AI Agents clearly explained (and why it matters)

AI agents are going to change the world in 2025. There is a very good chance this year AI agents will either get you or a lot of people you know fired from their jobs. AI agents are changing the world. By the end of this video, you'll know exactly what AI agents are. You'll have an entire army of AI agents working for you, and you'll even know how to build them yourself. This is the biggest trend in technology history, and it is critical you understand what they are. Let's lock in and get into it. So, I'm going to take you through a whole bunch of AI agent tools. I'm going to show you how to build them yourself. But first, it's important to understand what an AI agent is. There's a big difference between your chat GBT chat box and an AI agent that actually works for you. When you're in chat GPT, you're just asking questions, right? You ask the AI a question and then it gives you an answer. It's very one way. It's very back and forth. The AI is very dependent on you to give instruction. This is completely different when it comes to AI agents. AI agents work completely autonomously. They work by themselves. They're like an employee that works for you. They figure out what to work on. They figure out how to work on those things. They figure out what they need to complete those tasks and then they do it themselves. And on top of all that, and that this might be the most important part, they're able to reflect and improve and figure out what they can do better next time. It's literally like a human being. So first point work autonomously able to work by themselves. You give a task it goes it does a bunch of things for you right when you're in chat GPT now a normal AI chat box you ask a question it gives you an answer back and forth dependent on you AI agents it's not dependent on you they can go and figure things out to work on themselves. That's why a lot of people are scared AI agents will take their jobs. It's because they're very humanlike and able to work independently. Here's another big thing. They're able to break goals into tasks. Right? When I ask a question right now to chat GPT, I say, "What is the capital of France? " Right? It just looks in its knowledge base and it gives you an answer. But what if I were to say something like, "Plan a YouTube video for me based on trending topics right now. " Now, it gets a little bit more complex. The way an AI agent would figure this out is the first thing they do is they break it down to task. Okay, I need to figure out first of all what is trending right now in the world. Then I need to figure out okay, how do I turn those trends into scripts and then I need to figure out okay, how do I write those scripts? Right? It's multiple steps to complete that goal. So the AI agent goes and figures out every single task it needs to do to complete a goal. AI chatbots very singular. I answer a question. AI agents I break down complex goals into simple tasks. The next big important thing here is AI agents have access to tools. They're able to do a lot of different things. They don't just look in a knowledge base and answer a question. They have tools available to them to complete those complex goals and tasks. So things like web search, they're able to search the web to get answers. Codegen, they can write code for you. Computer use, they can take control of your computer and do things for you. AI agents have access to many different tools to complete tasks. And then lastly, this is a huge one here, reflection and improving on themselves. AI agents after they complete your goals by breaking it down into tasks, they're able to reflect, see how those tasks went, see if they were able to complete those goals, and then next time they do tasks, they're able to improve and do better because they saw how their last tasks went. So, they're self-conscious and able to improve based on what they've done. This is unbelievable. AI chat bots, when you chat with them, they don't reflect on themselves. Just to answer your question, you put all of these together and you have AI agents. And that's why there's a lot of hysteria right now around will people lose jobs over AI agents. This technology was just born. It has just started. It is cutting edge and it is just being developed. When this vision for AI agents comes to fruition, it is very possible a lot of people will lose their jobs. And that's why it's so critical you pay attention to this. By the way, if you've learned anything at all so far, make sure to leave a like, subscribe because all I do is make incredible videos about building stuff with AI. And let me know down in the replies what do you want to see a video on next. I'd love to educate you on anything that has to do with AI. Now

let's talk about popular AI agents out there that you can start using today. One of the most popular AI agents is Bolt. new. It's a developer agent. You give it a goal of what you want to build. It goes out, does research, and figures out how to build it for you. So, take a look at this prompt. I'm going to give this AI agent. I want to build a health tracking app. I can add in a food and the app tells me how many calories are in the food plus all the other nutritional information. Make it absolutely beautiful. So, now what this AI agent will do is it'll go research how health tracking apps work. It'll research how to figure out how many calories and protein are in food. It'll research what technologies are needed to build the app. And then it'll go and build the app for me. It'll break the goal down into tasks and then complete those tasks with tools. Let's hit enter. Okay. So now it's taking my prompt and it's going to start doing planning for this task. So it's thinking right now and this is a big thing. AI agents are able to think and reflect on the task to figure out the right ways to do things. Right? Let me analyze what I need to do based on the user's request health. See it's breaking it down into tasks. All right. Let me plan the key components. I'll need the main component the food search compon. Right? It's breaking the goal down into task. This is not how regular AI chatbots work. But this is how agents work. They're able to independently think about how to accomplish goals. Based on your request, I'll create a beautifully designed health tracking app that helps users track nutritional information for foods. It came up with all the features. So, the food search lookup and functionality, daily nutrition summary, personalized user dashboard, responsive design on all devices, clean modern interface with Apple inspired aesthetics. It came up with all this. It broke this down to all these different tasks. Now, the agent's using its code generation tools, right? We talked about tools earlier and all the tools available to the agent. It's now using the codegen tools to build all the code you see here. So, it's able to search the web to research how to build the app. Then, it uses its code tools to actually build it. AI agents in the future will have access to thousands of tools they can use to complete tasks, including controlling your computer. Boom. And here we go. Neutr, right? It came up with the app, the name, the logo, the dashboard, a breakdown of your daily nutrition. We can search for foods. It'll give a summary of all your macros and everything involved. It even has terms of service and a privacy policy. Right? The AI agent went out, figured out what we needed to build, broke it down into task, and built it for us. Let's take a look at another AI agent that's taken the world by storm right now. So, this is manis. im. This is probably the most powerful AI agent I've used so far. And this is a researching AI agent. So you give it a task to research and it goes out and it does a whole bunch of steps to give you really powerful research. Right? So I asked it what is the minimum Mac Studio specs I need to run DeepSeek R1 locally. This is something I'm thinking about doing running my own AI model locally with a Mac Studio. I want to know what's the cheapest Mac Studio I can buy that can run that model. So it goes in. I'll help you find information about the minimum Mac Studio specifications needed to run R1 locally. And what Manis does really cool that I've never seen from any other app is it breaks it down into a ton of tiny little steps it can do to complete this task. Right? So it searches for DeepSeek R1 system requirements and then it uses its webs web searching tool to search for it. Uh it looks for different websites. So it finds from different sources all the data. It found the detailed system requirements and it continues the research and you can see every website it browses to find that information. Right? And then it looks at Mac Studio specifications. And now it figured out, okay, now that we know how Deep Seek runs, let's figure out all the Mac Studio specifications out there and it searches Apple's site for all the specs of Mac Studio. You can even see it's scrolling down and looking at different parts of the site. And then it reports back to you. Okay, I've got all the details about DeepSeek. Mac Studio. Let me put this together for you. And then you scroll down here. It shows you all the information it discovered. So all the DeepSeek models, what Mac studio need you need to run each deepseeek model and at the very end it builds you an entire document that you can then download and use with all the information it found. And if I click on that document you can see the exact report it built for you in a really nice detailed structured document. This is what makes AI agent so powerful. You give it a simple goal, it goes works autonomously, works independently and completes a bunch of different tasks for you. It uses its different tools. It uses its web searching tool, its codegen tool, and even document creation tools like the one you see here to create this document. That's what makes AI agents so powerful. So, as I said earlier, I

really believe AI agents are going to cause a ton of destruction in the world. They're going to cause a lot of good things, but they're also destruction. What do I mean by that? A lot of people are going to lose jobs. AI agents are so capable beyond belief. There are a lot of jobs that are just going to be completely eliminated because AI agents can do that same task for basically free. So, how do you prepare yourself? How do you get ahead of this? How do you prevent yourself from being part of the destruction? Well, first you start using AI agents, right? You start using them yourself. You start using all the popular ones. You start building your own prototypes of boldnew. apps with wind surf and cursor. I like windsurf a little bit more right now. You start using manis to do research. You start using chat gpt03. 03 is chat gpt's newest model and it's actually amazing and it's very agentic in nature because it has access to a ton of different tools. It is the first AI model that actually has access to tools. So it is very close to being its own AI agent. I would definitely recommend trying that out. Here's my challenge to you. Every day, use an AI agent for at least 15 minutes. Build something with an agent. Send out manis to do a research task for you. Use 03 to brainstorm new ideas that you can build. Use it for 15 minutes every day and you will become a master of AI agents. But here's how you take it to the next

level. Here's how you not only prevent yourself from being part of the destruction of AI agents, but you actually enable yourself to capitalize on the great value creation that will come from AI agents as well. You build your own. You can build your own AI agent. In a world where now anyone can be a creator, you cannot be a consumer. You have to be a creator as well. Over the next two years, there are going to be multiple oneperson billion-dollar companies created off people building AI agents. Instead of being the ones feeding the money to those creators, be the ones creating that and actually getting the value yourself. The question becomes, how do you build your own AI agent? Well, I'm going to show you. First of all, you want to download an agent that helps you build apps. That's right. You're going to be downloading an agent that helps you build agents. I like Windsurf. It's my favorite AI agent right now. It helps me build out all the apps I build. I code every single day with it. I would check out Windinssurf windsurf. com. It's an agent that helps you write code and build apps. Download Windsurf. You'll be able to build your own code even if you've never programmed a day in your entire life. And then you need to be aware of the AI agent frameworks. Google just made their own Google agent developer kit. I tried a little bit myself. It's a little rough around the edges since it's so brand new. There's also Langchain. This is the one I'm currently use to build my AI agent right now. Lang Chain is a wellestablished open-source AI agent framework. Very easy to use. Every AI you use to build things, including in Windsurf, will know how to build it. So, you definitely want to be familiar with that. Next, once you have Windsurf downloaded, you can start building your own AI agent. So, open up Windsurf like you see here. Start a new project, open a new folder, and then in cascade, which is their AI agent builder, you can go in here and use this prompt. I want to build my own AI agent using lang chain. It should be able to take in a prompt to do research on the web based on that prompt. It should use Tavi for web searching. For those who don't know, Tavi uh is an API that actually enables AI agents to do web searches and crawl the web and find information. So, I use Tavi for all my AI agents. I'll put this prompt down below. Right, download Windsurf. Start building your AI agent. Take this prompt. Use it. It will start building your AI agent for you. In fact, here's an AI agent I built myself. I called it Chad. I want to give my AI agent a name. It's my Chad research assistant. And you can see all the research I've had this do for me, right? It's built all these documents for me. Is DeepSeek AI cheaper than OpenAI API? It went out and it did the research and built this document for me and did all this research. I just hit enter. It built this document. Now I can download it and I have all this research done for me. I have an AI agent. I built myself that works for me. I literally built my own employee. This is something everyone needs to do. If you do something like this, you will be a lot more safe at your job than you are now. Cuz I'm telling you, a lot of people are going to be in danger because of AI agents over the next year. And before we wrap up here, if you learned anything at all, make sure to hit the subscribe button, leave a like, and let me know in the replies what you want me to film next. All I do is make awesome videos about AI. So, as a recap here, AI agents about to change the world. They work autonomously. They break your goals down into tasks. They have access to many different tools, and they can reflect and improve on themselves. So, they're always getting better and always getting more productive. You need to be taking action now. Use the tools I showed you here. Take advantage of Bolt, Windserve, Manis, all them. Try them out. Get used to them. Understand how they work. And then here's my challenge to you. If you can do this next part, I promise you, you're ahead of the game. Build your own agent. It's possible. There's going to be billion-dollar AI agents made in the next three to four months. I'm working on one myself right now. You should be doing it, too. Try it out. Download Windsurf and test it out right now. Build something for yourself. I promise you, you'll not only be really safe at your job moving forward, but on top of that, you'll be a lot more satisfied. You wake up, you build stuff. It's really, really satisfying building stuff. I've been I'll just tell you this. I wake up so much more excited to get out of bed and do stuff after I started building. Building, I think, is the cure to all sadness. Honestly, give it a shot. Get out there, start building things. Enjoy it. AI agents are coming. Thank you for sticking around. Just by being here, you're ahead of 99% other people. Leave a like and subscribe. Help us see you in the next video.