Curriculum Phases
20 phases from setup & tooling to capstone projects. Click any phase to explore its lessons.
Setup & Tooling
Get your environment ready for everything that follows.
12 of 12 lessons available
Math Foundations
The intuition behind every AI algorithm, through code.
22 of 22 lessons available
ML Fundamentals
Classical ML — still the backbone of most production AI.
18 of 18 lessons available
Deep Learning Core
Neural networks from first principles. No frameworks until you build one.
13 of 13 lessons available
Computer Vision
From pixels to understanding — image, video, 3D, VLMs, and world models.
28 of 28 lessons available
NLP: Foundations to Advanced
Language is the interface to intelligence.
29 of 29 lessons available
Speech & Audio
Hear, understand, speak.
17 of 17 lessons available
Transformers Deep Dive
The architecture that changed everything.
16 of 16 lessons available
Generative AI
Create images, video, audio, 3D, and more.
15 of 15 lessons available
Reinforcement Learning
The foundation of RLHF and game-playing AI.
12 of 12 lessons available
LLMs from Scratch
Build, train, and understand large language models.
24 of 24 lessons available
LLM Engineering
Put LLMs to work in production.
17 of 17 lessons available
Multimodal AI
See, hear, read, and reason across modalities — from ViT patches to computer-use agents.
25 of 25 lessons available
Tools & Protocols
The interfaces between AI and the real world.
23 of 23 lessons available
Agent Engineering
Build agents from first principles — loop, memory, planning, frameworks, benchmarks, production, workbench.
42 of 42 lessons available
Autonomous Systems
Long-horizon agents, self-improvement, and the 2026 safety stack.
22 of 22 lessons available
Multi-Agent & Swarms
Coordination, emergence, and collective intelligence.
25 of 25 lessons available
Infrastructure & Production
Ship AI to the real world.
28 of 28 lessons available
Ethics, Safety & Alignment
Build AI that helps humanity. Not optional.
30 of 30 lessons available
Capstone Projects
17 end-to-end products + 9 deep-build tracks. 20-40 hours per project; 4-12 lessons per track.
85 of 85 lessons available
How this curriculum works
Each lesson follows the same loop: read the problem, derive the math, write the code, run the test, keep the artifact. No five-minute videos, no copy-paste deploys. Content is sourced from the open-source ai-engineering-from-scratch project by Rohit Ghumare (MIT license).