A platform to learn AI structurally, through visualizations.

Learn from the fundamentals up — CS, math, machine learning, deep learning, agents, loops and graphs, and the latest research — all 297+ pages built around real, computed, interactive visualizations, not static diagrams.

Start Learning →

or start from a real problem instead →

Interactive

Attention Step-Through

Sentence (up to 12 tokens)

Attention head

Every value below is real, computed live from “the” — color encodes sign and magnitude (green = positive, red = negative, opacity = size), the same value-as-color convention used throughout this walkthrough, now live instead of a fixed worked example:

Embedding

-0.62

0.34

0.55

-0.11

Query

-0.51

-0.14

0.18

-0.03

Key

-0.23

0.50

-0.26

-0.31

Value

-0.08

-0.59

-0.17

0.10

Attention from “the” to every token:

the

17%

cat

18%

sat

17%

the

17%

mat

15%

Real Q·Kᵀ/√d_k → softmax → weighted-sum-of-V math, computed on deterministic demo embeddings (see the component note for exactly what's simplified and why).

From Attention & Transformers — one of 297+ pages built the same way.

New

Test yourself

Real interview questions, click to reveal the answer — try one right here:

More real interview questions, across LLMs, RAG, and agents →

Write real code

Implement a function yourself in a real, in-browser Python sandbox — run it against real test cases (pass/fail, no LLM grading), then reveal a reference solution.

53+ problems: linear algebra to RL →