AI Agents Certification Course: Intro

Key Takeaways

Hi everyone and welcome to the RiseAI agent Mastery course. In this course, we're going to break down what it actually takes to build real working AI agents. Um you'll learn the core ideas, the tools, and the hands-on skill that you need to go from just taking a simple prompt and turning it into a full agent system. So, think of this as like the intro model as the foundation. I'll give you kind of a sense of who this course is for, what we'll cover, and how all the pieces fit together. My name is Shri. I'll be walking you through um the rest of the course. So you'll see me in all of these videos. So let's just dive right in. First and foremost, who is this for? So we're kind of designing this course for anyone who wants to build agents. So whether you're starting from zero or improving the agents that you already have, um if you're trying to understand things like rag or tool calling or MCP or eval anything like that, you're in the right place. And if you want like an end to-end example that you can actually run after the workshop, um we'll walk through exactly that. So [snorts] whether you're technical or more on the product side of things, um this course should still give you the concepts that you might need to build and ship agents that work. So your team might look like so most teams building agents include both technical users as well as kind of like domain experts. On the technical side, you might have engineers, developers, or data scientists. um they focus on the code, the automation, the pipelines, and kind of just making sure that everything runs fast and at a reasonable cost. So on the domain domain side of things, you might have product managers or subject matter experts. Um they're the ones that are kind of shaping the prompts. They're run running evals and they're making sure that the agent actually solves the user's problem. Now, the most important part is that both groups work together. Um, building good agents isn't only about the model, but it's really more about combining these technical skills um with domain knowledge. I keep saying the word agents a lot, agents this, agents that, but let's go ahead and start by defining what we mean by agents. So, agents are um software systems that can take actions on behalf of a user. So, they use reasoning powered by LMS to kind of figure out how to do that. I would say that there's three main pieces. So the first part of it is reasoning. It's basically the model's ability to kind of understand the request. Second is routing. Deciding which tools or actions make sense. And then the third is the actual action itself. So that means that calling the actual API or running code or even um invoking a model to kind of produce a response. So when you put these together, you get a system that can do more than just chat. It can actually take meaningful steps towards a specific task. Before we dive in, um I would say that here is a road map for the entire course. Um we'll start with the basics of an agent and then kind of move on to more engineering patterns and different agent architectures. Uh we'll cover tools MCP and then we'll look at rag and agentic rag. We'll send we'll spend some time on evaluation which is really the key uh to making sure that your agents are reliable. And finally, we'll talk through what happens after an agent is deployed. So, including things like monitoring as well as just production best practices. Regardless, by the end, you should kind of have a full picture of how to design, build, test, and ship real agents. I'll see you guys in the next one.