Vizuara BooksThe library
Every book, one library.
Six books are free forever; every other book opens its first chapter. Start with any book.
New collection
Charlie and the Intelligence Factory
Explore the world of language, vision, voice and reason. Four books, one per modality, each following a real Vizuara workshop, chapter for lecture.
Free forever
9 · 9 · no subscription neededStart here
13 · beginner
Build Decision Trees from Scratch
Deep Learning Fundamentals
Foundations for AI & ML
Generative AI Fundamentals
Git & GitHub Masterclass
FreeMachine Learning Fundamentals
Mathematical Foundations for Machine Learning
The Modern Software Developer
FreeNeural Networks from Scratch
FreePrompt Engineering
FreePython for Data Science
R Masterclass
SQL Masterclass
Go deeper
20 · intermediate
AI Agents Bootcamp
AI Context Engineering
FreeBuild Large Language Models (LLMs) from Scratch
Charlie and the Sound Room
Charlie and the Vision Room
Claude Certified Architect (CCAF)
Convolutional Neural Networks Fundamentals
Computer Vision Bootcamp
Data Structures & Algorithms in Python
LLM Finetuning
LLM Production & Deployment
Memory in AI Systems
Build a Mini ClawdBot from Scratch
Machine Learning & Deep Learning Mastery
NLP & Computer Vision Mastery
FreeRAG in Production
Reinforcement Learning
Transformers: Theory, intuition, and Building from Scratch
Build a Vision Transformer (ViT) from Scratch
Writing Journal & Conference Papers
For builders
16 · advanced
5D Parallelism for Large Model Training
Charlie and the Language Room
Charlie and the Reasoning Room
FreeDeepSeek Harness
Build a Data-Efficient Image Transformer (DeiT) from Scratch
Build a Diffusion Language Model from Scratch
Build Kimi K3 from Scratch
FreeHow to Host Kimi K3
Modern Robot Learning
Build a NanoVLM from Scratch
Pi vs Hermes vs Codex: Context Compaction and Memory
FreePretraining a Mini Kimi K3
RLHF from Scratch
Scientific Machine Learning (SciML)
Transformers for Vision & Multimodal LLMs
Vision-Language-Action & World Models for Robotics
Coming soon
3 · in the works · see the roadmap
Coming soonHarness Engineering
Build the layer that turns a model into an agent.
Coming soonKernel Engineering
From silicon to speculative decoding — GPU kernels for modern LLMs.
Coming soonInference Engineering
Serve LLMs fast — batching, KV-cache, paged attention, speculative decoding.