Probability, distributions, confidence intervals, hypothesis testing, linear and logistic regression, and Bayesian thinking — with Python code throughout. The statistics course that builds intuition before introducing formulas.
This is a text-first course that links out to the best supporting material on the internet instead of trying to replace it. The goal is to make this the best course on statistics you can find — even without producing a single minute of custom video.
This course is built by engineers who ship statistics systems in production. It reflects how these tools actually behave at scale.
Every day includes working code examples you can copy, run, and modify right now. Understanding comes through doing.
Instead of re-explaining existing documentation, this course links to the definitive open-source implementations and the best reference material on statistics available.
Each day is designed for about an hour of focused reading plus hands-on work. Do the whole course over a week of lunch breaks. No live classes, no quizzes.
Each day stands alone. Read them in order for the full picture, or jump straight to the day that answers the question you have today.
Sample spaces, events, conditional probability, Bayes' theorem with real examples, independence, and the probability intuitions that explain why A/B tests and machine learning models work the way they do.
The normal distribution and why it appears everywhere (Central Limit Theorem), binomial for success/failure counts, Poisson for event rates, and how to choose the right distribution for your data.
Standard error, confidence intervals, p-values (what they mean and what they don't), t-tests, chi-square tests, the multiple testing problem, and statistical power — with Python scipy code throughout.
Simple and multiple linear regression, the assumptions behind OLS, coefficient interpretation, logistic regression for classification, ROC curves, and the model evaluation metrics that matter for each use case.
Prior vs posterior, Bayesian updating, why Bayesian A/B testing gives more actionable results than frequentist p-values, minimum detectable effect calculations, and building a simple Bayesian A/B test in Python.
Instead of shooting our own videos, we link to the best deep-dives already on YouTube. Watch them alongside the course. All external, all free, all from builders who ship this stuff.
Complete statistics courses for data scientists — probability, distributions, hypothesis testing, and regression with Python examples.
p-values, confidence intervals, t-tests, and statistical significance — explained intuitively with examples that build real understanding.
Bayesian reasoning, prior and posterior distributions, and how Bayesian inference differs from frequentist methods — with visual examples.
Building, fitting, and evaluating linear regression models in Python with sklearn and statsmodels.
Designing valid A/B tests, sample size calculations, and the Bayesian vs frequentist approaches to interpreting results.
The best way to deepen understanding is to read the canonical open-source implementations. Clone them, trace the code, understand how the concepts in this course get applied in production.
Code for Statistical Rethinking — the best Bayesian statistics textbook. Python and R implementations of all models.
The SciPy source. The /scipy/stats directory has implementations of every distribution and statistical test used in this course.
Probabilistic programming in Python. The most accessible library for building Bayesian models — the examples directory covers all use cases in this course.
Statistical modeling in Python — the library for regression diagnostics, hypothesis tests, and time series analysis that scipy alone doesn't cover.
Most ML engineers can train a model but can't explain why the evaluation metrics are or aren't meaningful. This course builds the statistical foundation that makes ML work interpretable.
A/B tests, feature rollouts, and conversion optimization all require statistical thinking. This course gives you the conceptual tools to design and interpret experiments correctly.
Understanding confidence intervals, p-values, and regression makes your analysis rigorous. This course gives you the vocabulary to communicate statistical findings accurately.
The 2-day in-person Precision AI Academy bootcamp covers data science and statistics in depth — hands-on, with practitioners who build AI systems for a living. 5 U.S. cities. $1,490. 40 seats max. June–October 2026 (Thu–Fri).
Reserve Your Seat