# How the Microsoft AI for Beginners Curriculum Structures Its 24 Lessons

> Discover how the Microsoft AI for Beginners curriculum structures its 24 lessons across 12 weeks. Explore foundational AI concepts, neural networks, NLP, ethics, and more in this modular progression.

- Repository: [Microsoft/AI-For-Beginners](https://github.com/microsoft/AI-For-Beginners)
- Tags: architecture
- Published: 2026-08-28

---

**The Microsoft AI for Beginners curriculum delivers a 12-week, 24-lesson modular progression organized into seven thematic sections, moving from foundational concepts and symbolic AI through neural networks, computer vision, and NLP to advanced techniques, ethics, and multi-modal extras.**

The `microsoft/AI-For-Beginners` repository provides a comprehensive, self-contained educational framework. This **AI for Beginners curriculum structure** is defined in the root [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) ([source lines 81-108](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md#content)) and implemented through a hierarchical folder system under the `lessons/` directory.

## 12-Week Curriculum Organization

The curriculum divides 24 lessons across 12 weeks, grouping related topics into logical sections. Each section resides in a dedicated folder under `lessons/`, containing self-contained learning units.

### The Seven Core Sections

| Section | Theme | Lesson Folder | Key Topics |
| ------- | ----- | ------------- | ---------- |
| I | **Introduction to AI** | `lessons/1-Intro/` | Introduction and History of AI |
| II | **Symbolic AI** | `lessons/2-Symbolic/` | Knowledge Representation & Expert Systems |
| III | **Neural Networks** | `lessons/3-NeuralNetworks/` | Perceptron, Multi-Layer Perceptron, Frameworks, Over-fitting |
| IV | **Computer Vision** | `lessons/4-ComputerVision/` | OpenCV, Convolutional Nets, Transfer Learning, GANs, Object Detection |
| V | **Natural Language Processing** | `lessons/5-NLP/` | Text Representation, Embeddings, RNNs, Transformers, LLMs |
| VI | **Other AI Techniques** | `lessons/6-Other/` | Genetic Algorithms, Deep Reinforcement Learning, Multi-Agent Systems |
| VII | **AI Ethics** | `lessons/7-Ethics/` | Responsible AI principles |
| — | **Extras** | `lessons/X-Extras/` | Multi-Modal Networks (CLIP, VQ-GAN) |

## Standard Lesson Template

Every lesson follows a consistent four-component template, as documented in the "Each lesson contains" section of the root README ([lines 20-26](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md#each-lesson-contains)).

- **README**: Theoretical overview and learning objectives
- **Pre-reading**: Suggested background materials
- **Executable Jupyter notebooks**: Interactive code demonstrations (e.g., `Perceptron.ipynb` in `lessons/3-NeuralNetworks/03-Perceptron/`)
- **Lab (optional)**: Hands-on exercises for practical application

## Repository Structure and Navigation

The curriculum's modular architecture allows learners to enter at any topic while maintaining pedagogical coherence.

### Root Configuration and Setup

The entry point is [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) in the repository root, which provides the high-level overview and lesson table. Environment preparation is handled in [`lessons/0-course-setup/setup.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/0-course-setup/setup.md), containing prerequisite installation instructions.

### Lesson Directories

Each numbered folder corresponds to a curriculum section and contains a [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) for that section:

- [`lessons/1-Intro/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/1-Intro/README.md) – Foundational concepts and history
- [`lessons/2-Symbolic/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/2-Symbolic/README.md) – Symbolic AI theory and notebooks
- [`lessons/3-NeuralNetworks/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/3-NeuralNetworks/README.md) – Neural network fundamentals and labs
- [`lessons/4-ComputerVision/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/4-ComputerVision/README.md) – Computer vision pipelines and deep learning labs
- [`lessons/5-NLP/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/5-NLP/README.md) – NLP pipeline, embeddings, and transformers
- [`lessons/6-Other/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/6-Other/README.md) – Genetic algorithms, RL, and multi-agent systems
- [`lessons/7-Ethics/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/7-Ethics/README.md) – Responsible AI principles
- [`lessons/X-Extras/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/X-Extras/README.md) – Cutting-edge multi-modal content

## Pedagogical Progression

The curriculum follows a deliberate learning trajectory. It begins with **foundational concepts** (Sections I-II: intro and symbolic AI), advances to **core deep-learning building blocks** (Sections III-V: neural networks, computer vision, NLP), then explores **advanced techniques** (Section VI: reinforcement learning, multi-agent systems), and concludes with **ethical considerations** (Section VII) and **cutting-edge extras** (multi-modal networks).

## Practical Implementation

Learners interact with the curriculum through executable notebooks and utility modules.

To programmatically execute a lesson notebook, use `nbformat` and `ExecutePreprocessor`:

```python
import nbformat
from nbconvert.preprocessors import ExecutePreprocessor

nb_path = "lessons/3-NeuralNetworks/03-Perceptron/Perceptron.ipynb"
with open(nb_path) as f:
    nb = nbformat.read(f, as_version=4)

ep = ExecutePreprocessor(timeout=600, kernel_name="python3")
ep.preprocess(nb, {"metadata": {"path": "./"}})

```

Lesson-specific utilities can be imported for custom experiments. For example, the Transformers lesson provides a tokenizer in [`torchnlp.py`](https://github.com/microsoft/AI-For-Beginners/blob/main/torchnlp.py):

```python
from lessons.5_NLP.18_Transformers.torchnlp import simple_tokenizer

sentence = "Artificial Intelligence for Beginners"
tokens = simple_tokenizer(sentence)
print(tokens)   # → ['Artificial', 'Intelligence', 'for', 'Beginners']

```

Both patterns—**executing provided notebooks** and **importing lesson utilities**—represent the hands-on methodology encouraged throughout the curriculum.

## Summary

- The Microsoft AI for Beginners curriculum structure consists of **24 lessons over 12 weeks**, divided into seven thematic sections plus extras.
- Each lesson resides in a dedicated folder under `lessons/` (e.g., `lessons/3-NeuralNetworks/`) and contains a README, pre-reading materials, Jupyter notebooks, and optional labs.
- The pedagogical flow moves from **symbolic AI foundations** through **deep learning fundamentals** (neural networks, CV, NLP) to **advanced techniques** and **AI ethics**.
- All lesson content is defined in the root [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) ([lines 81-108](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md#content)) with standardized templates documented at [lines 20-26](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md#each-lesson-contains).

## Frequently Asked Questions

### How many lessons are in the AI for Beginners curriculum?

The curriculum contains **24 lessons** structured as a **12-week** course, as defined in the root [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md). Each lesson is designed to be self-contained while contributing to the overall progression from basic AI concepts to advanced applications.

### What is the standard folder structure for lessons?

Each lesson follows the path `lessons/{section-number}-{SectionName}/{lesson-number}-{Topic}/`. For example, the Perceptron lesson is located at `lessons/3-NeuralNetworks/03-Perceptron/`, containing the executable `Perceptron.ipynb` notebook and supporting files.

### Does every lesson include hands-on lab exercises?

Every lesson includes **executable Jupyter notebooks** and a **README**, but labs are explicitly marked as **optional** in the template structure ([source lines 20-26](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md#each-lesson-contains)). The notebooks themselves provide interactive coding experiences that serve as the primary hands-on component.

### What topics are covered in the Extras section?

The **Extras** section (`lessons/X-Extras/`) covers cutting-edge topics not included in the core curriculum, specifically **multi-modal networks** such as CLIP and VQ-GAN. This section extends the foundational knowledge into emerging research areas.