Conclusion - accentient/github-copilot-devs GitHub Wiki

Key Takeaways

• GitHub Copilot is an AI-powered pair programming assistant designed to help with code implementation rather than design
• GitHub Copilot Chat functions as an AI research assistant for developers
• Copilot's results are non-deterministic, meaning it can make mistakes or produce varying outcomes
• Training date matters: Copilot's knowledge is current as of October 2023
• The generated code may not be perfect or ideal—you may need to refine it (with or without Copilot's assistance), similar to how higher-level languages improved upon older ones like assembly
• Your programming skills remain essential: You are ultimately responsible for all the code in your codebase
• Prompt engineering is key: The quality of Copilot’s suggestions depends on how much context you provide
• Context management: Different IDEs handle context differently (e.g., typically limited to open tabs). Be intentional in providing hints and structure.
• Specialized tools still have their place: Continue using linting, static code analysis, and other tools alongside Copilot
• Don’t hesitate to ask Copilot for help with tasks related to the IDE, code, or workflows
• Use Copilot as a resource to expand your understanding of concepts or approaches
• Stay informed about updates: New features are regularly added to GitHub Copilot (check the GitHub Copilot changelog)