Compose Locally, Think Globally: On-Device LLM Apps with Kotlin, Compose Multiplatform & llama.cpp

Описание: In this technical workshop, we'll discuss how to leverage the power of Kotlin and Compose Multiplatform alongside llama.cpp to build performant, privacy-focused, and responsive on-device Large Language Model (LLM) applications. We'll dive into the full cycle -- from setting up automated native library compilation and seamless Kotlin Native interop, to crafting intuitive cross-platform user interfaces with Compose. Through practical examples and live demonstrations, we'll explore how to efficiently deliver locally-run LLM experiences on Android, iOS, and Desktop, empowering you to create powerful AI-driven apps that work offline and respect user privacy.

Speaker: Dmytro Minochkin