Introduction to Agents

Lecture Notes

# What is an AI Agent? Unlike traditional Large Language Models (LLMs) that function as static question-answering machines, an **AI Agent** is an autonomous entity capable of perceiving its environment, reasoning about complex problems, making independent decisions, and taking concrete actions using external tools to achieve a specific goal over a multi-step horizon. ## The Cognitive Architecture If an LLM is the "brain", an Agent provides the "hands" (tools), the "hippocampus" (memory), and the "prefrontal cortex" (planning). > **Definition:** An Autonomous Agent is an AI system that can act independently to achieve a given objective by breaking it down into manageable sub-tasks, executing them iteratively, and adapting its plan based on real-time feedback. ### Key Capabilities - **Perception:** Reading API outputs, web pages, or vision models to understand the current state. - **Decision Making:** Evaluating multiple potential actions and choosing the one that maximizes the probability of success. - **Action:** Interacting with the digital environment through APIs, DOM manipulation, or CLI commands.