I work on mobile apps and tools for software development, including AI agents and automation. This page also includes my earlier biomedical research and a local-LLM study.
I tested extraction and interpretation across local models, then benchmarked speed, memory and energy use on Android and iOS.
I joined Siemens Healthineers as the sole Flutter developer on a new mobile platform, established its application and delivery foundations, and helped it grow into three public apps now built by a full Scrum team. My contributions span architecture, native interoperability, accessibility and delivery.
Public applications · Google Play
Projects with substantial research, engineering or real-world validation behind them.
A Flutter health-app prototype that brings Android Health Connect records into observation and patient screens. It explores direct native access through JNI and generated bindings, mapping records into app-friendly and FHIR-oriented Dart models.
Phone + watchA private sensor recorder for Android and Wear OS. Capture raw samples and annotations as local CSV files, record independently on the watch, and sync measurements back to the phone.
Apps I build for myself, to try an idea or simply because I feel like making something.
New achievement: just one more chapter.
My offline audiobook player for Android and iOS. A home for your own books, with chapter navigation, listening history and text or voice notes.
For the Dungeon Crawler Carl kind of listening session. The sleep timer is there for when the next floor can wait.
A little more space. A little less distraction.
My minimal Android launcher. Keep a short list of apps on Home and let schedules change it by day and time. Everything else stays one swipe away.
Built with Kotlin and Jetpack Compose, with local settings and no account or internet permission.
Playing with Bluetooth and privacy.
An Android experiment exploring Bluetooth and privacy. It warns when Bluetooth is enabled without an active connected device—a small reminder to switch it off when it isn’t needed.
Focused libraries for biomedical signals and model interpretation.
Online and offline implementations of widely used ECG R-wave detectors for Kotlin applications.
Portable C++ implementations of the same detector family for native and embedded integrations.
Occlusion, saliency and Grad-CAM interpretation tools for one-dimensional TensorFlow models.
My current AI work focuses on developer tools, agent instructions and automation. My GitHub also includes mobile utilities, RAG examples and code from earlier research.