How to Contribute Code to AI For Beginners: A Complete 10-Step Guide

To contribute code to AI For Beginners, fork the Microsoft repository, create a feature branch, make your changes in the lessons/ or examples/ directories, verify with the provided conda environment, and submit a Pull Request through the Microsoft CLA workflow.

The AI For Beginners repository by Microsoft is an open-source AI curriculum containing 24 lessons, code examples, and interactive quizzes. Contributing follows a standard GitHub workflow with Microsoft-specific requirements for licensing. This guide walks through the exact steps based on the project's CONTRIBUTING.md and source structure.

Ready the Prerequisites

Before writing any code, understand the project's legal and technical requirements.

Sign the Microsoft CLA

The project requires the Microsoft Contributor License Agreement (CLA). You only sign this once across all Microsoft open-source projects. A CLA-bot automatically detects when you open your first Pull Request and guides you through the signing process in a comment on your PR【/cache/repos/github.com/microsoft/AI-For-Beginners/main/CONTRIBUTING.md#L5-L10】.

Set Up Your Development Environment

The repository includes an environment.yml file that pins exact versions of TensorFlow, PyTorch, and other dependencies. Create this environment to ensure your code runs identically to the production curriculum:

conda env create -f environment.yml
conda activate ai-for-beginners

Fork and Clone the Repository

Create your personal copy of the codebase to work on.

Fork on GitHub

Click Fork on the microsoft/AI-For-Beginners repository page. This creates <your-username>/AI-For-Beginners under your account.

Clone Your Fork

For most contributions, clone normally:

git clone https://github.com/<your-username>/AI-For-Beginners.git
cd AI-For-Beginners

The repository is large due to the translations/ folder. For translation work specifically, consider a sparse checkout to avoid downloading all language files【/cache/repos/github.com/microsoft/AI-For-Beginners/main/README.md#L33-L40】.

Create a Feature Branch

Isolate your work to keep the main branch clean:

git checkout -b fix-typo-lesson-3

Use descriptive branch names:

• fix-perceptron-notebook-typo
• add-translation-es-lesson-1
• update-quiz-app-dependencies

Navigate Source Code Locations

Understanding the repository structure ensures you edit the correct files.

Directory	Purpose
lessons/	Core curriculum content—Jupyter notebooks and README markdown files organized by topic
examples/	Standalone Python scripts demonstrating AI concepts
etc/quiz-app/	Vue.js frontend for the interactive quiz application
translations/<lang-code>/	Localized curriculum versions (e.g., translations/es/)

The master README.md at repository root contains the curriculum map and "Getting Started" instructions【/cache/repos/github.com/microsoft/AI-For-Beginners/main/README.md#L45-L52】.

Make Your Code Changes

Edit files relevant to your contribution type.

Fix Lesson Content

Example: correcting a typo in the Perceptron lesson.

Before (lessons/3-NeuralNetworks/03-Perceptron/README.md):

Perceptron is a **simple** binary classification model that uses a **single**

After:

Perceptron is a **simple** binary classification model that uses a **single** neuron.

Add Example Code

Creating a new helper function in examples/01-hello-ai-world.py:

def greet(name: str) -> str:
    """Return a friendly AI greeting."""
    return f"Hello, {name}! Welcome to AI for Beginners."

print(greet("Alice"))

Update the Quiz Application

For Vue.js changes in etc/quiz-app/, run linting before committing:

cd etc/quiz-app
npm install
npm run lint

Verify Your Changes

Always confirm modifications work as expected.

Test Jupyter Notebooks

Execute notebooks end-to-end in your conda environment:

jupyter notebook lessons/3-NeuralNetworks/03-Perceptron/Perceptron.ipynb

Run all cells and verify no errors appear. Notebooks are a primary delivery format for the curriculum.

Test Python Scripts

Run modified scripts directly:

python examples/01-hello-ai-world.py

Commit with Clear Messages

Describe what changed and why:

git add lessons/3-NeuralNetworks/03-Perceptron/README.md
git commit -m "Fix typo in Perceptron description: add missing 'neuron'"

Or for the greeting helper:

git add examples/01-hello-ai-world.py
git commit -m "Add greeting helper to hello-ai-world example

Provides a reusable function for personalized AI welcome messages."

Push and Open Your Pull Request

Upload your branch and initiate review:

git push origin fix-typo-lesson-3

Then on GitHub:

1. Click Compare & pull request
2. Select main as the base branch
3. Write a concise description referencing the issue (if applicable)
4. Submit the PR

The CLA-bot will automatically label your PR and comment if CLA signing is required.

Address Review Feedback

Maintainers may request:

• Link updates in markdown files
• Code formatting adjustments
• Additional test coverage
• Linting fixes for quiz-app submissions

Respond to comments promptly and push additional commits to the same branch—they automatically appear in the PR.

Special Contribution Paths

Add Translations

Place new language content under translations/<lang-code>/ following existing naming conventions. Translation-specific guidelines live in etc/CONTRIBUTING.md【/cache/repos/github.com/microsoft/AI-For-Beginners/main/CONTRIBUTING.md#L19-L22】.

Example structure:

translations/
├── es/          # Spanish

├── zh-cn/       # Simplified Chinese

├── your-lang/   # Your new translation

Contribute to the Quiz App

The Vue.js application in etc/quiz-app/ requires:

• npm install for dependencies
• npm run lint for code quality
• Manual testing of UI changes in a browser

Summary

• Sign the Microsoft CLA once before your first contribution
• Fork and clone from https://github.com/<your-username>/AI-For-Beginners
• Use the conda environment in environment.yml for dependency consistency
• Edit files in lessons/ for curriculum content, examples/ for Python scripts, or etc/quiz-app/ for the Vue.js frontend
• Verify notebooks run end-to-end before submitting
• Follow the 10-step workflow: branch → change → test → commit → push → PR → review → merge

Frequently Asked Questions

Does Microsoft AI For Beginners require a CLA for all contributions?

Yes. Every Pull Request triggers the Microsoft CLA-bot, which checks whether you have signed the Contributor License Agreement. If you haven't, the bot provides a direct link to complete the one-time signing process【/cache/repos/github.com/microsoft/AI-For-Beginners/main/CONTRIBUTING.md#L5-L10】.

Can I contribute without downloading the entire repository?

Yes. The translations/ folder contains large language-specific files. Use Git's sparse-checkout feature to clone only the directories you need, as mentioned in the README.md clone instructions【/cache/repos/github.com/microsoft/AI-For-Beginners/main/README.md#L33-L40】.

What dependencies do I need to run the curriculum code?

The environment.yml file pins exact versions of TensorFlow, PyTorch, and supporting libraries. Create the conda environment with conda env create -f environment.yml to ensure your code runs identically to the published curriculum.

How do I add a new language translation to AI For Beginners?

Create a folder under translations/ using the standard language code (e.g., translations/de/ for German). Follow the file structure of existing translations and consult etc/CONTRIBUTING.md for translation-specific formatting guidelines【/cache/repos/github.com/microsoft/AI-For-Beginners/main/CONTRIBUTING.md#L19-L22】.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}