The Many Ends of Programming
summary
Six concrete outcomes and a few measured surprises decide whether to watch: evidence of extreme completion with Cursor and an Open Hands agent producing a 48-file pull request for tedious fixes; a Parcel-style experiment turning 61 lines of prompts into a 220-line Python program that works as a Lisp interpreter; conflicting studies (Uplevel finds higher bug rates, GitHub Copilot finds quality gains) and industry data where copy-paste exceeded refactoring in 2024.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 5:55AI and programming scenarios
- 6:39Extreme completion limits
- 9:51Developer apocalypse
- 11:38Abstraction leap and DSLs
- 17:25Beyond code
- 18:51Review economy
- 20:57Garbage risk
- 24:07Scenarios interrelation
- 25:54Future of software
related talks
26mAll the projects I didn't finish, and why it was worth it
30mThe End of Fun
01h 12mKeynote: "Am I A Good Programmer?"
41mCoding for Curiosity
48mWhy Johnny Can't Code Good
01h 15m10x Development: LLMs For the working Programmer
01h 01mFixing the Ultimate Legacy System: You
43mYou Belong In A Room Full of Wizards