Lessons learned from using generative AI dev tools
summary
12 months of hands-on use shows generative AI speeds experimentation, lowers automation friction and cuts small project build time from days to hours — examples include a live five‑minute Chrome extension prototype, a 2.5‑hour reunion guessing game, and a GitHub-views ETL built in two hours. Tools used include Amazon Q for code + SQL generation and cursor as a daily IDE; offline local models run slowly on M1. Key failures: insecure or fabricated code snippets require trust‑and‑verify; model knowledge cutoffs and rapid library churn cause version hallucinations.
AI-written summary · about
chapters (6)
- 0:00Generative AI dev tools intro
- 2:12Loves and lessons learned
- 10:06Expectations vs reality
- 17:29Five success setups
- 24:15AI-assisted development future
- 28:31Q&A
related talks
39mBuild without limits: lessons learned from using generative AI dev tool
21mRevolutionizing software creation: Harnessing the power of GenAI
29mImpact of #GenerativeAI in Development and #SoftwareTesting
42mBuild Without Limits - Lessons Learned Working with Data and AI Coding Assistants
21mOpportunities of Generative AI
41mWeb Development in the Times of AI
36mUnleashing Generative AI: Improving Developer Productivity in SRE
29mGenerative AI: Path to Value in practice