Course details
AI & Data Scientist
Learn the working loop of a data scientist: turn a fuzzy request into a metric, measure uncertainty, run fair experiments, estimate effects you cannot randomise, evaluate models honestly and explain what you found. Every lesson follows Pantry, one grocery-delivery app, from its first question to its final report.
Course curriculum
Mathematics for Framing Questions4 lessons
Start where every analysis starts: the decision. Turn requests into metrics with clear numerators and denominators, report change in points and percents without confusion, weigh options by expected value, and use logarithms for skewed data and growth.
Statistical Inference4 lessons
Every number from a sample wobbles. Measure the wobble with a standard error, turn it into a confidence interval, test whether a difference could be luck, and size a study so it can actually find the effect you care about.
Experiments and Causal Effects5 lessons
Separate cause from coincidence. Spot confounders, design and read an A/B test, and when you cannot randomise, estimate effects with regression and difference-in-differences.
Exploring and Explaining Data5 lessons
Check what one row means, ask why values are missing, catch Simpson's paradox before it catches you, draw charts that make one point, and write up a result so a busy reader can act on it.
Evaluating Models5 lessons
Split data the way the model will be used, beat a simple baseline, read precision and recall from a confusion matrix, choose a threshold from real costs, and hunt down leakage before it fools you.