Lessons Learned in AI Engineering
summary
AI engineering becomes a distinct role as LLMs dominate software and testing, with concrete QA problems: models favor popular stacks so AI-generated code often defaults to React + Tailwind bias, it injects harmless-looking side effects (extra variables, dead lines) that need code review and targeted regression tests, and context windows cause memory/context failures so conversation-length tests must be designed for nondeterminism. Use desktop IDE/agent setups and smoke tests rather than copy‑paste web chats — IDE + agent workflows speed work but require extra verification.
AI-written summary · about
related talks
32mQuality engineering in the AI age: From testing AI to leveraging GenAI
07mAI's impact on programming jobs and developer skills
41mThe Changing Face of Engineering What Teams Need Now 002
14mFull-Stack AI Engineering from AI-Native Coding to Knowledge, Models, Infrastructure & OSS
50mtl;dr: Chill, y’all – AI will not devour SE
30mAgentic QA vs LLM Dev: A Live AI Showdown in the DevOps Arena
37mHow AI & automation are changing software engineering
41m#define AI Engineer