mAI Coach

Real-time motion capture & full-featured workout tracker.

mAI Coach brings advanced computer vision to the weight room. By running Google's MediaPipe pose detection directly on-device, it provides instant form feedback—no cloud uploads, no latency. The app also includes a complete workout tracker with templates, supersets, personal records, scheduling, and more.

Available on iOS and Android, with a companion desktop research toolset.

Key Features

Feature	Description
Privacy-First AI	All pose estimation runs on-device via MediaPipe. Video never leaves your phone.
Live Form Coaching	Real-time skeletal overlay with audio cues for grip, tilt, lockout, and depth.
Workout Tracking	Custom workouts, templates, exercise reordering, superset groups, and notes.
Personal Records	Auto-detected PRs with trophy badges, confetti, and a dedicated PR board.
Scheduling	Recurring weekly schedules with calendar integration and automatic reminders.
Rest Timer	Auto-start countdown after set completion with configurable duration.
Interactive Tutorial	24-step spotlight walkthrough that guides new users through every feature.
In-App Help Guide	Comprehensive PDF user guide accessible from Settings.
Research Tooling	Desktop app for labeling video, tuning poses, and training classifiers.

Repository Structure

Path	Description
`ios/`	iOS app — SwiftUI, MediaPipe inference, CoreData.
`android/`	Android app — Jetpack Compose, Room, Hilt.
`tools/`	Desktop research toolkit — PySide6/Qt labeler, trainer, tuner.
`backend/`	FastAPI auth server (Firebase-backed).
`shared/`	Cross-platform resources: exercise library, audio clips, ML models, user guide.
`docs/`	Technical design briefs and architecture documents.
`tests/`	Automated test suite for tools.
`temp/`	Scratch files (gitignored).
`archive/`	Archived files.

Quick Start

📱 iOS App

Requires Mac, Xcode 16+, iOS 17+.

Navigate to ios/.
Open mAICoach.xcodeproj and hit Run on a physical device.

👉 Full iOS instructions

🤖 Android App

Requires Android Studio (Hedgehog+), SDK 26+.

Open android/ in Android Studio.
Sync Gradle and run on a device or emulator.

👉 Full Android instructions

🖥️ Developer Tools

Requires Python 3.10+ and ffmpeg.

Run ./scripts/setup_unified_env.sh.
Run ./scripts/run_unified_tool.sh.

👉 Full tools instructions

🚀 Deployment

Deploy to TestFlight (iOS) and Google Play Internal Testing (Android) with one command using Fastlane:

fastlane deploy_all      # Deploy both platforms
fastlane deploy_ios      # iOS → TestFlight only
fastlane deploy_android  # Android → Play Internal Testing only

Credentials are stored in fastlane/ (gitignored). See fastlane/.env for config paths.

Maintainers

Travis Whitney (whitnetr@oregonstate.edu)
Cole Seifert (seiferco@oregonstate.edu)

License

Proprietary. See LICENSE.md.

TWhit229/mAI_Coach