DSTA Docs

Dr. Strange Trading Analysis — Platform documentation, powered by MkDocs Material.

📖 Live site: https://minhdqdev-org.github.io/dsta-docs

---

What's Inside

Section	Content
Getting Started	Quickstart, dev setup, configuration, code quality, testing
Architecture	System overview, microservices design, database schema & models
Services	Per-service deployment runbooks
Trading Engine	Strategy development, risk management, RL trading, regime detection
Data & Analytics	Market data, backtesting (guide, architecture, best practices), sensitivity & capacity tools
Machine Learning	Feature engineering, price prediction, RAG, quant libraries
Infrastructure	Docker, CI/CD, secrets, security scanning, Prometheus, Grafana, Kubernetes
API Reference	OpenAPI 3.0 specs for all four FastAPI services
Reference	Changelog, troubleshooting, PRD, resources

---

Local Development

Prerequisites: mise + uv

# Install dependencies
mise run install

# Start hot-reload dev server at http://localhost:8080
mise run up

Task	Description
mise run up	Start docs server (hot-reload, port 8080)
mise run down	Stop the docs server
mise run build	Build static site → site/
mise run clean	Remove generated site/
mise run lint	Strict build check (fails on warnings)
mise run deploy	Push to GitHub Pages (gh-pages branch)
mise run version	Publish a versioned release via mike
mise run version-list	List all published versions
mise run check	Full quality gate

---

Project Layout

dsta-docs/
├── docs/
│   ├── index.md                  # Home page
│   ├── getting-started/          # Setup & development guides
│   ├── architecture/             # System & database design
│   ├── services/                 # Per-service runbooks
│   ├── trading/                  # Trading engine & strategy docs
│   ├── data/                     # Market data & backtesting
│   ├── ml/                       # Machine learning & predictions
│   ├── infrastructure/           # DevOps, CI/CD, monitoring
│   ├── api/                      # OpenAPI YAML specs
│   ├── reference/                # Changelog, troubleshooting, PRD
│   ├── diagrams/                 # Architecture diagrams
│   └── assets/                   # CSS, JS, favicon
├── mkdocs.yml                    # MkDocs configuration
├── requirements.txt              # Python dependencies
└── mise.toml                     # Task runner

---

Deployment

Docs are automatically deployed to GitHub Pages on every push to main via the deploy workflow.

To deploy manually:

mise run deploy

---

Contributing

1. Edit or add Markdown files inside docs/
2. Run mise run up to preview changes live
3. Run mise run lint to catch build errors before opening a PR
4. Open a PR against main

---

Source Platform

The DSTA platform source lives across multiple private repositories under minhdqdev-org:

• dsta-core-svc — Auth, config, notifications (Django 5.2)
• dsta-trading-svc — Strategies, execution, risk (FastAPI)
• dsta-data-svc — Market data, backtesting (FastAPI + TimescaleDB)
• dsta-ml-svc — ML features, predictions, RL agent (FastAPI)
• dsta-exchange-svc — Exchange gateway + WebSocket (FastAPI)
• dsta-web — Frontend (React 19 + Vite 7)
• dsta-cli — CLI tool (Typer)
• dsta-qa — Quant analytics + E2E tests

---

Proprietary — DSTA Team