Data science, explained from absolute zero. This course follows one simple idea – STAR: See, Transform, Analyze, Reflect – and a single small cafe dataset that runs through all 65 short lessons, so every concept stays concrete and easy to picture.
You’ll start with what data even is (rows, columns, types, and file shapes), learn to clean and explore it, then build your first models and understand how models actually learn – loss, gradient descent, and overfitting – without heavy math. You’ll meet the common model families (linear and logistic regression, decision trees, random forests, k-nearest neighbors, and a taste of neural networks) and learn a simple map for choosing one.
The second half is what makes this course different: you’ll see how an AI agent actually does data science, following the DS-STAR loop of Analyzer, Planner, Coder, Verifier, Router, Debugger, and Retriever. Then you’ll put it all together in five hands-on projects, from a data-file detective to building your very own mini DS-STAR agent.
