---
title: "Part 1: Agentic AI Full Course | Foundations (LLMs, ReAct Loop, Agent Architecture) 2026"
source_url: https://youtu.be/f5j99Lfm-dU
captured_at: 2026-06-09
author: "SGITMYSTR "
contributor: "ชิดชนก"
platform: youtube
---
# Part 1: Agentic AI Full Course | Foundations (LLMs, ReAct Loop, Agent Architecture) 2026

youtube video by **SGITMYSTR **, 6m57s, uploaded 2026-06-08. URL: https://youtu.be/f5j99Lfm-dU

## Shared in LINE Note+
- 2026-06-09 07:35 by ชิดชนก

## Description
🤖 Welcome to Part 1 of the SGITMystr Agentic AI Series — your complete hands-on 
guide to understanding how AI Agents actually work under the hood.

Whether you're a developer, DevOps engineer, or IT professional — this is the 
foundation you need before building your first AI Agent.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📚 WHAT YOU'LL LEARN IN THIS VIDEO
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
✅ Module 1 — What is Agentic AI? (00:00)
   • AI Evolution: Chatbots → Copilots → Agents → Multi-Agents
   • Agentic AI vs Traditional AI — key differences
   • Real-world examples: Devin, AutoGen, Claude Code, Perplexity
   • Future of Autonomous AI Systems

✅ Module 2 — LLM Fundamentals (00:45)
   • How Large Language Models work (tokens, embeddings, context windows)
   • Prompt Engineering: Zero-Shot, Few-Shot, Chain-of-Thought, ReAct
   • Reasoning Models vs Standard Models (o1, Claude 3.7, GPT-4o)
   • LLM Hallucinations and how to fix them

✅ Module 3 — AI Agent Architecture (01:30)
   • Agent components: Perception, Memory, Planning, Tools, Reflection
   • Memory systems: In-context, Vector DB, Episodic, Procedural
   • The ReAct Loop: Thought → Action → Observation
   • Multi-step decision makin