This post explores building intelligent Android apps using Firebase AI Logic for cloud-hosted and hybrid AI features. It details how to ground AI answers in real-world context using URL, Google Search, and Maps grounding. The article also explains hybrid inference, allowing dynamic routing between cloud and local execution for features like restaurant review drafting. A museum assistant example demonstrates using web context and search for up-to-date information. The hotel support chat feature showcases custom hybrid routing for translations, utilizing ML Kit for language identification and both on-device and cloud models for translation. Firebase App Check is introduced to secure AI pipelines against API key abuse and unauthorized billing. The combination of cloud and on-device capabilities results in a smart, secure, and offline-available travel app. The Jetpacker app's full source code is available on GitHub. This is the third part of a series on building intelligent Android apps.