From the course: Introduction to Large Language Models (LLMs) and Prompt Engineering by Pearson
Join today to access over 25,200 courses taught by industry experts.
Expanding into AI agents
From the course: Introduction to Large Language Models (LLMs) and Prompt Engineering by Pearson
Expanding into AI agents
A natural extension of retrieval-augmented generation is AI agents, where instead of simply retrieving information to keep the AI current, we actually want our AI to quote unquote do something. And as we'll see, a lot of this boils down to a similar construction of RAG, meaning a good prompt. So when we talk about agents, there's a lot of definitions floating around. So I'm gonna make things quite crystal clear. An agent is simply some AI, usually generative, some AI with an access to tools in which the AI can perform actions with those tools, observe the result and use it in context. So for example, if I were to ask an agent, who is that person? The AI would have to go, well, hold on, I have a thought. Let me look this up. Let me perform the action by looking that thing up. Let me observe the result of that looking something up and then respond to the user using that observation. This thought action observation response pattern is not always what you need to do to make an agent work,…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.