🔥 Generative UI: The Future of Frontend (And Why You Should Care)
summary
Argues that real-time generative UI works by using LLMs only for intent classification while keeping private data inside micro‑frontends, shows a Payoneer‑style demo where the model returns strict JSON contracts and a frontend updates components, uses JSON schema via zod to describe intents and fields, recommends starting with large LLMs then fine‑tuning smaller ones, enforces 80% confidence fallback and reports ChatGPT drove a 1M‑user spike in five days.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:49Speaker background
- 2:59OpenAI growth impact
- 5:29Natural language UI
- 8:20AI intent-based search example
- 10:08Fintech intent challenges
- 11:55Intent-oriented micro frontends
- 13:43Fine-tuning LLMs
- 14:27Live demo intent UI
- 19:27Micro frontend implementation
- 25:12JSON schema for intent mapping
- 27:43Form population with AI
- 29:30Training data for intent variation
- 30:33Q&A
related talks
12mNavigating the World of AI Development Without Going Insane
38mGenerative UIs: Using LLMs to Create Flutter Screens at Runtime
19mGenerative AI for Front-end Developers
17mBeyond Components: Designing Generative UI for MCP Apps
19mGenerative UI: Bring your React Components to AI today!
29mImpact of #GenerativeAI in Development and #SoftwareTesting
32mGenerative UI in Flutter
33mIs Frontend Still a Thing? A UI Reset in the Age of AI Disruption