, >> Too much! >> Right? Those who are older will remember. Super Downloads. >> Download here. >> I used to download, my hobby was downloading 10 software programs a day and testing them all. I swear, >> there in that [snoring] demo area. >> Exactly. I would choose a category and download. It was my hobby, just like traveling is yours. >> The moment has really come back. Same. >> Exactly. Mine was downloading applications just to see what happened. And then after a year, people would ask me, "Do you know any program to do X?" Of course, XPTO programs. I became a reference. >> Of course. Uh-huh. >> In knowing which program was used for what. It was a curation. Because I did that, right? >> But you waste a lot of time. >> A lot of time. >> So you have to choose the battles you want to fight, right? I was single, 20 years old, today >> infinite time. Almost. >> Today I had infinite time. Today I'm married, have a daughter, mother-in-law, a company, a million things to do. So, what did I decide to do? I decided to focus on one thing, because in today's market it's very easy to lose your attention, because something new comes out every day, especially in the war of. Today there's Cloud, then Gemini 4 is coming out, >> is >> GPT10, right? Perplexity, I don't know what, Open Cloud, which went viral, Manuso. So what did I think for myself, man? I'm going to choose one based on some metrics that I consider important, like which company is growing a lot, which company is innovating, because it's no use, right? There has to be innovation too. And which company do I like too? The philosophy, etc. There's that. >> And Antropic is one of the few companies that has a philosophy, right? It's one of the few, and they say, right, that it follows a constitution. That's why they were banned from the Pentagon, because they were not in favor of using AI for LL for war. Yes. Ah, so there's something behind it. And because of that, I decided to focus on Antropic, because in my opinion, it's the best there is. Any other tool, like Open Cloud, these tools, man, they use ah, they used, in fact, they used Cloud behind the scenes, right? Today, Cloud is the big brain of everything. So I keep an eye on it, I know, but I don't waste any more time. Unless it's something extremely absurd, like, "Man, let me see, let me see," >> unless it's absurd. >> I understand. >> But to this day, I haven't installed an Onepay Cloud. >> I understand. >> Just so you know, [clears throat] >> because I understood that >> Antropic is not stupid. What is it doing? If you notice the updates, it's bringing everything from OpenCloud. So today I can use my Cloud Code here for the phone that's at home, the computer, I can do almost everything. So in a few more months it will be there. Why would I waste time? >> Makes sense. >> So that's a good answer. About four times you said, "That's why I chose, I was going to say Gronk." About four times I was going to make the joke because almost nobody mentions Gronk from [clears throat] from [laughs] >> But I use it. You use it? >> I use it for research. For research? >> For news research, recent news. >> Recent news research because it's all there, right? >> What are the places I research to stay updated? Grocken, Reddit, and one I love, that I don't see anyone talking about, Medium. I love reading. >> Ah, Medium is cool. >> Articles. So I use these three. For me, they are the top three to keep me updated. Now, man, let's bring it down to earth for the crowd that is like, >> how do you use eh today Cloud within your company, right? I'm less about external, more about internal, because I know you've done a lot of stuff about it. So how do you use it to help those on the other side think about how they can use it within their own businesses? Let's divide this conversation into two aspects here: within the company context and personal context. >> Ah, good, good. Let's start with the company, >> okay, >> right? >> So, within the company, I told my collaborators, I said the following: "Look, everyone will use Cloud Code. If you don't want to use it, man, too bad." >> You can't work here. >> You can't work here. Especially because I talk about, ah, it would be absurd, right, to sell something that someone doesn't use. It's like the company's mission. Perfect. >> Okay. And Antropic fits very well with this, because as I said before, it has grown a lot in companies. So now I'm using Cloud for companies. This is wonderful, because any collaborator can use something you share there, like, for example, accessing Hotmart, right? So [snoring] how do I use it today? I use it to create campaigns, to analyze campaigns, especially to analyze campaigns. Want to create a campaign? Okay. Your traffic manager will do it. But, man, what people often can't see is why something went wrong is in the analysis. And sometimes even a traffic manager might not be able to, because it's very analytical data there, right? So, for me, the greatest power today lies in this triangulation of data. I take, for example, data from Hotmart, data from Meta, data from Pixel, from Google, data from what else? from Zoom, because I do things on Zoom. If it were YouTube, it would be from YouTube to see what the peak was, where people entered, where people left, where there was the most engagement in the chat, where there was the least. This is gold, man. >> Totally, a lot. >> For example, what I do most today is create skills. Skills, for those who don't know, literally skill is an ability, right? But they gave this name "skill" to something you create that will reproduce a task you do. So, for example, if every day I have to log into my Gmail and reply, take the email someone sent me, a support email, let's say, and reply, I can train and create a skill, it will reproduce, reproduce this every day. >> Produce this behavior. >> Produce this behavior every day. So, for me, man, this is the only and perhaps the last skill we will learn about skills, right? Going back to what I do for the company, imagine I'm in my day-to-day work. [clears throat] So every time I see a process, and I see myself doing that process manually, I think, "Oops, I can turn this process into a skill." I'll give you a very clear example here. I teach a lot. So every time I teach, I have this thought in my head, "Man, was it a good class? What can I do to improve the next class?" So, what do I do? I download the Zoom file, right? A video file, transcribe that file, I take the comments, cross-reference the two, and see what was good, what I said, what might be untrue. I take everything I said and do research to see if there's any incorrect fact or not, right? I ask it to analyze this. And now my teaching format, I create a guide, a material. So I cross-reference all this information with this material, right? And then I used to do this manually. So I thought, "Damn, why am I doing this manually?" I went there and created a skill. So today, after each class, I call this ability, this skill. For those who don't know, a skill is just a file on the computer, a text file you create, right? I call this skill. This skill takes what I created, this PDF, this guide. It will take this written audio and also the comments. It will cross-reference this. It will tell me what needs to be done, what needs to be corrected or not. You don't know how much time this took me. Hours. Now this is resolved. This is just one of the things that can be done, right? Think about the traffic manager for campaign analysis. Instead of wasting one or two hours, man, they waste 5 minutes. That's if they are the ones doing it, because I think the big breakthrough now for AI in companies is to empower anyone on your team. It will no longer be restricted to your traffic manager. The strategist, when she wants, she goes and asks. When I want, I go and ask. >> Support. >> Understand? Support. Anyone. >> Exactly. So I think it's about giving this power that was previously restricted to a certain profession, right, like the traffic manager, the CS, the social media manager, to everyone, and everyone shares the same thing. Now, this has greatly facilitated things within my company, >> okay? But the absolute gain is the creation of skills, which is turning something, a process I used to do manually, >> repetitive >> into an ability for Cloud to use, >> which I can even share with someone else. >> You can share it with someone else. Exactly. So, from my point of view, this should be, I think, one of the only things people should focus on right now, learning to take something they do today and turn it into an ability for Cloud to use, because it will free up time. By freeing up time, you can think about how to earn more money, how to live your life more. Man, we joked that it's the best moment for the intelligent lazy, for the ADHD person, but it's also the apocalypse for the junior professional, right? So, the junior professional, man, like, >> eh, they will essentially be replaced with quality today. >> They are being, right? >> Right? And it's already reaching the intermediate level, right? [laughs] It's the junior professional apocalypse. >> Exactly. >> Perhaps the path for the junior professional is exactly to learn a lot about >> and I have concrete proof of this. >> Ah, good. So go on. >> Okay. No use talking. >> It has to be grounded. It has to be grounded. Proof. >> So let's go. There's a young man who is an affiliate of my wife, >> okay? He approached me about six months ago and said the following: "Uncle Thales, >> he's 16 years old." Good. >> Yes, the age. I agree. He's finishing high school. Finishing high school. >> Exactly. What am I going to do with my life, Uncle Thales? I wanted to learn automation. I saw it on YouTube, but I got confused. Man, what do you recommend? I saw this kid five years ago, >> right? Good guy. >> So I looked and said to him: "Man, forget that [ __ ] because AI is already doing that, okay? It's interesting to learn, if you want, no offense, but it's much healthier to learn something else that I'm going to tell you here, okay? You're going to learn to use Cloud Code." I went straight to Cloud Code. Why? For us older folks, right? 40 plus, a terminal might be more difficult, especially for those not in technology. Of course, for those not in technology, for those not in parentheses, it is. But for a 16-year-old kid, anything you present, it doesn't matter. They will learn. Their mind is bubbling, man. He took it, he started learning, and when he was having difficulty, I told him: "Figure it out, go on YouTube, ask Cloud." Sure enough, he sent me a message yesterday, after a few months, we keep talking, right, on WhatsApp, of course, he sent the first, the first proposal, >> hmm, >> to create a skill, a repetitive process that we might turn into an application for a company. He was so happy, man, because he had no prospects six months ago, and today his life has changed. He already knows that this might be what he'll do for the rest of his life, who knows, because he's there, he's excited. You, like the other one, don't know, right? It's exciting to use something that gives you the possibility to create things. Now imagine this for a young person, man. >> It's too empowering. >> It's empowering. But this in the hands of a young person is >> infinite, >> it's exponential. >> Yes. >> Understand? So I think young people should go down this path. Yes, man, I agree 1 million percent. I think, man, that's it, like, forget, man, doing that career path we did and so on, it doesn't work anymore. You have to do something else, right? Another path, like >> it might even work better in another context. Law and eh, I don't know. >> Perfect. >> But not, maybe in the traditional context. Yes, one thing I think is cool, we were even talking about it before here, right, in our chat before, like, eh, man, let's demystify something here, right, that like, >> for an entrepreneur who has never programmed in their life >> Uh-huh. >> More than that, who has never opened an LLM in their life, who, never opened one. Exactly. >> Exactly. >> Where do they start with Cloud Code? >> Okay. There's a fallacy, right? Limiting belief. >> Exactly. Limiting belief. Better. There's a limiting belief that to use AI, you have to know prompts, >> be a prompt engineer, >> a prompt engineer, a profession that started like this and in the same way it did this. Today, we don't even talk about it, >> right? I remember when I was listening, I read an article by someone who wrote that it doesn't make sense, because it will get there and it will do it. And I remember a study they did where they took laypeople and gave them an ultra-detailed prompt, and laypeople wrote anything, as if they were very, very laypeople. The best result was from the laypeople, because the AI understood better, because the others, by filling it with so much information, you limit the AI, >> let's say. Now, back to your question about the limiting belief, everyone thinks they need to be a prompt engineer or a programmer or be a genius. You don't need any of that. The only thing you need to do well is communicate clearly. In other words, you need good context. Folks, the secret of AI is just one. It's good context, in the end. It's not some secret tool hidden from normal people. No. When you provide good context with clarity, with facts, and say everything you want, the AI itself does the job well, and that's it, right? I think there's a lot of this fear, and I think our role is precisely to break this fear, right? Like, open the thing and start. Like, open the thing and start. I speak from personal experience. I was very, eh, eh, for a long time I was intimidated, eh, by it, and I'm a guy, [ __ ] in technology, okay? I'm from communication, but I'm a guy who has worked in technology for 10 years. Eh, and then, [ __ ] I was using my little ChatGPT and so on. And then I saw the crowd on Cloud, and I said, "Damn, man, I was really intimidated, because I said, 'Man, I don't know.'" Then, man, one day I sat down, started using it. Lucas Cudeler, I owe this to him, eh, sat next to me, started saying, "Open that damn thing, open it >> Now, now, write what you want to do." And I said, "Damn, is that it?" It's like, I go, you ask, it answers you, and you build things. So, you don't need, man, to be, eh, you can be an idiot, man. It's idiot-proof. If you are capable of, right, asking good questions, you'll get there, >> man. Eh, we're talking today about, a lot of people are saying that the future of the digital market, of digital products, is AI and so on. And I think that's nonsense, because AI is already the present. Like, man, if you don't have AI in your product, you're already a bit behind, right? So, like, you should already be there. >> Eh, where do you think this market of ours is going? Because at the same time it's scary, because like, the changes are very rapid, it's enchanting because it's like you start to have infinite potential, right? So we have a lot of difficulty grounding ourselves and thinking about where it's going. I know you think about this. >> I think and use it, exactly. So tell us where you think it's going. It would be too arrogant to try to predict that, but I'll bring a case I'm working on today, which might be a path. I've always been a person who bought a lot of courses and watched few >> few >> like the vast majority, perhaps >> vast majority, >> right? For lack of time. Anyway, we always have, >> you buy, I'll start. >> Yes, already with [ __ ] what a cool course. I'll do it on the weekend, never do it, okay? I'm that person, I won't lie. And then I stopped buying courses a bit because of that, knowing how I work. Exactly. [clears throat] >> And then I thought, man, there has to be a way, right, to solve this. What if I put AI to watch for me? That was the first thought, okay? But no, that's not it. Okay? Today I'm talking a lot about skills. What if I took the course, extracted the methodology, and took not the transcription, but how the person thinks, why they created it? What was the turning point? >> What was the turning point? >> What did they think for that course to be created, because if they created a course, it's because they have a different way of thinking than the majority that isn't written down somewhere. So, I created a skill that does exactly that. And when I did this and it worked, I thought, it's time to start buying courses again. >> Yes, [laughs] it's perfect. >> Because it makes total sense, because now I don't need to learn something anymore. Obviously, >> ah, cooking, no, that will continue, right? I'm talking more about things like, >> ah, computers, let's say. So I started buying courses again. So I'm not a guy who produces a lot of content. I bought a content course, threw it into Cloud, turned it into an ability so I don't have to study 40 hours about it. I don't want to learn about it. >> So wait. So, the ability validates your idea or helps you build based on the methodology of the >> ability, I extract the ability from the person to use that ability. It's like having the mentor by my side doing it for me, not teaching me, understand? >> Today, what the agent does traditionally, you go, you ask, it says, "Do this, this, and that." No, I want it to do it for me. So by turning a course into a skill, I can achieve this result. When I did this and validated it and it worked, man, I went back to consuming courses, because it made sense to me. I don't want to waste time, but I know there are many good people out there in certain niches that are very worthwhile. They are better than me at things I'm not good at. >> Perfect. >> I'm good at staying in the cave creating. I love staying in the cave creating things. So I won't have time to do other things. This is wonderful. So, going back to your question, I think the future of courses will literally be a skill. Imagine you sell course X. I think it would be very smart if you did an upsell of the skill for it. >> Skill, man. >> Got it? Why? >> That's good, huh? People don't want to waste time anymore, but if you sell something that will execute what you're teaching them, it makes much more sense, and you also help people use it. >> That's [ __ ] huh? >> Before I taught you, now I do it. >> Now I do it with you or for you. You >> It's exactly that. A skill has a minimal difference, right? The skill does it with you. Sometimes it stops, asks for your opinion. When you create an agent, it's autonomous, so it does it for you. >> Yes. >> Let's say. >> Man, with this crowd here, I can go back to the year 2000, 2001 to 2001. >> You can. >> When The Matrix was released and Neo learns from Fu. >> It's funny that this is exactly the reference, right? >> I pass, I pass this reference in my workshop. Pass, pass. [laughs] >> It's the reference. If you could, exactly, exactly that. >> That's it. I know with fu. Yes, yes, it's kind of like that. You're saying, it's like, >> it is. >> I know >> paid traffic, right? [laughs] That's it, man. Because I never thought it would be like this, right? But this is what's happening, man. Like, >> eh, at least intellectual things, right? Because obviously, because in this case it's a business that goes to the physical part, right? But, but from an intellectual point of view, this is where we're arriving, which is like, man, I acquire, >> but I think people are still worried about, like, [ __ ] if I deliver this skill, >> tomorrow, I'm delivering gold, my gold. >> Generally, those who think like that perhaps lack repertoire or the ability to create more. And then there's a bit of this belief, >> as if I had only one shot among the >> Yes, that's exactly it, because the big breakthrough is, man, okay, sell this, but sell a subscription to it where the person can have updates, because, for example, I create several skills, right? >> I'm always updating them, >> until you have to update, >> until because it's my daily life, things change, so it makes more sense from this point of view too, to sell a skill and also the option for the person to use a subscription. And that's okay, folks. You'll use it, and in a year it will be completely different, there will be a better version. It's a new concept, right? A paradigm shift. Then >> Yes, man, eh, I wanted to try to help the person who is watching us and is convinced, okay? I'm going to start using it. Good. What are the first two things they should delegate >> to go, like, >> Yes, in general. You can be creative, but like, the first things you think are worth delegating, to save time? >> I don't know if I'll have a simple answer for you, right? Because >> it's harder than it looks. >> It's harder than it looks. Each person will have a different thing. But in general, generally, right? Eh, content, generally everyone has this problem of stopping, generating content. It's boring. >> Everyone has it. I have it, you have it, right? Everyone has this problem. So I think the first should be that, but it's not what I think. >> Cool, >> okay? It's not what I think. I think ah, what you should do is map out what takes you a long time in your day-to-day, because that's where you'll get the real gain from things. >> Because, okay, creating content, maybe it's not even taking up one or two hours of your daily hours. What is it taking up? What box is taking up your hours in your day? That's what you should focus on, put your brain more on solving it. >> Exactly. Exactly. Because with content, you go there, maybe you won't be very excited, it will take time, it won't become a product right away. Man, it's totally different to take a process that you already execute, that already exists, and see its gain right away. Man, you know what I think you just gave an insight, which is the following: >> maybe this answer isn't in the things they'll do at work, but in personal life. >> Exactly. >> Because if you map out sometimes in your personal life, there's something in your personal life that takes you two and a half hours, and maybe that's what you'll automate, not from your company. >> What we call the "O ele" effect, right? >> Yes, [ __ ] >> I don't do this anymore, that I used to do for two hours a day. >> So, let's go. I think the biggest gain from AI, yes, it's in the professional context, but I strongly believe it's in the personal context, and that's what I've been working on a lot in the personal context in recent months. I'll give some examples for you. I'm an extremely disorganized and chaotic person, [laughs] very much so. I forget things. I remember that I came to Hotmart last time, I didn't save the PDF, I called this guy at midnight, "What's the address, >> man? What's the voucher for the flight? >> For the flight? >> What's the flight ticket number, because I lost it, man? Where is it? >> Because the person had sent it on WhatsApp with that rule that the message disappears. >> I'm that person, right? And I thought, "Okay, I'm using this for work, but I think before using it for work, my personal life should be better." I should use it better in my personal context, because otherwise I'll bring this mess from my personal life to my professional life. That's usually what happens. Perfect, right? If you're chaos in personal financial management, you're usually chaos, unless you have a controller, a controller, etc. So, what did I do? I created a process for hiring AI agents in the simplest way possible, as if it were a traditional company. So, what did I think? I worked in companies for many years. My wife worked as a psychologist and she worked in recruitment and selection. Uh-huh. >> Okay. I kept talking to her, Dani, what do we do to hire someone? Ah, the manager ah, sends a message here with the skills they need, the skills, right? The skills. >> I go to the market, do research, bring back five resumes here, he will interview, evaluate, go through the hiring process, and if everything goes well, he hires. >> Hires. >> Then I, hmm, that's it. What did I do? I created a company, and the only and first employee is not the CEO, it's HR. You start this company with HR because they will do the hiring of others, understand? So imagine I'm going to hire a traffic manager here, right? Something simple there. I call HR, I want to hire a traffic manager, they will analyze my entire company context first, because they need to have context about the company, otherwise they will hire something generic. "Okay, you are ah, I need to know who you are, what is your company. It's Talis, who has an automation school, has product A, B, C, and D, has this funnel, runs launches here, runs this, runs that, because otherwise they might hire a traffic manager for perpetual instead of launches, let's say." Perfect. >> Right? >> Okay? After this context is included for HR, they start doing research, and they go, like, to LinkedIn. I understand. >> They go on the internet and ask, "What are the skills of a traffic manager in 2026?" and they bring this back. So they come back knowing, "Hmm, they need to know this, this, this. They need to know that Meta updated to Andromeda or something, otherwise they'll be outdated." So they come back with this scope, and after that, they look for skills in the market, because there are various skill directories. You don't need to create a skill from scratch, an ability from scratch. There are some sites where you search for a skill, download it, execute it, test it, improve it, and use it. >> Perfect. >> So this process also looks for a skill. After all this is done, they start as a trainee to see if it's in line, if they will do their job properly. After a while,
I can promote him to be hired. The big breakthrough, however, is that after I hire this guy, how do you update yourself? Do you update yourselves? Do you take courses? Something like that? It's courses, so I'll study alone on YouTube, I'll buy the book, I'll go... Exactly. Exactly. So imagine, it's master's, it's harder than... Done. So imagine that I can, I can buy, for example, Sobral's product there, I can download, it's mine, I download the transcript and train AI with Sobral's product. In other words, I'm actually treating AI as a human employee. That's the big breakthrough for me. Don't invent trends. What has worked for 300 years in the corporate market? I'll take everything from the corporate market and translate it to AI. Man, it's working in my context. So, I hired a traffic manager, I hired a strategist, I hired a slide creator for me, okay? And all of this is working beautifully in this hiring process. However, the big turning point didn't come from there, where was it? The big turning point came from the personal context. That's when one day I thought, man, my personal life is a mess, it can't go on. Let me take this that I created for the company. Let me see what it would be like for me, Thales as an entity. I, the company. I, the company. Then I started there and then I thought the following, let's go. I need to divide my personal context into domains. So Thales, he has several domains. He has the domain, let's say, health, financial. Financial. Ah, children. Perfect. Marriage. Home. And so on. Let's take those. And for each one, I was giving context. So, for example, for the home context, I put all the home bills, right? Everything, everything. But I went further. I didn't just give the home bill, because every first of the month I lost time having to go to the website to download the slip. [ __ ] What do you mean lose time, man, doing these stupid things. I get very pissed off about this, okay? So I taught this agent to download everything. Everything that had to be downloaded, it was about seven sites, even from the government, DARF, I don't know what. In no-code, this? Yes. Okay. No, in no-code. No-code. Exactly. I taught him to download. So he had skills for everything, okay? And then after that, he would go and add it to the calendar. Exactly. So as not to forget. So if you look at my calendar today, it's crazy, man. Let me see if I can show you here. Quick game. Look at this. All of this here are reminders so I don't forget anything anymore. Damn. And it's impossible now to forget things, because before I forgot because I didn't have control of everything. So this is a personal context, and I kept forgetting to pay the bill. They've tried to cut off the electricity three times at my house because I forgot the light downstairs. Oh, they're coming to cut it off. Why? Because you didn't pay. Pay it now, but, you understand? Perfect. Okay. And then I went to the health domain, I took 10 years of Sérgio Franco, right, of exams that I do, I always do all those exams, right, a battery of them, and I put it there, I took genetic exams, I took everything I had and put it there, man. Then AI started to triangulate everything and started to find things I didn't even know. I started scheduling doctor's appointments, man, for things. It's surreal. Because sometimes, we can't see the patterns in things, and AI does it very well. Then I went to the children's part, right? I kept scheduling school events, for example, on the day there was a school event. [ __ ] Horrible. And now, no, it's already there, it's already blocked on my calendar for the whole year. Oh, this Saturday there can't be an event because it's family day at school. It's family day at school. The only time I'll be there, I can't, I have to pay attention. And all of this was working very well, and I thought, can I go further? So imagine that I moved recently, okay? I bought an apartment, and there were several problems. I'll take any example here. In an apartment, there was mold in the bathroom, and I hate having to research suppliers, send messages. Man, you waste time. What did I do? I have a skill, for me, the best skill there is, it's a simple skill, but it's a research skill. It researches in depth. So I gave the context, look, I have this problem here, here's this photo, research the best supplier in the South Zone of Rio de Janeiro and send a message. Man, it went there, researched all of them, looked at Google reviews, analyzed, pá, p pa pa pá pá, and in the end, I had already connected with WhatsApp, right? And it put the following, oh. I found this one, can I send this message? Yes. It went there, sent it, and [clears throat] then the guy scheduled it, the guy went there and resolved it. So what used to take me a lot of time in the personal sphere started taking less time and less time. This is one of the things. I'll give another example now. We delivered another apartment, okay? And then we, the company made the technical report, the technical report took 5 hours to do, and then they sent us a message saying that the new tenant, when he pulled off a sticker, it was moldy and that we would have to pay. I had a security deposit there, right, to receive. No, we're going to have to use this money. Man, I was very pissed off because this guy from this company was already acting in bad faith, right? Anyway, what did I do? I went to the coworking space. So, in the coworking space, for those who don't know and maybe it answers a question from the beginning, the coworking space has 12 plugins from various areas, for accounting, for law, for productivity, for everything. You just have to go there and use them. It's wonderful. So, what did I do? Remember that context is the soul of the business? [snoring] I went to the coworking space and said: "Look, this is happening. I took the entire WhatsApp conversation with the real estate agency, it analyzed all the emails, it analyzed all the photos of the initial report, it went online to research jurisprudence in Rio de Janeiro and put together a 10-page document. And then I took this legal plugin, right, for law, it reviewed everything, adjusted it, I sent it to the company, the next day they had signed the contract. [laughter] So [snoring] you see that it's not just for the professional sphere, and I think there's a lot of gain for the personal context, man, a lot. But nobody is talking about it. It's just the sales funnel, man. No, man, lead generation or whatever. Danger, man. I think first [snoring] we should use AI for the personal context. I even compare this to the part of maybe creating applications. People spend their whole lives, they need an idea. No, man, take a problem you have, create something for yourself, if it solves your problem, then you go and create it for others later. Boom, I think we can start there, Thales. I, I would make 1 hour and 40 minutes of episodes, but you know that I didn't even count the time, but I suppose we have already [laughter] beautifully exceeded the time. So I'm going to ask people for two things. First, follow Thales, because Thales is talking about this all the time, he's been doing it for a long time. So he's one of the guys who is ahead of this, he's a great partner of Hotmart, he even helps us to form our base. I think he thinks well about the future, and more importantly, he has his heart in the right place. He's a guy who has always been very correct, always very reliable, and that's as important to us as being competent, especially when we talk about a superpower like AI, right? So if you're not serious, man, this can be incredibly damaging. So, I'm going to invite you to follow Thales, to comment on this episode to be able to watch his immersion. And we're still going to choose the best one to offer even more to Thales' community. And what do I need from you too? That you like this episode, that you follow this channel, because I've already noticed that when I don't say this, you don't follow. So follow the channel, activate the bell. I know, man, they always say this, but it's because otherwise you don't do it. So go ahead, follow us, activate the bell, and come back every week, because the content is good, it's difficult, it's the damn notification. So I'll see you in the next episode. A kiss. Bye. Bye. Ciao.