Q&A #1
summary
Six-step design workflow becomes concrete: prompts break problems into substeps and can spawn autonomous AI agents that book holidays or assemble multi-step solutions; Llama 2 small/medium models and RAG reduce compute and privacy exposure versus monolithic LLMs; recommendation systems can use user-intent prompts to predict demand and generate posters or briefs for creators; mitigation relies on mitigation patterns plus human-in-the-loop, anonymization, fine-tuning in-house models for sensitive domains like mental health; several prototype designs (AI-generated packable sofa) and rapid university adoption are already working.
AI-written summary · about
chapters (6)
- 0:00Introduction and first question
- 1:48Future of prompt engineering
- 8:13Hallucinations and mitigation
- 15:15AI adoption in architecture
- 18:45Career and energy concerns
- 21:54Wrap-up and closing
related talks
01h 00mPanel Discussion: AI Meets Design: A Dialogue
48mMastering the AI Shift: Adaptive Strategies for Software Evolution
27mThe Shape of AI : How designers integrate AI into their workflow
25mHuman-Centered AI: What I've Learned From Building Two LLM-Powered Products Thi...
01h 39mMastering LLMs: From Prompt Engineering to Agentic AI
44mFuture-Proof Your...
26mFour Steps to Embed AI Agents into Workflow
45mPanel Talk