Presentation: Designing Fast, Delightful UX With LLMs for Mobile Frontends
Balakrishnan Ramdoss discusses how to architect production-grade, AI-powered conversational apps at scale. He explains how to overcome model latency, leverage server-driven UI and Backend-for-Frontend patterns to dynamically render multi-modal interfaces, optimize prompts for UI selection, and integrate low-latency, privacy-first on-device AI for mobile applications. By Balakrishnan Ramdoss
Title: Designing Fast, Delightful UX With LLMs for Mobile Frontends
Bala Ramdoss, a Senior Software Engineer at Amazon, presents on architecting production-grade, AI-powered conversational apps at scale. He discusses overcoming model latency, server-driven UI, and Backend-for-Frontend patterns to dynamically render multi-modal interfaces. Additionally, he covers optimizing prompts for UI selection and integrating low-latency, privacy-first on-device AI for mobile applications.
Bala highlights that 89% of top 100 apps have AI, and many have explicit chatbots. While plug-and-play chatbot-as-a-service solutions exist, they don't integrate well with mobile apps due to deep integration requirements. Bala shares his experience building Amazon Lens, an AI-powered feature for finding products using a camera or image on iOS and Android. Lens Live, released to the public, allows live scanning and follow-up questions on scanned items.
Bala explains that AI conversational experiences have evolved from search systems, where ChatGPT initially only answered questions. Most of the talk focuses on the frontend and how it plays a crucial role in building AI-powered conversational apps at scale.
Written by urgent.news from InfoQ's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
