Trading Infrastructure

A systematic trading research and execution infrastructure.

Components

Component	Status	Entry point
Marcus — macro regime classifier	✅ Live	systems/signals/regime_classifier.py
Sarah Stage 1 — daily vol pipeline	✅ Complete	systems/sarah/daily_vol_run.py
Sarah Stage 2 — analytic greeks engine	✅ Complete	systems/sarah/greeks_tool.py
Sarah Stage 3 — scenario P&L engine	✅ Complete	systems/sarah/scenario_engine.py
Sarah Stage 4 — pre-trade dashboard	✅ Complete	systems/sarah/pretrade_dashboard.py
Sarah Stage 5 — regime library / analog search	✅ Complete	systems/sarah/regime_library.py
Jordan — risk layer	⬜ Not built	systems/risk/
Priya — research backtesting	⬜ Not built	research/
Kai — execution layer	⬜ Not built	systems/execution/
Alex — orchestration	⚠️ Framework only	scheduler.py

Databases

File	Purpose
data/processed/macro.db	Marcus: macro series, regime history, COT, calendar
data/processed/trading.db	Sarah: vol signals, vol surface, VVIX daily

User-facing interfaces

Marcus dashboard (Dash web app):

python systems/dashboard/macro_dashboard.py
# → http://127.0.0.1:8050

Sarah — programmatic (no GUI):

python systems/sarah/daily_vol_run.py          # Stage 1: refresh vol signals
python systems/sarah/greeks_tool.py            # Stage 2: greeks demo

See docs/architecture/current_state.md for full API reference.

Setup

bash setup.sh

Verification

python scripts/verify_phase0.py

Pipeline

The scheduler runs automatically:

• Weekdays 08:00 — Sarah vol surface run (requires Marcus regime_state.json)
• Weekdays 18:05 — Marcus macro pipeline (FRED → regime classify → snapshot)
• Sunday 20:00 — Full FRED history refresh

python scheduler.py