Future of Software Programming with AI Bots at Hand
summary
Claim: large language models materially change developer workflows but do not yet replace engineers. Demonstrations use ChatGPT/GPT‑4 and GitHub Copilot to validate requirements, generate C++ functions (distance with chrono-style units), and produce Java BDD tests; some outputs compile and some contain subtle bugs that require human review (one fix is as small as adding _t to a type). Presenter cites model benchmarks at 85–87% pass rate on coding challenges, warns of hallucinations, token costs and security risks, and recommends use LLMs to draft and test but always read and review the code.
AI-written summary · about
chapters (8)
- 0:00AI bots in programming
- 10:19Large language models and limits
- 21:17AI silver bullet for development
- 32:39Coding and testing with AI
- 50:33Debugging AI-generated code
- 1:01:02C++ with AI considerations
- 1:15:08AI-driven programming era
- 1:25:42Q&A
related talks
01h 01mC++hatGPT: How AI Tools Impact Your Life as a Systems Programmer
01h 02mPANEL - Unleashing the Future: The Impact of Gen AI on Development Productivity
45mProgrammer’s Apprentice Season 2: Future Directions in AI-assisted Coding
26mAl and Software: The Good, The Bad, The Scary
43mBecome a 10x developer by harnessing the power of Generative AI
13mMeet the Speakers
40mBuilding the Future Developer: Trends in Open Source, Education...
45mThe AI Revolution in Software Development: From Code Writing to Quality Assurance