← All courses
AI
AGENTIC AI
Agentic AI refers to AI systems that can independently make decisions, plan tasks, take actions, and adapt to achieve a goal with minimal human intervention. Unlike traditional AI that only responds to prompts, Agentic AI can: Analyze goals Break tasks into steps Use tools or APIs Learn from feedback Execute actions autonomously
4.6 (1062)10,262 learners21 lessons1h 24m
Curriculum
Topic
- Why an AI Agent Is More Than a Chatbot4:15
- LLM Brain vs Agent Body3:57
- AI Autonomy Levels: Copilot to Fully Autonomous Agent4:14
- The AI Agent Loop: Observe, Think, Act, Reflect4:27
- When to Use AI Agents vs When They’re Overkill3:38
- ReAct: The Reason-and-Act Framework Behind Modern AI Agents3:53
- Tool Calling: How an LLM Chooses the Right Action4:14
- Function Schemas: Teaching AI Models What Tools They Can Use4:01
- AI Planning: Breaking Big Goals Into Actionable Steps4:00
- AI Reflection: Self-Checking Before Responding3:53
- Chain-of-Thought vs Scratchpad Reasoning in AI Agents3:47
- Context Window: An AI Agent’s Working Memory & the Lost-in-the-Middle Problem3:40
- Short-Term vs Long-Term Memory in AI Agents3:54
- RAG: Giving AI Agents Knowledge Without Retraining the Model3:45
- Agentic RAG: AI Agents That Decide What to Retrieve and When3:33
- Agent State: The Memory Layer Behind Production-Ready AI Agents3:46
- MCP: The Universal Connector for AI Tools and Agents4:04
- Why MCP at the Linux Foundation Reshaped AI Frameworks in 20263:30
- Structured Outputs: When AI Must Return JSON Instead of Text3:49
- AI Sandboxes: Safe Environments for Agents to Run Code and Tasks4:40
- Computer-Use Agents: AI That Operates Software Like a Human4:47