Building real-time user insights into your design workflow
summary
Claims a concrete system: a JavaScript-triggered in-app AI interviewer that records screens, asks follow-ups, and pushes video feedback into Marvin for instant analysis and sharing to Figma, Slack, Teams. It supports 40 languages, runs at scale so thousands of live intercepts collect moments-of-use insight, and routes clips into workflows immediately. The built product—branded Marvin Live Intercept—ships as a platform update, demonstrably working in real demos and handling real user speech, aiming for a one researcher per user experience.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:57User feedback challenge
- 6:54In-app AI interviewer
- 8:19Real-time AI interviewer
- 10:52Traditional feedback failures
- 12:57AI interviewer inspiration
- 14:44Closing insight-decision gap
- 16:29Human-centric AI design
- 17:10Emotional impact of feedback
- 18:35One researcher per user
related talks
20mHow fast feedback enables smart design
16mBetter research, better design
01h 40mIntegrating AI into your Design workflow
27mPixels and prompts: building products in the AI era
20mInteractive Insights : Using Figma for UX research
43mPanel Discussion: Unraveling AI & Design
17mDesign systems in an AI first ecosystem
31mLive design critique with the Figma design team