Developing Applications using GitHub Copilot
This comprehensive one-day online class, “Developing Applications using GitHub Copilot” equips participants with advanced skills to maximize productivity using AI-assisted development.
It begins with a thorough introduction to GitHub Copilot, exploring its configuration options, prompt engineering techniques, and comparing various AI models before diving into practical code generation across multiple files and contexts. The next module focuses on extending GitHub Copilot with agent mode and custom tools via the Model Context Protocol, leveraging AI for git operations and source control, and applying Copilot to refactoring tasks and error resolution. Participants will also learn to generate high-quality documentation including code comments and architecture diagrams using Mermaid.
The class concludes with testing strategies across various programming languages and frameworks, including integration testing with Playwright MCP, preparing developers to integrate AI-assisted development seamlessly into their workflow.
Samples will be provided in Python, TypeScript, .NET and Java, but the concepts are applicable to any programming language. Various Cloud scenarios will be addressed.
GitHub Copilot Introduction
- Overview GitHub Copilot
- Enabling & Configuring Feature using Setting
- Prompt Engineering & Reusable Prompts
- Ask, Edit & Agent Mode
- AI Model comparison
Generate Code using GitHub Copilot
- Inline Code Generation and suggestions
- Understanding and using Context
- Chat Participants & Slash Commands
- Mastering Multi-File Edits
- Integrating Code from Online Resources
- Overriding LLM Data using Instructions
Using & Extending Agents Mode
- Github Copilot Agent Mode Overview & Best Practice
- Agent related Settings
- Adding Tools using Model Context Protocol (MCP)
- Manage Tool approvals
- Using Custom Models
Copilot for Git & Source Control
- Effective Commit Messages, Pull Requests
- Using Code Reviews
- Resolving Merge Conflicts
- Creating DevContainers & CodeSpaces
- Using the GitHub MCP
Refactoring, Fixing Errors & Reducing Repetitive Tasks
- Semantic Search & Explaining Code
- Using Thinking Models for Architecture and Code Optimization
- Using Screenshots & Vision for Error Fixing
- Implementing Instructions for Repetitive Tasks
Testing using Copilot
- Implementing Unit Tests
- Fixing Test Errors
- Integration Tests using Playwright MCP
Using Copilot for Documentation
- Code Comments
- Generate Markdown Documentation
- Creating Mermaid Architecture Diagrams