Intermediate
AI Agents: Core Capabilities
Go deeper into the technical building blocks that make agents powerful: tool use and function calling, four types of agent memory, planning and reasoning strategies, retrieval-augmented generation (RAG), agentic design patterns, multi-agent coordination, and the MCP protocol.
01
Tools and Function Calling
Tool use is what moves AI agents from "reasoning about the world" to "acting in it." Learn exactly how function calling works under the hood, what makes a good tool, and what can go wrong.
›
02
Memory in Agents: Four Types
Agents need memory to build on their own work. Learn the four distinct types — in-context, episodic, semantic, and procedural — and how to choose the right one for each situation.
›
03
Planning and Reasoning Patterns
How an agent reasons before acting determines how reliable it is. This lesson covers the four main reasoning patterns — chain-of-thought, ReAct, Tree of Thoughts, and reflection — and when to use each.
›
04
RAG for Agents
Retrieval-Augmented Generation gives agents access to knowledge beyond their training data. Learn how RAG works, why agents need it, and what agentic RAG adds on top of the standard setup.
›
05
Agentic Design Patterns
Agentic design patterns are reusable architectural solutions to common agent problems. This lesson covers the four patterns Andrew Ng named in 2024 and how Anthropic extended them for production systems.
›
06
Multi-Agent Systems
Some tasks are too large, too complex, or too diverse for a single agent. Multi-agent systems distribute the work. Learn the coordination patterns used in production and what makes them fail.
›
07
MCP and Agent Protocols
Model Context Protocol (MCP) is an open standard that lets AI agents connect to any tool or data source through one consistent interface. Learn how it works and why it matters.
›
Prefer to learn offline?
Download the complete workbook — yours to keep, print, or use in teaching.
Download Workbook →
Online lessons always free · PDF includes personal & educational use license
Enjoying this track?
Get weekly AI learning picks — prompts, tools, and tutorials every Monday.