Creating fluent Frontend Experiences in an age of AI
summary
Demonstrates turning blocking AI calls that caused ~15s waits into responsive UIs by streaming partial responses and switching from plain text to NDJSON newline‑delimited JSON, then flushing one valid object per chunk so the frontend can incrementally parse and render lists (examples: streaming 20 items). Implements streaming with OpenAI chat completions (stream=true), shows a Movies++ demo, uses Vercel AI SDK to stream component JSON, and builds an MCP-backed assistant with Langflow, Docker and Composio tools to read calendars and add events.
AI-written summary · about
chapters (7)
- 0:00AI-driven frontend intro
- 3:11Blocking vs streaming AI responses
- 5:42Streaming JSON for frontend
- 11:23DOM components as JSON
- 15:21Redefining frontend with AI chat
- 17:47Model Context Protocol evolution
- 27:19Future of conversational UX
related talks
38mUI Engineering for the AI Age
44mUsing AI Effectively in 2025
14mStreaming UI and Scaling Full-Stack AI Like a Boss to Millions with Next.js
24mFull Stack Web Development with AI
31mAdaptive UIs with Vue and AI: What If Your UI Could Build Itself?
17mReact Server Components in AI Applications
06h 28mDay 5
27mAI-Generated React Server Components