At I/O '25, we announced new Gen AI tech for Android devices. This session will explore how these technologies work and how you can leverage them in your apps. I will also detail the key updates released last month (Aug). Introduction: Why Gen AI in Your Android Apps? I. Gemini Nano: The fastest On-Device Gen AI model on Android - Prompt Engineering 101 for Android Developers - How to Write Mobile-friendly Prompts - Using the ML Kit GenAI API for Specific Use Cases - The Future is Now: The Evolution of Gemini Nano II. Gemma 3n: Leveraging Open Models - Why Choose Gemma? - SDKs for Gemma-Based Gen AI - ICL & LoRA: Fine-Tuning Strategies for Your App's Needs III. Last, but definitely not least - Traditional Machine Learning in 2025 - H/W Acceleration with LiteRT Next - Play Delivery: Distribution Strategies for Large-Scale Models
Sa-ryong Kang Developer Relations Engineer @ Google
- Those who have been devoted to Android development…but feeling a bit restless with the recent surge in generative AI - Those about to give up because Google has too many AI solutions (Gemini Nano, Gemma Nano, TensorFlow Lite / LiteRT, MLKit / MLKit Gen API, MediaPipe solutions / MediaPipe Inference API, Google AI Edge SDKs, etc.) - Those considering implementing AI-powered features - Those who have at least a basic understanding of Kotlin syntax
Beyond Mobile: Building Android Auto and Android TV Applications with React Native
Aashima Wadhwa
#Cross-platform Development
No More Getting Lost! A Practical Guide to Writing Jetpack Compose
b4tchkn
#Jetpack Compose
Worry free TV App development for users and developers - a focus on Compose internal implementation.
taked137
#Jetpack Compose
© 2014- DroidKaigi Committee