The present and future of programming with AI (Keynote)
summary
Progress from 2023 AlphaCode to 2025 shows LLMs go from messy contest solutions to clean, efficient code (Claude reproduces AlphaCode’s backspace solution with fewer bugs), and logic puzzles move from 0/9 LLMs solving Cheryl’s birthday to a majority by May; prototyping becomes 10x faster while whole-life-cycle gains sit around 20–50%, with some users reporting short-term productivity drops. Real-time IDEs aim for subsecond responses, teams experiment with agent ensembles, and a live app littlepiper.ai demonstrates conversational, Socratic programming and critique.
AI-written summary · about
chapters (10)
- 0:00The evolution of coding
- 1:55Ancient history of AI
- 4:08DeepMind and AlphaCode
- 6:09Modern AI capabilities
- 8:31AI-driven development
- 9:45Agents and architecture
- 12:31Interactive learning tools
- 13:23Redefining programming
- 16:15Future languages and OS
- 17:40Discussion and conclusion
related talks
45mProgrammer’s Apprentice Season 2: Future Directions in AI-assisted Coding
43mFrom Lambda to LLMs: AI Coding and the Future of Programming
27mWill AI (Finally) Replace Developers?
24mHow AI is increasing the surface area for serendipity
01h 28mFuture of Software Programming with AI Bots at Hand
30mThe AI Revolution in Coding
45mThe AI Revolution in Software Development: From Code Writing to Quality Assurance
32mComputer Programming Is Dead; Long Live AI-First Programming