In this session, I’ll show how Android developers can build real-time AI experiences by combining Gemini-powered intelligence with Agora Conversational AI and a native Android app built using Kotlin and Jetpack Compose.
We’ll go beyond a traditional chatbot and look at how to design voice-first Android experiences where users can speak naturally with an AI agent, receive low-latency responses, and see real-time conversation states such as listening, thinking, speaking, reconnecting, and transcript updates.
The session will cover Android-specific implementation patterns including microphone permissions, audio lifecycle, Compose state management, Kotlin Flow-based event handling, foreground behavior, and designing a responsive UI for real-time AI interactions. I’ll also demo how Gemini can act as the reasoning layer while Agora handles real-time conversational infrastructure, giving Android developers a practical blueprint for building the next generation of AI-powered mobile apps.