π Android & KMP Engineer | π§ On-device AI | β‘ Performance-focused Developer
I build high-performance Android & Multiplatform apps focused on:
- β‘ Smooth, responsive UI with Jetpack Compose
- π§ On-device AI (no cloud dependency)
- π Clean, scalable architecture
Currently exploring how to push fully offline, privacy-first applications using
Kotlin Multiplatform + Local LLMs.
On-device image classification using KMP + ONNX
- β‘ Runs fully offline (no API calls)
- π§ ONNX Runtime for local ML inference
- π Shared business logic with Kotlin Multiplatform
- π± Designed for Android + Desktop scalability
π§ In Progress
Privacy-first interface for local LLMs
- π§ Run LLMs locally with real-time streaming
- β‘ Smooth UI with Compose (Desktop + Android)
- π Fully offline AI interaction
- π Clean architecture + reactive state handling
π https://github.com/dontknow492/ollama-gui
Offline-first finance tracking app
- π¦ Room DB + offline-first architecture
- π Data insights & analytics
- β‘ Fast, responsive Compose UI
- π§© Clean Architecture + MVVM
π https://github.com/dontknow492/expensetracker
High-performance Android music player
- π§ Built using Media3 APIs
- β‘ Smooth playback & queue management
- π Clean architecture implementation
π https://github.com/dontknow492/musify
Core
- Kotlin, Jetpack Compose, Coroutines, Flow
Multiplatform
- Kotlin Multiplatform (KMP)
Data & Backend
- Room, DataStore, Firebase, REST APIs
AI / ML
- ONNX Runtime, Ollama, Local LLMs
Tools
- Git, GitHub, Gradle, GitHub Actions
- β‘ Compose performance (recomposition, state optimization)
- π§ Running LLMs locally on mobile (latency + memory tuning)
- π Scaling real-world KMP architectures
- Android / Kotlin Multiplatform roles (Intern / Junior)
- Projects involving on-device AI or performance-critical apps
- GitHub: https://github.com/dontknow492
- LinkedIn: https://www.linkedin.com/in/manish-bhatt-181363390/
β If you find my work interesting, consider starring a repo!