FREE LEARNING PATH
Learn AI from the mechanisms.
Study search, probability, model fitting, evaluation, neural networks, and attention through eight chapters and a reproducible Python experiment.
Before you begin
Basic algebra and Python variables, loops, and functions. Every chapter states its prerequisites. All reading is open, and the Python examples use no paid APIs or packages.
Search and state
Represent a problem as a graph, trace uniform-cost search, and state the conditions behind A*.
CHAPTER 2 · 35 MIN WITH PRACTICEProbability and evidence
Compute conditional probabilities, update a belief with Bayes’ rule, and make a decision from explicit costs.
CHAPTER 3 · 45 MIN WITH PRACTICELinear models and gradients
Define a loss function, derive its gradients, and implement simultaneous parameter updates in plain Python.
CHAPTER 4 · 40 MIN WITH PRACTICEClassification and decisions
Turn a linear score into a probability estimate and measure the consequences of changing a threshold.
CHAPTER 5 · 40 MIN WITH PRACTICEGeneralization and leakage
Design an honest experiment, keep related observations together, and explain what a test score can support.
CHAPTER 6 · 45 MIN WITH PRACTICENeural networks and backpropagation
Follow one computation forward, apply the chain rule backward, and verify gradients numerically.
CHAPTER 7 · 45 MIN WITH PRACTICEAttention and language models
Calculate an attention row, trace a causal mask, and distinguish token prediction from factual verification.
CHAPTER 8 · 60 MIN WITH PRACTICEA reproducible experiment
Build a small classifier from scratch, choose a threshold on validation data, and write an honest results report.
Each chapter includes an original reading, a worked exercise, understanding checks, and saved progress. Start with the Computer Science Core if programming is new.