A software entity that uses AI techniques to autonomously perform tasks or make decisions on behalf of a user
A combination of all tools needed and the LLM that reasons and gets the task done (which the LLM can’t do on its own)
Agent is designed to interact with their environment, process input, and provide outputs based on the data they receive and the goals they’re supposed to achieve
Key characteristics
Autonomy
No human intervention by nature
But we can also add human input as well
Learning and adaptation
Learn from experiences and adapt their behavior over time
Usualy achieved through ML techniques
Interactions
AI agents should be able to interact with their environment (other software systems, databases, human users, etc)
Goal driven
Typically designed with specific goals/tasks in mind
Examples
Example 1
Example 2: Online travel agency
Book flights and get recommendations
Agent gives our LLM a brain and reasoning capabilities to assemble the correct tools to get the needed information to give to user
Agent gives our LLM a brain and reasoning capabilities to assemble the correct tools to get the needed information to give to user