Robots Are After Your Job: Exploring Generative AI for C++
summary
Claim: no-risk performance improvement for std::binary_search by replacing the usual approach with a skewed binary search that biases toward boundaries, implemented as a center-left upper_bound using a 1x4 bias. ChatGPT supplies math, plots and draft code; the talk builds and runs benchmarks (including a 465k-word equal_range test) showing roughly 50% fewer comparisons on that workload and measurable runtime gains with small worst‑case regressions. Experiments and code exist; tool output needs human fixes and prompt tuning.
AI-written summary · about
chapters (22)
- 0:00Welcome
- 2:23Not fluffy keynote
- 3:06Binary search history
- 8:17Algorithm comparisons
- 10:24Binary search performance
- 13:23Classic binary search
- 17:15Math and AI comparisons
- 21:32AI accelerates research
- 27:58Biasing binary search
- 33:21Out-of-bounds handling
- 37:58Extreme search algorithms
- 41:32Skewed binary search
- 44:00Skewed binary search implementation
- 47:58Equal range optimization
- 53:17Equal range bounds
- 1:00:28AI algorithm testing
- 1:06:32Creativity in coding
- 1:17:45Judgment over skill
- 1:21:40Closing finale
- 1:22:04Q&A session part one
- 1:25:59Q&A session part two
- 1:29:28Q&A session part three
related talks
57mRethinking Binary Search: Improving on a Classic with AI Assistance
01h 01mC++hatGPT: How AI Tools Impact Your Life as a Systems Programmer
44mToxic Trust: The Psychology of Automation Bias in Generative AI
39mBack to Basics: Cutting Through the AI-Generated Code Noise
01h 30mFairness Tales: How To Measure And Mitigate Unfair Bias in Machine Learning Models
43mBecome a 10x developer by harnessing the power of Generative AI
50mIdentifying and fixing Issues in Code using AI based tools
55mAI Showdown: The Good, The Glitchy, and The Risky