OnDemands New AI Changes AI Agents Forever! (Ondemand AI Tutorial)

So if you've ever imagined building and automating your own AI agents without needing endless resources or technical roadblocks, you can actually meet On Demand, the ultimate AI operating system designed for tech enthusiasts, developers and businesses alike. Whether you're looking to automate tasks, or build custom AI workflows, or even test the latest AI agents, this is the platform that has got you covered. But here's some big news for my viewers. Okay, you guys get $150 in free credit, no credit card required to dive in and start exploring so you can build agents, create automations, test workflows at absolutely no cost. And crazily, starting in January, things get even better. You'll be able to monetize the AI agents you create by sharing them in a decentralized marketplace. And in this video, I'm going to take you guys through an in depth walk around of On Demand, exploring the Agent Marketplace where you'll find ready to use AI agents, everything from social media automation to to automated trading. And I'll also show you how to use the Agent Flow Builder to assemble your own workflows, schedule tasks and integrate outputs seamlessly. So let's not waste any more time and jump right in to see how this platform can help you automate AI. So one of the first things you're going to want to do when you log into the On Demand website is go to the Agent Marketplace. So the Agent Marketplace is one of the standout features. Think of it as a hub where you can discover, test and customize pre built AI agents to suit your needs. As you can see here, the marketplace is organized into things like data analysis. We've of course we've got, you know, things like finance. We've got every single thing you would want that is organized by topics that you might like. So there's pretty much something for everyone here. All you'd have to do is just simply scroll around, look around. For example, if you're a marketer, you'd be using these marketer agents. Or if you're a researcher, you'd likely be using these research agents. And you can see there are many different agents for every single category. And the best part of this AI agent marketplace is that you don't even need to create these agents yourself. You can simply copy them and use them in all in just a few clicks. So whether you're automating reports, sending notifications, or crunching financial data, the possibilities here are endless. Now a really cool thing about this platform is that you can actually create your own agents. I'm going to show you guys how to create two agents. So one of the first things I'm going to do here is I'm actually going to create my own agent. So for example, we've got three different types of agents. First we're going to go ahead and I'm going to show you guys how to actually create a knowledge base agent. This is pretty much a basic agent, but it's really, really useful. So if we click Knowledge based agent, then right here you're going to see that we can basically create our own agent that is trained on our own knowledge. So I'm actually going to call this agent the AI Investment agent. And then I'm going to put this into the category of finance. And then my agent description, I'm just going to name this an AI agent that helps you invest into AI. As always, you want to add a conversation starter. So I'm going to put what's the best stocks to invest in in for a good return on investing in AI. Then this is where you're going to upload your knowledge base. And of course when you click the conversation starter, you just want to click add and it will be added right there. You can also add it up to around five or so. You can add a lot, of course. I'm just going to go ahead and change the profile picture. I've just added a really old profile picture just for the sake of this tutorial. Then you can see I've now clicked upload and this file is uploading. So I can upload some PDFs here and it's completely up to you but what kind of files you do want to upload. So now I can save and this agent is going to be here right for me to be able to use. So I've got my AI investment agent. I can of course publish this to the marketplace if I wish to, or I can chat with this privately. So what you want to do is you want to just click Add. Then I'm going to go my AI investment agent. I'm going to just click Add. Now this is added to my session. So what I will do now is I will go ahead and I can use this prompt right here and then I can say what's the best document to get A blah blah blah. And then I can click Run. You can see right here it says based on the information provided, it's giving me the information that this agent was trained on a really useful way for you to use different agents for different workflows. So this is really nice and this is something that's really easy for you to use. So now let's talk about one of the most powerful features of On Demand. The ability to export your AI agents into different programming languages. Let me break down why this is incredibly useful. So all you want to do is you want to click here and you can basically pick any kind of programming language. Now this is really useful because this allows integration flexibility. If you're building a Python web application, you might want the Python code. If you're working On a Node JS background, you'll want the JavaScript code, which I can easily get here. Just enter JavaScript and click Get Code. For mobile apps, you might need Swift or Java, the code export feature basically lets you integrate your AI agent into any existing system. And looking at this JavaScript code example here, the code we're seeing is what you'd use to integrate this AI agent into a JavaScript application. And it provides two main functions. One to create chat sessions with your AI agent and another to actually send queries to that agent. And this actually matters quite a lot because instead of being locked into the on demand platform's interface, you, you can actually run your AI agent on your own servers, integrate it directly into your applications and customize how it works with your specific needs and scale it according to your requirements. Think of it like building with Lego blocks. You can create the AI agent here in the playground and then export it to whatever programming language fits your project. And this makes On Demand incredibly flexible for businesses and developers. What we can also do is we can also create a REST API agent. So for example, if we click here, then we click next, you can see we've got the same information. So for this we're just going to create a very simple basic agent, since it's going to be able to demonstrate exactly what's going on with no authentication required. So let's name this agent the HTTP CAT Agent. Let's just add another basic icon here. Then from the drop down menu, let's just put this into. I guess it's just put into general. Of course, this is a chat agent. Then for the description we've got this. This agent provides images of a cat that represents HTTP status codes. Want to see what a 404 error looks like as a cat? This agent has you covered for the starter conversations. Let's just add this. Show me a cat status for code error404. And of course, give me a cat image for 201 created. Let's just add that too. Now, we don't actually need any configuration fields for this. But of course, if you want to know more about the configuration fields, you can head on over to the documentation. And then this is pretty much everything you'll need to know. So you can just pause the video here or just, you know, check out the documentation. Now, the important part here, from the example, we can just grab this HTTP cat. Let's just actually remove this configuration field right there, because we actually don't need that. And then now we can actually go ahead and deploy this AI agent. So you see right there, that is the agent that has been successfully deployed. And that is of course how you can create REST API agents really easily. And then of course, if you come on over to the area, you add that REST API agent once again. Of course you create a new session and you use any one of these. I give me a cat image for 201 created. It actually gives you the image. And then when you actually download the image, you can see this is the image that you do get. So that is how you do that section. So let's actually explore the agent workflow builder. And this is basically like building AI. So let's explore the agent's workflow builder. And this is like building with Lego blocks, but for AI processes. So you can see right here, if I click Create Workflow, this is the first thing that we do see. First you can see the trigger section. It's set to API, which means that this flow will start when we make an API. First we see the trigger section. You can see see that this is set to API, which means that this flow will start when we make an API call. This is like setting up a doorbell. When someone rings, it sends a request. This is where our flow starts in the middle. This is where we have our LLM block. We have On Demand GPT4O selected, which is a powerful AI model. And then this is where we're going to need to fill in two important fields. For example, in the fulfillment prompt, we can input this, such as process the user's inputs and provide a helpful response. And if they ask a question, answer it. If they share a statement, provide relevant insights. And of course, for the prompts, this is where you'd actually put the user's prompt on the right. This is where you have the output blocks, and this is where your results will appear. And it's automatically connected to show what our AI processing returns. And this is a very basic but powerful setup. So when someone sends a message in this flow, the trigger receives it. Then of course, the LLM processes it. Using our instructions. Then the then of course the output returns the enhanced response. Of course you can go ahead and run this to test to see if it is working. What's also cool is they also have a bring your own model section. This is where you can actually deploy your own model. For example, you can search hugging face for your own models. You can also sign in with Hugging face and find your model that way. And this is going to be something that allows just a little bit more customization because there are so many different AI models that are currently recently released. So if you want to use your specific own models, this is going to be the area for you. You can see there's tons and tons of different models there. So you're not just bound to what you are using with GPT4, Gemini, Flash and those standard models. So if you do enjoy the on demand platform, don't forget to get started and enjoy your $150 free credit grant to get started with no credit card required. If you enjoyed this video, don't forget to leave a comment below. Let me know what you're building with this platform and I'd love to see you guys in the next video.