Track overview: Foundations for AI

Why start with foundations. Here's a situation you'll meet again and again. Someone hands you a sales export with fifty thousand rows, dates in three different formats, missing customer IDs and a few orders worth minus ten dollars. Before any AI model can help, someone has to load that data, clean it, and understand what's actually in it. That's not a side task. In practice, it's most of the work. The Foundations track gives you exactly these skills, so that everything later in the course is built on solid ground instead of guesswork.

Before you start this track. To start this track, you should be comfortable with basic programming in some language, whether that's Python, JavaScript, C sharp or even SQL and shell scripts. You should be able to open a terminal and install software. If you already know Python, great, parts of the first course will be a quick refresher. If you don't, don't worry. We'll cover the pieces of Python that data work actually uses, and you'll pick up the rest as you go.

Three courses, one path. The track has three courses, and they build on each other. First, Python for AI gives you the tools. Second, Math Essentials for ML gives you the ideas behind the tools, without heavy formulas. Third, Data Handling and Preparation shows you how to turn raw data into something a model can learn from. By the end, you'll be ready for the Classical Machine Learning track, where you'll train your first models.

Course 1: Python for AI. In Python for AI, you'll set up a workstation that doesn't break, with isolated environments, VS Code and Jupyter notebooks. Then you'll learn NumPy for fast calculations on large arrays, and pandas for loading, cleaning and joining tables. You'll pull data from the places it really lives: CSV files, JSON, Excel sheets, SQL databases and web APIs. And you'll turn numbers into clear charts. The course ends with a project where you explore an online store's sales dataset and answer real business questions.

Course 2: Math Essentials for ML. Math Essentials for ML covers only the math that machine learning truly relies on, and it's explained visually, with code instead of proofs. You'll see how a customer or a product becomes a vector, a list of numbers, and how distance between vectors measures similarity. You'll get an intuitive feel for gradients, which is how models improve step by step. And you'll learn the probability and statistics you need to read results honestly, like averages, variance, correlation and what a p-value does and doesn't tell you.

Course 3: Data Handling and Preparation. Data Handling and Preparation is where many real projects succeed or fail. You'll deal with missing values and outliers, convert text categories into numbers a model can use, and scale features so they're comparable. You'll learn why we split data into training, validation and test sets, and how data leakage can make a useless model look brilliant. Then you'll build a reproducible pipeline, so the same steps run the same way every time. The project is preparing a customer churn dataset, ready for modelling.

What you'll be able to do. Let's make this concrete. When you finish this track, you'll be able to set up a Python environment that anyone on your team can rebuild. You'll be able to take an unfamiliar dataset, explore it, clean it and chart it on your own. You'll be able to explain ideas like vectors, similarity and gradients in plain words. And you'll be able to hand over a clean, model-ready dataset, with a pipeline that produces it the same way every time.

How to approach this track. A few suggestions. Take the setup lesson seriously and finish its exercise, because every later lesson depends on it. Whenever you can, repeat the examples with data from your own work, like an export from a system you support. And don't rush the math course. If an idea doesn't click the first time, replay the lesson and try the code yourself. These ideas come back in every track that follows.

Recap and next step. To recap, the Foundations track has three courses: Python for AI, Math Essentials for ML, and Data Handling and Preparation. You'll finish with two projects, exploring an online store's sales and preparing a customer churn dataset, and with skills you'll use in every other track. In the next lesson, we'll start at the very beginning and set up your AI workstation. See you there.