Quantitative Finance · Risk · Data Analytics · Machine Learning

Financial Engineer with hands-on research experience in high-frequency liquidity modeling, regime-aware trading strategies, and quantitative risk infrastructure. Designed a 6-metric Liquidity Stress Score framework processing 20M+ NASDAQ TAQ records, achieving a Sharpe Ratio of 5.43 with 100% win rate on high-confidence signals. Built regime-conditioned portfolio systems using HMM, Graph Attention Networks, and Reinforcement Learning that outperformed benchmark drawdowns by over 20%. Passionate about building systematic, data-driven frameworks at the intersection of market microstructure and machine learning.

Stevens Institute of Technology — MS in Financial Engineering (GPA: 3.97/4.00) | Expected May 2026 Graduate Certificate: Algorithmic Trading Strategies · Concentration: Quantitative Finance & Statistical Modeling Coursework: Algorithmic Trading, Market Microstructure, Computational Methods in Finance, Pricing & Hedging, Advanced Derivatives, Stochastic Calculus

Jaypee University of Engineering & Technology — B.Tech in Computer Science & Engineering | May 2024

Stevens Institute of Technology | Research Assistant — Market Liquidity & High-Frequency Trading (Prof. Dragos Bozdog) May 2025 – Aug 2025 | Hoboken, NJ

• Engineered cross-asset risk pipelines processing 20M+ NASDAQ TAQ records; developed a 6-metric Liquidity Stress Score (LSS) framework that directly identified top 1% rare intraday stress events for alpha generation
• Achieved RMSE of 0.323 and R² of 0.197 on out-of-sample data by building minute-ahead execution risk models using Random Forest, XGBoost, and Elastic Net on NASDAQ TAQ data
• Designed a mean-reversion trading strategy triggered on top 1% LSS events, achieving a Sharpe Ratio of 5.43 on a $50,000 capital base with 100% win rate across 5 high-confidence signals

Lintel Technologies Pvt. Ltd. | Python Developer Intern April 2023 – Oct 2023 | Gujarat, India

• Developed Python analytics solutions for a fintech platform, integrating telephony APIs to automate KPI reporting
• Automated data processing workflows using Python and SQL, reducing manual reporting effort and improving accuracy

Credencys Solutions Inc. | Software Engineer Intern Jun 2022 – Jul 2022 | Gujarat, India

• Automated reporting workflows for an FP&A client using SQL/Python, improving data processing efficiency
• Built reporting dashboards to monitor business metrics; improved code quality via testing and reviews

Regime-Aware Pairs Trading Strategy — Python, HMM, GAT, Reinforcement Learning | 2026

• Screened 741 pairs across 39 healthcare equities; identified 36 cointegrated candidates via Johansen testing
• Trained a Graph Attention Network achieving Spearman rank correlation of 0.56 (p = 0.0004) between model scores and realized P&L
• Deployed a PPO agent with regime-conditioned execution achieving 94–96% win rates across COVID and post-COVID market regimes with controlled drawdowns

RegimeSense: Market Regime Detection & Portfolio Optimization — Python, HMM, K-Means | 2026

• Built a 3-state HMM on 20 years of SPY, VIX, and yield-curve data (2005–2025) to classify Calm, Neutral, and Turbulent market regimes
• Regime-conditioned allocation achieved 8.6% annual return, 12.1% volatility, -34.6% max drawdown vs SPY's -55.2% drawdown

Factor-Based Long-Short Portfolio Allocation under Beta Constraints — Python, Fama-French, CVXPY | 2025

• Constructed two long-short strategies over 12 global ETFs using Fama-French 3-factor model with weekly rebalancing from 2007–2025
• Strategy II achieved 828.80% cumulative return vs SPY's 535.62% with Sharpe ratio of 0.66 vs 0.54; lower CVaR (2.94% vs 3.08%)
• Analyzed sensitivity to estimation horizons and risk-aversion parameter across 5 market regimes including GFC and COVID-19

American Options Pricing & Delta Hedging on SPY — Python, LSMC, Binomial Tree, Crank-Nicholson | 2025

• Implemented and compared 3 numerical methods for pricing American options on SPY using real Bloomberg data
• Computed Greeks (Delta, Gamma, Vega, Theta) across all methods; validated against market prices
• Implemented dynamic delta hedging strategy maintaining delta-neutral position across 5 trading days

BTC Market Microstructure — Volatility, Liquidity & Market Resilience — R, High-Frequency Data | 2025

• Analyzed tick-level BTC/USD data (14,618 observations) from BitMex; estimated realized volatility at 30.76% and Roll Model volatility at 28.97%
• Computed quoted, effective, and Roll spreads to quantify liquidity; found effective spread (0.406 USD) significantly below quoted spread (0.569 USD) indicating hidden liquidity
• Analyzed immediate and cumulative price impact of large trades (>40,000 units) on market resilience

Sensitivity of Airline Stock Returns to Oil Price Changes — R, Correlation & Regression Analysis, Bloomberg | 2024

• Tested whether Delta and United Airlines stock returns move inversely with crude oil returns using 5 years of weekly Bloomberg data (2019–2024)
• Found weak positive correlations instead (Delta: 0.161, p=0.009; United: 0.088, p=0.157), rejecting the inverse-relationship hypothesis
• Cross-validated R regression outputs against Bloomberg's native regression tool, confirming consistency with industry-standard calculations

Programming: Python (pandas, NumPy, scikit-learn, statsmodels) · C++ · SQL · R

Finance & Math: Fixed income pricing · Derivatives valuation · Black-Scholes · Monte Carlo · VaR · CVaR · Risk decomposition · HMM · ARIMA · GARCH · PCA · OLS · Cointegration · Machine Learning

Tools: Bloomberg Terminal/API · Git · Excel

Certifications: Baruch C++ for Financial Engineering

• Beta Gamma Sigma (BGS) — Top 20% of master's students in AACSB-accredited programs
• Led digital literacy programs for 200+ underprivileged children; improved computer skills by 90%
• Managed fundraising Cyclothon for Shivam NGO; raised $5,000

Bharatnatyam · Skating · Athletics · Dance · Reading · Sketching · Painting · Art · Music