The Time Traveler's Playbook for Coding with AI
summary
Shows step-by-step tactics for getting GitHub Copilot to produce usable code: use big contextual comments, give one‑to‑few examples (zero/one/few‑shot), and iterate prompts until results match specs. Claims developers adapt in seven days to adapt and demos a Next.js markdown editor built with react-markdown and a pregnancy due‑date calculator whose logic Copilot generates and saves to a decentralized web node (Web5 DWN). Includes concrete prompt texts, API names, and live debugging moments.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker intro
- 1:33Three types of technologists
- 3:40Adapting to culture and tech
- 8:14Investigating GitHub Copilot
- 10:43Adapting to AI coding workflow
- 12:55Zero-shot, one-shot, few-shot learning
- 15:52Context and examples for Copilot
- 17:59Demo: Pregnancy tracker with Copilot
- 22:51Debugging and final demo steps
- 24:00Closing and Q&A
related talks
25mImproving developer happiness with AI
49mUnlocking Success: Navigating the Code of Life with GitHub CoPilot
40mLevel Up with Copilot
40mUnlocking Productivity with GitHub Copilot Best Practices and Tips
31mPrompt Like a Pro & Vibe Code with AI Tools
01h 20mGithub Copilot - The World's Most Widely Adopted AI Developer Tool
01h 02mSkills for the age of AI developer tools
59mBeyond The Hype: Every Day, Real World Uses for Copilot