Evidence-governed quantitative cyber risk — a trustworthy CLI and scenario engine where every number traces to a reviewed public source.
-
Updated
Sep 15, 2026 - Python
Evidence-governed quantitative cyber risk — a trustworthy CLI and scenario engine where every number traces to a reviewed public source.
Cost-sensitive credit card fraud detection on 1.3M transactions. Picks the decision threshold that minimizes expected dollar loss, avoiding $1.25M vs a naive baseline (XGBoost, PR-AUC 0.91). Adds a quant-risk layer (VaR/Expected Shortfall, walk-forward backtesting, PSI drift) and an agentic RAG investigation copilot. Interactive Streamlit app
Sequential comparison of probabilistic forecasts and anytime-valid inference in R.
End-to-End Market & Credit Risk Analytics Engine under Basel III / EBA standards. Features Parametric & Historical Value-at-Risk (VaR), Expected Shortfall, GARCH(1,1) Volatility Forecasting, Basel III Portfolio Stress Testing, ALM Interest Rate Sensitivity (EVE/NII), Basel Traffic Light Backtesting, PostgreSQL, and Power BI.
Cyber Risk Posture Calculator — AI-powered financial exposure estimates for SMBs
Actuarial & quantitative risk portfolio — non-life insurance (GLM, Monte Carlo, Solvency II)
FAIR Monte Carlo cyber risk quantification: translates technical vulnerabilities into probable financial loss distributions, then has Claude draft the board narrative. Next.js + Recharts + Trigger.dev + Supabase.
Risk Assessment & Management Labs
Financial Risk Management
To associate your repository with the quantitative-risk topic, visit your repo's landing page and select "manage topics."