Curriculum

24 lessons across 4 phases. Each lesson builds on the last. Phase 1 is completely free — no card required.

Phase 1: Foundations

FREE

Understand the building blocks — neural networks, transformers, attention, and how LLMs work.

1
What is a Neural Network?
The fundamental building block of all modern AI
Free
2
How Transformers Work
The architecture behind every modern LLM
Free
3
The Attention Mechanism
The mathematical heart of every transformer
Free
4
How LLMs Actually Work
The complete picture — from text to intelligence
Free
5
Why LLMs Are the Future of AI
The thesis: LLMs as the nucleus of artificial intelligence
Free

Phase 2: The Build

PRO

Build an LLM from scratch — tokenization, embeddings, training, and architecture.

6
Tokenization Deep Dive
BPE, WordPiece, and SentencePiece — how text becomes numbers
7
Word Embeddings & Vector Spaces
How meaning becomes math
8
Build Your First Neural Network
From zero to a working network in Python
9
Build a Tiny Language Model
Your first LLM — 1M to 10M parameters
10
Training: Gradient Descent & Optimization
Adam, learning rate schedules, and training stability
11
Build GPT from Scratch
The complete transformer — coded line by line
12
Scaling Up: GPT-2 Architecture
From toy model to 124M parameters
13
Evaluation & Benchmarks
MMLU, HumanEval, and measuring what your model can do

Phase 3: Make It Useful

PRO

Fine-tune, add retrieval, run locally, build agents — turn your LLM into a product.

14
Fine-Tuning with LoRA/QLoRA
Adapt any model to your domain — efficiently
15
RAG: Retrieval Augmented Generation
Ground your LLM in real, up-to-date knowledge
16
Running LLMs Locally
Ollama, llama.cpp, and local inference
17
Open-Source Models: Llama, Mistral, Qwen
The landscape of models you can use and modify
18
Building AI Agents on LLMs
From chat to autonomous action
19
Prompt Engineering Mastery
The science of communicating with LLMs

Phase 4: Deploy & Scale

PRO

Ship your LLM — APIs, infrastructure, enterprise use cases, and the frontier.

20
Deploy Your LLM
From trained model to production API
21
GPU Infrastructure
What you need, what it costs, and how to optimize
22
Private LLMs for Enterprise
Build AI that keeps your data yours
23
Multimodal: Vision + Language
Extend LLMs to understand images and beyond
24
The Future: Where LLMs Are Heading
Scaling laws, AGI, and what comes next