Skip to content

README.md

Latest commit

 

History

History
232 lines (166 loc) · 6.89 KB

README.md

File metadata and controls

232 lines (166 loc) · 6.89 KB

TellMeMo

Turn project chaos into clarity. AI-powered platform that transforms meetings into searchable knowledge—upload content, ask questions, get instant answers.

✨ Features

  • 🎯 Real-Time Meeting Intelligence - Live question detection with four-tier answer discovery (RAG → Meeting Context → Live Monitoring → AI-Generated)
  • 🤖 Automatic Action Tracking - Extract tasks, owners, and deadlines as they're mentioned
  • 🔍 Ask questions - Search in plain English, get answers with sources
  • ✅ Track actions - AI extracts tasks automatically
  • ⚠️ Detect risks - Spot problems before they become critical
  • 📝 Generate summaries - Executive, technical, or stakeholder formats
  • 📊 Portfolio view - Organize by Portfolio → Program → Project
  • 🔓 100% Open Source - Self-host or use our cloud

🆕 Recent Updates

See CHANGELOG.md for latest features, improvements, and bug fixes.

🏗️ Tech Stack

Core Services

  • Frontend: Flutter Web (Nginx-served)
  • Backend: FastAPI (async Python)
  • Database: PostgreSQL 18.0 (metadata & relational data)
  • Vector DB: Qdrant v1.15.5 (semantic search & embeddings)
  • Queue: Redis 8.2 + RQ (background job processing)
  • Workers: RQ Worker (2 workers for high/default/low priority queues)

AI & ML

  • Primary LLM: Claude 3.5 Haiku (Anthropic)
  • Fallback LLM: GPT-4o/GPT-4o-mini (OpenAI) - automatic fallback on Claude overload
  • Real-Time LLM: GPT-5-mini (OpenAI) - streaming intelligence for live meetings
  • Real-Time STT: AssemblyAI - speech-to-text with speaker diarization
  • Embeddings: Google EmbeddingGemma-300m (local, 768-dimensional)
  • Circuit Breaker: Purgatory library for production resilience

Monitoring & Observability (Optional)

  • RQ Dashboard: Web UI for queue monitoring (port 9181)
  • Langfuse: LLM observability & cost tracking (optional)
  • ELK Stack: Elasticsearch, Logstash, Kibana for centralized logging (optional)

Deployment

  • Containerization: Docker Compose (7 services: frontend, backend, postgres, qdrant, redis, rq-worker, rq-dashboard)
  • Authentication: Built-in JWT auth (backend) or Supabase (configurable)

🚀 Quick Start

Prerequisites

  • Docker ≥20.10
  • Docker Compose ≥2.0

Everything else runs in containers.

Installation

# 1. Clone the repository
git clone https://github.com/Tell-Me-Mo/tellmemo-app.git
cd tellmemo-app

# 2. Create .env file with your API keys
cat > .env << EOF
ANTHROPIC_API_KEY=sk-ant-api03-your-key-here
HF_TOKEN=your-huggingface-token-here
EOF

# 3. Start all services
docker compose up -d

# 4. Access the application
# Open http://localhost:8100 in your browser

Getting API Keys

ANTHROPIC_API_KEY (Required): Get from console.anthropic.com → API Keys

HF_TOKEN (Required): Get from huggingface.co/settings/tokens → Create token (Read permission)

OPENAI_API_KEY (Optional but Recommended): Get from platform.openai.com → For automatic fallback when Claude is overloaded + powers GPT-5-mini for real-time meeting intelligence

ASSEMBLYAI_API_KEY (Optional): Get from assemblyai.com → For real-time meeting transcription with speaker diarization

Useful Commands

# View running containers
docker compose ps

# View logs
docker compose logs -f

# Stop all services
docker compose down

# Restart services
docker compose restart

# Update to latest version
git pull
docker compose pull
docker compose up -d

📖 Documentation

🛠️ Development

For development and customization:

# Clone repository
git clone https://github.com/Tell-Me-Mo/tellmemo-app.git
cd tellmemo-app

# Set up environment
cp .env.example .env
# Edit .env with your configuration

# Start services
docker compose up -d

# For frontend development (Flutter)
cd frontend
flutter pub get
flutter run -d chrome --web-port 8100

# For backend development (Python)
cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

Project Structure

tellmemo-app/
├── frontend/          # Flutter web application
├── backend/           # FastAPI backend
├── docs/              # Documentation website
├── docker-compose.yml # Docker services configuration
├── .env.example       # Environment variables template
└── README.md          # This file

🔧 Configuration

Key environment variables:

# Required
ANTHROPIC_API_KEY=sk-ant-api03-xxx
HF_TOKEN=hf_xxx

# Optional (High Availability)
OPENAI_API_KEY=sk-proj-xxx           # Automatic fallback on Claude overload
ENABLE_LLM_FALLBACK=true             # Enable provider fallback (default: true)

# Optional (Circuit Breaker - Production Resilience)
ENABLE_CIRCUIT_BREAKER=true          # Enable circuit breaker pattern (default: true)
CIRCUIT_BREAKER_FAILURE_THRESHOLD=5  # Failures before circuit opens (default: 5)
CIRCUIT_BREAKER_TIMEOUT_SECONDS=300  # Circuit open duration in seconds (default: 300 = 5 min)

# Optional (Other)
JWT_SECRET=your-secret-here
SENTRY_ENABLED=false

See .env.example for all options.

📚 More Info

🤝 Contributing

We follow the Feature Branch Model:

  1. Fork the repo
  2. Start from main: git checkout main && git pull origin main
  3. Create feature branch: git checkout -b feature/amazing-feature
  4. Make changes and run tests
  5. Commit: git commit -m 'feat: add amazing feature'
  6. Push and open a Pull Request to main

See CONTRIBUTING.md for full workflow and guidelines.

🐛 Troubleshooting

Services won't start?

docker compose down
docker compose up -d

Check logs:

docker compose logs -f

See full troubleshooting guide.

📄 License

MIT License - see LICENSE file.

🔒 Security

Found a vulnerability? See SECURITY.md for reporting.

🌐 Links

Built with ❤️ using Flutter, FastAPI, and AI