CoachFlow is a premium, minimalist mobile application that democratizes access to personal coaching. By combining Advanced AI (Llama 3.1 405B) with a frictionless user experience, it allows anyone to create, customize, and converse with expert personas tailored to their specific goals.
"Productivity = (Guidance × Context) / Friction"
- Custom Coaches: Create unique AI personas with specific expertise (e.g., Marcus Aurelius for Stoicism, Steve Jobs for Product Design).
- Real-time Preview: See your coach come to life as you design them.
- System Prompts: Define the exact personality and behavioral constraints of your AI.
- Universal Context: Define your goals, values, and constraints once.
- Adaptive Advice: All coaches automatically understand who you are (e.g., "I know you value sustainability, so here is an eco-friendly solution...").
- Llama 3.1 405B Integration: Powered by Bytez API for reasoning capabilities that rival top-tier proprietary models.
- Voice-to-Voice (TTS): Listen to your coaches with natural, neural voices.
- Rich Text Support: Markdown rendering for structured advice (lists, code blocks, bold text).
- Share Wisdom: Export entire conversation threads to friends using a unique link.
- Network Management: Build your contact list and share insights directly within the app.
- Biometric Login: Secure access via FaceID/TouchID.
- Offline-First: Powered by Hive for instant load times and offline history access.
- Supabase Backend: Robust real-time syncing and Row-Level Security (RLS) for data protection.
CoachFlow is built with a focus on Clean Architecture, Scalability, and Performance.
| Layer | Technology | Description |
|---|---|---|
| Framework | Flutter (Dart) | Cross-platform high-performance UI. |
| State | Riverpod | Compile-time safe dependency injection and caching. |
| Navigation | GoRouter | Declarative routing with deep linking support. |
| Backend | Supabase | Auth, Database (PostgreSQL), and Real-time subscriptions. |
| AI Data | Bytez API | Access to Open Source models (Llama 3.1, Gemma, Mistral). |
| Local DB | Hive | NoSQL local storage for |
| UI/UX | Flutter Animate | fluid, engagement-driven micro-interactions. |
- Flutter SDK (latest stable)
- Supabase Account & Project
-
Clone the repository:
git clone https://github.com/Donchaminade/CoachFlow.git cd CoachFlow -
Install dependencies:
flutter pub get
-
Environment Setup: Copy
.env.exampleto.envand fill in your keys, or updatelib/core/config/supabase_config.dart:cp .env.example .env
-
Run the app:
flutter run
- MVP: Core Chat, Local Storage, Mock AI.
- V1.0: Real AI (Bytez), Biometrics, User Context, Supabase Sync.
- V1.1: Direct Share (in-app contact sharing), message deletion, clear history, settings links, i18n polish.
- V1.2: Voice Mode (Full duplex conversation).
- V2.0: Coach Marketplace (Community extensions).
This project is being submitted to the GitHub Finish-Up-A-Thon — reviving an abandoned side project with a clear before/after story and GitHub Copilot assistance.
| Resource | Link |
|---|---|
| Submission article (draft) | docs/FINISH_UP_A_THON_SUBMISSION.md |
| Before snapshot (git) | git checkout 941d27b |
| V1.1 release tag | v1.1.0-finishathon (create after final push) |
Quick demo path: Create coach → Chat → Share to contact → View in Shared tab → Continue conversation.
Contributions are welcome! Please read our CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.
Built with ❤️ by the CoachFlow Team.