Welcome to the hands-on portion of the Introduction to GitHub Copilot session.
This repository contains five guided exercises (plus a bonus) that take you from your first Copilot autocomplete all the way to exploring GitHub.com features and custom instructions.
Before you start, make sure you have:
- VS Code installed (code.visualstudio.com)
- GitHub Copilot extension installed in VS Code (
GitHub.copilot+GitHub.copilot-chat) - A GitHub account with Copilot access (your company license)
- Git basics done (clone, commit, push, pull requests)
- Python 3.8+ installed (check with
python --version)
- Fork this repository to your own GitHub account (button top-right on GitHub)
- Clone your fork locally:
git clone https://github.com/<your-username>/copilot-introduction.git cd copilot-introduction
- Open in VS Code:
code . - Open the Copilot Chat panel (the chat bubble icon in the left sidebar, or
Ctrl+Alt+I) - You're ready — start with Exercise 01 below!
Work through these in order. Each exercise has its own README.md with step-by-step instructions.
| # | Folder | Title | What you'll practise |
|---|---|---|---|
| 01 | exercises/Data engineering/01-first-contact/ |
First Contact | Inline completions + Copilot Chat basics |
| 02 | exercises/Data engineering/02-sql-assistant/ |
SQL Assistant | Generate & optimise SQL queries |
| 03 | exercises/Data engineering/03-custom-instructions/ |
Custom Instructions | Teach Copilot about your project |
| 04 | exercises/Data engineering/04-github-features/ |
GitHub.com Features | Copilot on pull requests & issues |
| 05 | exercises/Data engineering/05-skills/ |
Skills & Slash Commands | /explain, /fix, /tests, @workspace |
| 🎁 | exercises/Data engineering/bonus/ |
Agent Mode Pipeline | Open-ended: build an ETL script with Agent Mode |
Each exercise takes roughly 10 minutes. The bonus is open-ended — jump in if you finish early.
- Tab accepts an inline suggestion. Esc dismisses it.
- Open Copilot Chat with
Ctrl+Alt+I(Windows/Linux) orCmd+Option+I(Mac). - Highlight code, right-click → Copilot to access quick actions.
- If a suggestion isn't great, try rephrasing your comment or prompt — context matters.
- Copilot learns from the files open in your editor. Keep relevant files open!
All exercises share a common dataset in data/sales_sample.csv — a small fictional sales table with orders, customers, products, and regions. You don't need to load it anywhere; the exercises will point you to it as needed.