AI
// FULL COURSE · 2026 EDITION · UPDATED
Roadmap to
Agentic AI
A complete self-paced course covering everything from ML fundamentals to production multi-agent systems — with theory, visuals, step-by-step labs, and a direct line to Claude for help.
16
SECTIONS
9
HANDS-ON LABS
7
QUIZZES
2026
UPDATED
// FOUNDATION
01
NOT STARTED
02
NOT STARTED
03
UPDATED 2026
04
NOT STARTED
Introduction to Agentic AI
What agents are, how they differ from regular AI, the core perceive→reason→act loop
AI & ML Fundamentals
Supervised learning, neural networks, transformers, gradient descent
Programming & AI Frameworks
LangChain, LangGraph, CrewAI, Anthropic SDK, Mastra — when to use which
Large Language Models
Transformer architecture, tokenization, model families, cost vs. capability trade-offs
// CORE AGENTS
05
NOT STARTED
06
NOT STARTED
07
NOT STARTED
08
NOT STARTED
09
NOT STARTED
10
NOT STARTED
Understanding AI Agents
ReAct, Reflection, Tool Use patterns — the design vocabulary every AI engineer needs
AI Memory & Knowledge Retrieval
4 memory types, vector databases, embeddings, semantic search, chunking
Decision-Making & Planning
Task decomposition, hierarchical planning, error handling, conditional branching
Prompt Engineering & Adaptation
Few-shot, chain-of-thought, system prompt design, prompt compression
Reinforcement Learning & Self-Improvement
RLHF, PPO, DPO, reward modeling, self-reflection patterns
Retrieval-Augmented Generation (RAG)
Naive RAG, advanced techniques, HyDE, re-ranking, RAGAS evaluation, GraphRAG
// PRODUCTION
11
UPDATED 2026
12
NOT STARTED
Deploying AI Agents
Docker, async execution, token cost management, observability, FinOps
Real-World AI Agent Applications
Business automation, enterprise integrations, human oversight workflows
// NEW IN 2026
13
NEW 2026
14
NEW 2026
15
NEW 2026
16
NEW 2026
MCP & Agent Communication Protocols
Model Context Protocol, A2A, building and consuming MCP servers
Multi-Agent Orchestration
Orchestrator-subagent patterns, specialist agents, parallel execution, debugging
Agent Security & Adversarial Robustness
Prompt injection, sandboxing, least privilege, supply chain attacks, audit logging
Long-Horizon Agents & Production Reliability
Checkpointing, graceful degradation, error propagation, token budget optimization