How I Made Peace with AI
summary
A hands-on report on using GitHub Copilot over months, showing it excels at generating boilerplate, translating between programming languages, and speeding repetitive tasks within VS Code and Android Studio for Flutter/Dart. The talk gives measured adoption (27% of suggestions accepted) and notes success cases (e.g., producing Swift from commented Dart, scaffolding JSON models) plus concrete failures: subtle algorithmic bugs, hallucinated or copyrighted code, unclear data reuse, and the service’s cost (~€10/month).
AI-written summary · about
chapters (6)
- 0:00Introduction
- 5:40GitHub Copilot stats
- 14:11Live demo of Copilot
- 21:36Copilot limits and risks
- 35:44Best practices
- 39:52Conclusion & Q&A
related talks
25mImproving developer happiness with AI
36mME, MYSELF AND AI
01h 00mLevel up with GitHub Copilot: using AI to learn, code, and build
32mBeyond the Hype: A Pragmatic Look at How Developers Actually Use AI Tools
53mImprove Your Code Quality and Performance using AI and GitHub Copilot
01h 20mGithub Copilot - The World's Most Widely Adopted AI Developer Tool
01h 02mLevel up with GitHub Copilot: using AI to learn, code, and build
01h 00mGitHub Copilot, using AI to help you learn, code, and build