Lesson 1. The AI Landscape
1.1 From Rule-Based to Learning-Based Systems
Case Study: Elevator Access Control
The Rule-Based Approach (Card Access System)
The Learning-Based Approach (Face Recognition System)
Understanding the Architectural Difference
Exercise 1
1.2 The Five Waves of AI
Wave 1: The Birth of AI and Symbolic Reasoning (1950s-1970s)
Wave 2: Expert Systems and Knowledge Engineering (1970s-1980s)
Wave 3: Machine Learning and Statistical Methods (1990s-2000s)
Wave 4: The Deep Learning Revolution (2010s)
Wave 5: Foundation Models and Generative AI (2020s-Present)
Exercise 2
1.3 Understanding AI Terminology: A Three-Layer Framework
1.4 Layer 1: Objectives, Discriminative vs. Generative
1.5 Layer 2: Training Methods, How AI Systems Learn
Supervised Learning: Learning from Labeled Examples
Unsupervised Learning: Discovering Hidden Patterns
Reinforcement Learning: Learning Through Interaction
Hybrid Approaches
1.6 Layer 3: Architecture, From ML Models to Deep Learning
The Feature Engineering Paradigm
Deep Learning’s Representational Revolution
Theoretical Foundations: The Power of Depth
1.7 From Foundation to Agency: Why This Landscape Matters
Quiz 1
3 attempts allowed
Lesson 2. What Is a Language Model?
2.1 The Language Modeling Problem
Why Natural Language Is So Hard
2.2 The Starting Point: Bag-of-Words
2.3 The Four Fundamental Problems of Bag-of-Words
2.4 Solving Problem #1: Equal Weighting (TF-IDF)
2.5 Solving Problem #2: Word Order and Syntax
N-Grams: Local Sequential Patterns
RNNs: Processing Text as a Sequence
LSTMs: Solving the Vanishing Gradient Problem
Exercise 3
2.6 Solving Problems #3 & #4: Representation (Embeddings)
From Sparse to Dense Representations
Word2Vec: Learning Embeddings from Context
What Embeddings Solve
Exercise 4
Quiz 2
3 attempts allowed
Lesson 3. Large Language Models
3.1 The Remaining Challenges
3.2 The Transformer: A Bird’s-Eye View
3.3 Self-Attention: The Core Mechanism
From Text to Vectors
The Intuition: Selective Focus
Step-by-Step Mechanics
Exercise 5
Multiple Perspectives and Masking
3.4 Architectural Variants
3.5 Scaling Laws: The Physics of Intelligence
3.6 Training a Functional LLM
Phase 1: Pre-Training
Phase 2: Post-Training (Alignment)
3.7 Text Generation: Decoding Strategies
Exercise 6
Quiz 3
3 attempts allowed
Lesson 4. Vision-Language Models
4.1 The Text Foundation We’re Extending
4.2 Three Fundamental Challenges
4.3 The VLM Skeleton
4.4 Component A: The Image Encoder
4.5 Component B: The Multimodal Projector
4.6 Component C: Fusion Strategies
4.7 Putting It Together: The VLM Recipe
4.8 Current Limitations
Exercise 7
Exercise 8
Quiz 4
3 attempts allowed
Lesson 5. Building with Large Language Models
5.1 The Love/Hate Relationship
5.2 Hallucination
5.3 Frozen Knowledge
5.4 The Reversal Curse
Exercise 9
5.5 Flat Latency
5.6 Misalignment
5.7 Underspecification
5.8 Visual Illusions and VLM Failures
5.9 From Limitations to Techniques
Exercise 10
Quiz 5
3 attempts allowed
Lesson 6. Prompt Engineering
6.1 Why Prompt Engineering?
6.2 Anatomy of a Prompt: The Six Elements
6.3 The Prompt Engineering Process
6.4 Core Prompting Techniques
Exercise 11
6.5 Technique Selection Guide
6.6 Key Principles for Prompt Design
Exercise 12
Quiz 6
3 attempts allowed
Lesson 7. Retrieval-Augmented Generation (RAG)
7.1 The Motivating Problem: When LLMs Don’t Know
7.2 The Significance of RAG
7.3 RAG at a Glance: The Three-Stage Architecture
7.4 Stage 1: Ingestion
7.5 Stage 2: Retrieval
Exercise 13
7.6 Stage 3: Synthesis
7.7 Evaluating RAG Systems
Exercise 14
Quiz 7
3 attempts allowed
Lesson 8. Fine-tuning and RLHF
8.1 The Limits of In-Context Learning
8.2 The Post-Training Landscape
8.3 Supervised Fine-Tuning (SFT)
8.4 Parameter-Efficient Fine-Tuning (PEFT)
Exercise 15
8.5 Why RLHF? The Limits of Imitation
8.6 The RLHF Workflow
8.7 Beyond PPO: The Baseline Problem
Exercise 16
Quiz 8
3 attempts allowed
Lesson 9. Agent Building Blocks
9.1 What Is an AI Agent?
Control Flow: Three Dimensions
The Agent Loop
9.2 The Six Levels of Agentic Autonomy
Exercise 17
Exercise 18
Quiz 9
3 attempts allowed
Lesson 10. The ReAct Pattern: Reasoning and Acting in Harmony
10.1 Anatomy of an Agent: Brain, Body, and Memory
The Brain: Language Model and Instructions
The Body: Tools
The Memory: Context and State
10.2 The ReAct Framework
10.3 Inside a ReAct Trace
10.4 Why Interleaving Matters
Exercise 19
10.5 Beyond ReAct
Exercise 20
Quiz 10
3 attempts allowed
Lesson 11. Context Engineering: Memory Systems
11.1 Memory as Context Curation
11.2 The Attention Budget and Context Rot
11.3 A Lightweight Taxonomy: Three Tiers of Memory
11.4 Just-in-Time Context vs. Upfront Loading
11.5 Long-Horizon Memory Strategies
11.6 Memory and the Three Agent Patterns
11.7 Open Challenges
Exercise 21
Exercise 22
Quiz 11
3 attempts allowed
Lesson 12. Tool Use and Function Calling
12.1 What Is a Tool?
12.2 The Function Calling Protocol
12.3 Designing Good Tools: The Agent-Computer Interface
12.4 Tool Execution Patterns
12.5 Error Handling and Risk
12.6 Evaluating Tool Use
12.7 MCP: Toward a Standard Tool Interface
Chapter Wrap-Up
Exercise 23
Exercise 24
Quiz 12
3 attempts allowed
Lesson 13. Multi-Agent Architectures and Design Patterns
13.1 Do You Actually Need Multi-Agent?
Workflows vs. Agents: The Production Reality
The Safe Path to Multi-Agent: The Agent-as-Mega-Tool
When to Consider a True Multi-Agent System
Exercise 25
Exercise 26
Quiz 13
3 attempts allowed
Lesson 14. Architectural Patterns: From Workflows to Swarms
14.1 The Network View
14.2 The Five Patterns
14.3 Choosing the Right Pattern
Exercise 27
Exercise 28
Quiz 14
3 attempts allowed
Lesson 15. Planning Systems: Goal Decomposition and Task Hierarchies
15.1 The Decomposition Problem
15.2 Task Graphs and Dependency Management
15.3 The Planner Agent
15.4 Decomposition Evaluation
Exercise 29
Exercise 30
Quiz 15
3 attempts allowed
Lesson 16. Communication, Coordination, and State Management
16.1 How Agents Talk to Each Other
16.2 The Agent-to-Agent Interface (AAI)
16.3 Context Boundaries and Information Hiding
16.4 Coordination Failures
16.5 Human-in-the-Loop in Multi-Agent Systems
Exercise 31
Exercise 32
Quiz 16
3 attempts allowed
Lesson 17. Production-Ready Agentic AI
17.1 The Foundation of Trust
The Demo-Deploy Gap
The Framework Stack: A Three-Layer Taxonomy
The Modular Agent: Versioning and Independence
Exercise 33
Quiz 17
3 attempts allowed
Lesson 18. Evaluation for Agentic Systems
18.1 The Attribution Problem
18.2 Evaluating at Three Levels
18.3 The Evaluation Framework
18.4 Testing Strategies
18.5 Reliability Patterns
Exercise 34
Exercise 35
Quiz 18
3 attempts allowed
Lesson 19. Observability and Cost Control
19.1 The Observability Stack
19.2 Key Metrics for Agentic Systems
19.3 Loop Divergence Detection
19.4 Cost, Latency, and the Efficiency Stack
Exercise 36
Quiz 19
3 attempts allowed
Lesson 20. Safety, Accountability, and Governance
20.1 The Threat Model
20.2 Prompt Injection
20.3 Tool Risk Auditing
20.4 Guardrails Architecture
20.5 Confirmation Gates and Human-in-the-Loop
20.6 Governance Frameworks
Exercise 37
Exercise 38
Quiz 20
3 attempts allowed
The Agentic AI Course
From Language Models to Multi-Agent Systems
It's never been easier to build an AI agent — and never been harder to make one that actually works. This course takes you from language-model foundations to production-ready multi-agent systems, with the depth to predict failure before it happens and take absolute architectural ownership.
Minimum price
$49.00
$99.00
You pay
Author earns
About
About the Course
The barrier to building autonomous AI systems has completely collapsed, but the chasm in true engineering understanding has never been deeper. Systems that impress in demos barely survive day one in production.
This course is the definitive, framework-agnostic guide for practitioners who want to move past fragile "prompt-and-pray" scripts and architect agentic systems that hold up under real production conditions.
Across twenty lessons, it builds from first principles to full multi-agent orchestration — not just what these systems do, but how they work architecturally and why they fail. You'll trace the complete arc from bag-of-words to self-attention, master RAG, LoRA, and the alignment frontier (DPO and GRPO), navigate the Six Levels of Agentic Autonomy, design three-tier memory systems that defeat context rot, orchestrate multi-agent systems with DAG-based task graphs without paying the complexity tax, and take absolute engineering ownership of production: evaluation frameworks, observability stacks, guardrail architectures, and kill-switch protocols designed before you need them.
Every lesson is reinforced with checkpoints to keep the concepts anchored: 20 quizzes and 38 exercises placed exactly where they consolidate what you've just learned, plus 4 bite-size FHD video lessons that use graphical demonstrations to make the hardest ideas click.
This course is for serious practitioners who aren't strangers to code or machine-learning fundamentals. If you're a software engineer, data scientist, or technical leader who wants to bridge the gap from "I can call an LLM" to "I can design, ship, and diagnose an autonomous system," it was written to equip you — not to impress you.
📖 This interactive course — twenty lessons with quizzes and exercises — is exclusive to Leanpub. For the full written treatment, see the book it's based on: The Agentic AI Book, also in print on Amazon.
Instructor
About the Instructor
Dr. Ryan Rad is an AI researcher, professor, and industry advisor with over 15 years shaping how artificial intelligence is built, taught, and deployed. He holds faculty positions at Northeastern University and the University of British Columbia, where he has designed curricula and graduate programs in Machine Learning, Computer Vision, Generative AI, and Data Science—reaching thousands of students across six universities and online. His industry work spans research, engineering, and leadership at startups and tech giants including Microsoft, scaling AI solutions across 40 countries for Fortune 500 companies. A sought-after speaker with 50+ international talks and tens of top-tier peer-reviewed publications, Dr. Rad's research focuses on resource-efficient generative AI for edge deployment. He is the author of The Agentic AI Book—distilling years of building, teaching, and deploying AI into actionable insights—and shares AI perspectives on his YouTube channel.
Material
Course Material
The Leanpub 60 Day 100% Happiness Guarantee
Within 60 days of purchase you can get a 100% refund on any Leanpub purchase, in two clicks.
See full terms...
Earn $8 on a $10 Purchase, and $16 on a $20 Purchase
We pay 80% royalties on purchases of $7.99 or more, and 80% royalties minus a 50 cent flat fee on purchases between $0.99 and $7.98. You earn $8 on a $10 sale, and $16 on a $20 sale. So, if we sell 5000 non-refunded copies of your book for $20, you'll earn $80,000.
(Yes, some authors have already earned much more than that on Leanpub.)
In fact, authors have earned over $15 million writing, publishing and selling on Leanpub.
Learn more about writing on Leanpub
Free Updates. DRM Free.
If you buy a Leanpub book, you get free updates for as long as the author updates the book! Many authors use Leanpub to publish their books in-progress, while they are writing them. All readers get free updates, regardless of when they bought the book or how much they paid (including free).
Most Leanpub books are available in PDF (for computers) and EPUB (for phones, tablets and Kindle). The formats that a book includes are shown at the top right corner of this page.
Finally, Leanpub books don't have any DRM copy-protection nonsense, so you can easily read them on any supported device.
Learn more about Leanpub's ebook formats and where to read them
Write and Publish on Leanpub
You can use Leanpub to easily write, publish and sell in-progress and completed ebooks and online courses!
Leanpub is a powerful platform for serious authors, combining a simple, elegant writing and publishing workflow with a store focused on selling in-progress ebooks.
Leanpub is a magical typewriter for authors: just write in plain text, and to publish your ebook, just click a button. (Or, if you are producing your ebook your own way, you can even upload your own PDF and/or EPUB files and then publish with one click!) It really is that easy.