Production-ready containerized stack for backtesting and live trading quantitative strategies.

Architecture

Remote (You) → [VPN:10.8.0.1] → Internet → [Firewall:Port 1194] → Docker Host:192.168.1.1
│ ┌───────────┴───────────┐

│ Docker Bridge │

│ 172.18.0.0/16 │

│ │_______________________________

│ [Jupyter:8888] ←→ [Prometheus:9090] → [Grafana:3000] │

│ ↓ ↓ ↑ │

│ [MLflow:5000] ←→ [Redis:6379] [Loki:3100] │

│ ↓ ↓ │

│ [MinIO:9000] ←→ [TimescaleDB:5432] │

└──────────────────────────────────────────────────────┘

• JupyterLab: Strategy development (port 8888)
• TimescaleDB: Market data storage (port 5432)
• MLflow: Model tracking (port 5000)
• MinIO: Artifact storage (port 9000)
• Redis: Signal cache (port 6379)
• Prometheus + Grafana + Loki: Observability (ports 9090, 3000, 3100)

Quick Start

# 1. Clone
git clone https://github.com/raynayam/quant-devops-stack
cd algorithmic-trading-platform

# 2. Configure
cp .env.example .env
# Edit .env with your credentials

# 3. Start
docker-compose up -d