Build intelligent experiences: Deep dive into AppFunctions
summary
Shows how to expose app capabilities to the OS using the AppFunctions Jetpack API so agents can call annotated suspend functions, register schemas, and act like clients in an Android MCP flow. Demo builds on-device receipt parsing with Gemini Nano (already on ~140 million devices) to extract title, amount, category and it works live. Covers testing via ADB and a development agent, toggling functions at runtime, and upcoming alpha 10 changes (service entry point, experimental access controls).
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:54AI on Android
- 2:18Implement in-app intelligence
- 6:45Integrate with the intelligence system
- 10:08Task automation
- 12:18AppFunctions
- 14:47Android MCP powered by AppFunctions & Demo
- 19:25Deep dive into AppFunctions
- 23:12New features in alpha 10 & Experimental status
- 25:05Testing AppFunctions
- 27:14Experiment with AppFunctions today
related talks
28mBuild intelligent experiences: Deep dive into AppFunctions
24mAppFunctions: Making Your Android App Agent-Ready
36mElevating OS Experiences with AI-Driven Velocity
41mAgent skills for Android app development
19mProtecting Your Users in an Agentic World
37mDevelop faster with Android Studio, AI and skills
02h 58mDeveloper keynote deep dive
21mAdd Generative AI capabilities to your Android app