welcome

This book is about learning to think the Bayesian way: reasoning under uncertainty, updating your beliefs as new information comes in, and making better decisions when data are noisy, incomplete, or ambiguous. If you’ve ever asked yourself questions like “What’s the probability this treatment works?” or “How confident should I be in my model’s prediction?”, this book is for you.

To get the most out of Grokking Bayes, you should be comfortable with Python and have some exposure to basic probability or statistics. You don’t need to be a math expert. The goal of the book is to make Bayesian reasoning intuitive, visual, and directly useful for real-world problems.

Along the way, you’ll see how Bayesian methods power applications from A/B testing to predicting house prices, and even provide a way to think about modern AI systems like large language models. Through many examples, vivid illustrations, and hands-on code (using PyMC and ArviZ), you’ll learn how to:

Define and use priors, likelihoods, and posteriors.

Apply conjugate priors where the math works out cleanly.

Use MCMC and variational inference when exact solutions aren’t possible.

Check and compare models to avoid being misled by spurious results.

Train dedicated Bayesian models such as mixtures, hierarchical models, state spaces model, Bayesian neural networks.

Make practical, uncertainty-aware decisions in data science and AI.