ANDROID, RESEARCH, PLATFORM, ARCHITECTURE

Modular Data Tracking Application Platform

Research Period: 2025.03 - Present

Overview

This project is a plugin-based, configuration-driven mobile tracking platform for research studies. It lets researchers compose and run study protocols through JSON (including schedules, instruments, and UI blocks), while developers focus on building a stable runtime and reusable modules.

Why This Project

Most research platforms lean toward “survey-first” systems with fixed sensing pipelines. This platform treats surveys, sensors, health data, routines, notifications, and visualization as equal protocol components, making it closer to a Study Runtime / Protocol Engine than a single survey app.

Problem

  • Each study often requires repeated custom development and brittle one-off app changes.
  • Protocol changes are hard to track and reproduce, weakening methodological clarity.
  • Mobile OS constraints make reliable collection and scheduling difficult.
  • Researchers need operational support such as monitoring coverage and exporting data.

Goals

  • Enable protocol composition through configuration rather than rebuilding the app per study.
  • Separate a stable core runtime from flexible plugins.
  • Support multi-source collection and study-specific scheduling.
  • Treat study configuration as a versionable artifact to improve reproducibility.

System Design

  • Config-driven execution (JSON): configuration controls plugin composition, schedules, surveys/routines, and SDUI-style UI blocks.
  • Plugin-based architecture: modules can include collection logic and (depending on the plugin) UI components.
  • Local-first storage: data is persisted locally with Room, then synced to Firebase as the server layer.
  • Platform-level scheduling support: baseline scheduling infrastructure is provided so plugins can reuse reliability-sensitive patterns.

What I Built / Contributed

  • Designed the overall platform architecture (core runtime vs plugins).
  • Implemented JSON-driven protocol execution, including schedules and UI blocks.
  • Built multi-source collection foundations (survey, sensors, health integrations, routines, notifications, visualization).
  • Implemented local-first persistence with Room and Firebase-backed syncing.
  • Started researcher tooling to reduce friction from editing JSON directly.

Key Features

  • Plugin-based module composition for study protocols
  • JSON configuration bundles for instruments, schedules, and UI blocks
  • Offline-first data capture (Room) with Firebase sync
  • Reproducible, versionable protocol configuration as a first-class artifact
  • Researcher tool in progress: drag-and-drop protocol builder with live preview

Tech and Keywords

Android, Kotlin, Jetpack Compose, Modular Architecture, Plugin System, Protocol Engine / Study Runtime, JSON-driven Configuration + SDUI, Room, Firebase, Offline-first Data Collection, Research Operations Tooling

Outcome

A reusable study protocol engine that reduces repeated study-specific development by turning protocol customization into configuration and composable plugins. It enables researchers to act as protocol designers without programming, while keeping developers focused on building a stable, extensible runtime.