FeedbackShare feedback

← All sections·Learning path

IV

Agentic AI18

Core

AI that plans, acts, and uses tools autonomously - moving beyond question-and-answer into doing.

Agent Memory

The mechanisms by which AI agents store and retrieve information across tasks, sessions, and time - enabling them to learn from experience and maintain context beyond a single conversation.

Agent Orchestration

The system that coordinates multiple AI agents - deciding which agent handles which task, managing their communication, and ensuring the overall workflow stays on track.

Agent State Management

The system that tracks everything an AI agent knows and has done so far during a task - the equivalent of working memory and a to-do list for a multi-step AI system.

All concepts

A

C

F

Fallback Strategy
The plan an AI agent follows when its primary approach fails - ensuring that failure has a graceful outcome rather than a crash or silent bad output.

H

Human-in-the-Loop
A design pattern in which a human is kept in the decision-making process for an AI system - reviewing outputs, approving actions, or providing corrections at key points rather than allowing full autonomy.

L

Long-Horizon Planning
The ability of an AI agent to decompose complex multi-step goals, sequence actions across many decisions, and maintain coherent progress toward an objective over an extended period.

M

P

Prompt Injection
An attack on AI agents in which malicious instructions hidden in external content - websites, documents, emails - hijack the agent to perform actions the user did not authorise.

R

S

Self-Healing
The ability of an AI agent to detect when something has gone wrong and automatically attempt to fix the problem - recovering from errors without human intervention.

T

Tool Calling / Function Calling
The mechanism that lets AI models request to run external tools - searching the web, executing code, querying databases - and incorporate the results into their responses.