I combine CS training and finance study to build volatility, alpha and risk models, valuations, and data-driven dashboards with Python, R and SQL.
Projects include a multi-model SPY volatility engine, a multi-regime alpha engine, a 13-model VaR study, and an FP&A and valuation externship for a private IT firm.
.png)
Attronica · Oct 2026 · ✅ Verified by Extern
A 40-year-old, $80M family-owned IT firm is preparing to sell — and they need real FP&A work to back it: clean financials, a defensible forecast, and a valuation. Organize the books, model the forecast, and deliver a valuation that lands on the CEO's desk and on a buyer's table.
Built a robust out-of-sample volatility forecasting framework for SPY, integrating GARCH-family models, gradient boosting (XGBoost, LightGBM), and stacked ensemble techniques; engineered advanced quantitative features including fractional differentiation, Garman-Klass and Parkinson volatility estimators, overnight gap signals, VIX term structure, and credit stress indicators; achieved 15 620% improvement in predictive accuracy over baseline models, validated using QLIKE loss, RMSE, and Diebold-Mariano tests.
I combine CS training and finance study to build volatility, alpha and risk models, valuations, and data-driven dashboards with Python, R and SQL.
I am Rajvardhan Jadhav, a graduate student in M.S. Quantitative Finance at Temple University and a Junior Consultant. I trained in computer science (B.Tech) and have interned in data science and engineering roles. I build quantitative models, valuations, and risk tools using Python, R, SQL and financial econometrics.
Externships
Attronica Financial Planning & Analysis and Private Company Valuation Externship
Completed · Oct 2026
Experience
Data science Intern
Hinduja Global Solutions · June 2024 - July 2024
Engineering Intern
Hinduja Global Solutions · June 2023 - July 2023
Junior Consultant
Principio Holding · March 2025
Startup advisor
Varidus · December 2024
Education
National Institute of Technology, Warangal
B.Tech Computer Science and engineering · Class of 2025
Fox School of Business(Temple University), Philadelphia
M.S. Quantitative Finance · Class of 2027
Skills
✅ Verified by Extern
A 40-year-old, $80M family-owned IT firm is preparing to sell — and they need real FP&A work to back it: clean financials, a defensible forecast, and a valuation. Organize the books, model the forecast, and deliver a valuation that lands on the CEO's desk and on a buyer's table.
The externship produced financial planning and valuation deliverables for a family-owned IT firm preparing for a potential sale. The work involved organizing historical financials, building a forecast model, and producing a valuation analysis. Deliverables included cleaned financial statements, a forecast model, and a valuation report.
I consolidated audited FY2022–2025 statements into a single, formula-driven workbook with income, balance sheet, cash flow, and equity tabs, added a Ratio Analysis and 15 audit checks, and documented data sources and known limitations.
Built a robust out-of-sample volatility forecasting framework for SPY, integrating GARCH-family models, gradient boosting (XGBoost, LightGBM), and stacked ensemble techniques; engineered advanced quantitative features including fractional differentiation, Garman-Klass and Parkinson volatility estimators, overnight gap signals, VIX term structure, and credit stress indicators; achieved 15 620% improvement in predictive accuracy over baseline models, validated using QLIKE loss, RMSE, and Diebold-Mariano tests.
Developed a multi-regime systematic portfolio construction framework combining machine learning-based return forecasts with Bayesian Black-Litterman optimization across 15 multi-asset ETFs; engineered cross-sectional factor signals (momentum, RSI, MACD, Bollinger Bands, realized volatility, higher-moment statistics); integrated GJR-GARCH volatility forecasts, regime-switching logic, and CVaR/mean-variance optimization; conducted performance validation using Deflated Sharpe Ratio, block bootstrap simulations, and information coefficient analysis.
Built and evaluated 13 advanced VaR models including GARCH variants, Extreme Value Theory (EVT), and Filtered Historical Simulation (FHS) applied to a 20-year multi-asset ETF portfolio; performed backtesting using Christoffersen tests and Expected Shortfall validation; quantified a ~89% increase in portfolio VaR during crisis periods via Component VaR decomposition.
