Intermediate
View course →Experimental Design
Plan randomized studies with estimands, power, validity checks, and review-ready design memos.
Courses · Open Textbook · Certificates
Learn the methods behind trustworthy adaptive decision systems
Structured courses and an open textbook for experiments, causal inference, Bayesian reasoning, A/B testing, and reinforcement learning.
Structured courses
Build portfolio-ready artifacts
Enroll in guided courses with lesson notebooks, capstones, rubrics, templates, and certificates. Each course ends in a concrete work sample.
Advanced
View course →Causal Inference
Use DAGs, matching, DiD, IV, RDD, and sensitivity analysis to make defensible causal claims.
Intermediate
View course →A/B Testing at Scale
Run trustworthy experiments with metric contracts, diagnostics, platform checks, and review boards.
Open textbook
Read the foundations for free
The textbook remains open and interactive. Use it as a reference, or pair it with the courses for assignments and certificates.
Why Randomize?
Understand why randomized experiments are the gold standard for causal claims — and when they're not enough.
Causal Graphs
Learn to draw and reason with DAGs: spot confounders, find adjustment sets, and avoid classic traps like collider bias.
Bandits & RL
Go beyond A/B tests: learn how multi-armed bandits balance exploration and exploitation in real time.
Causal AI
See how machine learning and causal inference combine — from doubly-robust estimators to counterfactual prediction.
Research roadmap · P11+
11 upcoming research directions
Each idea extends P1–P10 to close a specific methodological gap in behavioural self-experimentation.
Table of Contents
All 10 Chapters
Read in any order, though 1 through 10 works best.
- 118 min
Why Experiment?
The surprisingly tricky problem of figuring out what actually causes what
- 220 min
Designing Experiments
Sample sizes, randomization schemes, and how not to fool yourself
- 322 min
Causal Graphs
Drawing the invisible: how to see confounders, mediators, and colliders
- 420 min
Counterfactuals
What would have happened? The language of causation
- 522 min
Bayesian Thinking
Updating beliefs with evidence, from Bayes' theorem to posterior distributions
- 625 min
Multi-Armed Bandits
The exploration-exploitation trade-off and why A/B tests waste half your traffic
- 728 min
Reinforcement Learning
Sequential decision-making: when actions have long-term consequences
- 824 min
Effect Estimation
Getting the numbers right: regression, matching, and doubly-robust methods
- 920 min
When Experiments Fail
Sensitivity analysis, E-values, and what to do when you can't randomize
- 1026 min
Causal AI
Combining machine learning with causal reasoning