Data science is the practice of asking questions of data and finding honest answers — a skill used everywhere from public health to business to sports. This track starts with statistics fundamentals, moves through Python's most-used data libraries, and ends with a first taste of machine learning.
Builds on PythonReal DatasetsFirst ML Model4 Phases
Mentor TipAlways plot your data before you model it — a five-second chart often reveals what an hour of code would miss.
The Roadmap
01
FoundationsThinking like a data scientist
Before any code, this phase builds the statistical intuition that makes every later tool make sense.
01
What data science actually isThe difference between data analysis, data science, and machine learning.
02
Statistics basicsMean, median, mode, and standard deviation — the vocabulary of every dataset.
03
From spreadsheets to codeWhy code eventually beats spreadsheets for serious analysis.
04
Asking good questionsThe most underrated data science skill: framing a clear, answerable question.
Practice ProjectTake any spreadsheet of data you have access to and write down five questions you could ask of it.
02
Python for DataThe core toolkit
Two libraries — NumPy and pandas — do the vast majority of real-world data work in Python.
01
NumPy basicsFast numerical arrays, the foundation everything else is built on.
02
pandas DataFramesLoading, viewing, and slicing tabular data like a spreadsheet in code.
03
Cleaning messy dataHandling missing values, duplicates, and inconsistent formatting.
04
Grouping & aggregatingSummarizing data by category — the pandas equivalent of a pivot table.
Practice ProjectLoad a public CSV dataset (e.g. from Kaggle) and produce a clean summary table by category.
03
Visualization & AnalysisMaking data speak
A great chart communicates an insight in seconds — this phase teaches you to build ones that do.
01
Matplotlib basicsLine charts, bar charts, and histograms from scratch.
02
Seaborn for polishStatistically-aware charts with far less code.
03
Exploratory data analysisA repeatable process for getting to know any new dataset.
04
Drawing honest conclusionsAvoiding the most common ways charts mislead, even accidentally.
Practice ProjectProduce a 3-chart mini-report on a dataset of your choice, with one written insight per chart.
04
Intro to Machine LearningYour first predictive model
This phase is a gentle first step into machine learning — enough to understand what it is and build one real model.
01
Core ML conceptsTraining data, features, and predictions, explained without the jargon.
02
scikit-learn basicsThe standard, beginner-friendly Python ML library.
03
Your first modelTraining a simple regression or classification model.
04
Evaluating a modelUnderstanding whether your model is actually any good.
Practice ProjectTrain a simple model to predict a value in a beginner Kaggle dataset (e.g. house prices or Titanic survival).
Free Video Courses
Curated, highly-rated, and free — pick whichever teaching style matches how you learn best.