TokenBar: A macOS Usage Tracker for Claude Code and OpenAI Codex
TokenBar is a lightweight macOS menu bar app that helps you track Claude Code and OpenAI Codex rate limits, quotas, and token usage locally, with no telemetry.
Read article
Lead Mobile Engineer
Building intelligent mobile apps for the edge.
I bring over 7 years of mobile development experience across Android, Flutter, and React Native. Today, my focus is advancing those platforms with on-device AI to build smart, private, and highly responsive apps.
Expertise
At a glance & core stack
At a glance
0+
Years in mobile engineering
0
Core platforms
0+
Domains in production
Core stack
Mobile Platforms
AndroidFlutterReact NativeKotlinJetpack ComposeJavaDart
On-Device AI
Edge AIMediaPipeLlama.cppExecuTorchML KitTensorFlow LiteOpenCV
Architecture
Clean ArchitectureMVVMBLoCRedux
IoT & Real-time
IoTBLEMQTTSocket.ioWebRTCArduino
Experience
Work history
Leading Android development across connected-device streaming, Edge AI, and embedded systems.
Android TV Transmitter App
Audio Streaming Platform
- Built a scalable transmitter app to manage and synchronize wireless speakers across devices
- Architected Socket.io–based real-time communication and optimized multi-threading for low-latency audio control
- Developed JNI integrations and modular MVVM architecture for high-performance native interactions
- Refactored legacy components to improve stability, scalability, and maintainability
Companion Mobile App
Multi-Device Control
- Developed a mobile app to manage multiple transmitters, speakers, and audio configurations across networks
- Enabled centralized control for audio enhancements and real-time device orchestration
- Ensured responsive UI and reliable synchronization across distributed devices
Smart IoT App with Edge AI
- Built on-device AI capabilities using MediaPipe, Llama.cpp, and ExecuTorch inference engines
- Designed fully offline AI pipelines with zero cloud dependency, ensuring privacy and low latency
- Optimized model performance using quantization and efficient runtime strategies
Embedded Wearable Device
i.MX 8M Mini
- Developed an Android-based application for a wearable device to monitor health sensors and track user health data
- Integrated on-device AI insights for real-time health analysis and feedback
- Optimized performance for embedded constraints including limited compute, memory, and power
Tech Stack
KotlinAndroidAndroid TVEmbedded AndroidJetpack ComposeMVVMSocket.ioJNIMediaPipeLlama.cppExecuTorch
Blog
Latest thoughts
Agent Workflows for Vibe Coding: Idea to Product Guide
Learn how agent workflows, rules, context files, and review loops turn AI vibe coding into a reliable path from idea to product.
Read article
Google I/O 2026: Entering the 'Agentic Gemini Era' of Autonomous Systems
An in-depth developer review of Google I/O 2026, breaking down the Agentic Gemini Era, Gemini 3.5 Flash, TPU 8t/8i, Antigravity 2.0, Android XR, and the shift to autonomous workflows.
Read article
Android’s New Era of Intelligence: What Developers Need to Know from The Android Show (I/O Edition 2026)
Discover how Gemini Intelligence, AppFunctions, adaptive apps, Android Auto, XR, and RemoteCompose are reshaping the Android developer experience in the I/O Edition 2026.
Read article
Webshotio Screenshot API: Turn Any Public URL into a PNG in Seconds
Webshotio is a simple screenshot API for developers: pass a URL, get a PNG. Try the live GitHub Pages frontend or run it locally in minutes.
Read article
Running On-Device AI Models on Android: MediaPipe, Llama.cpp, or ExecuTorch?
Explore the best ways to run on-device LLMs and AI models on Android. I compare MediaPipe, Llama.cpp, and ExecuTorch to help you choose the right backend for offline AI Android edge computing.
Read article
Contact
Let's connect
Open to new opportunities, collaborations, or just a friendly chat about mobile development and AI.
meet30997@gmail.com