Welcome, This is a place where I document my learning, research, some revision notes, and evolving ideas
in data engineering, machine learning and quantitative finance.
I enjoy exploring markets and non-market problems constantly experimenting and uncovering deeper
insights.
Feel free to reach out to me.
I am an engineer driven by a love for solving complex problems. My journey began at National Institute of Technology, Trichy and led to PayPal, where I worked as a Software Engineer tackling data at extreme volumes.
I enjoy engineering challenges and problem solving by seeking deeper insights and utilizing data. I am currently pursuing a Master's in Financial Engineering at Stevens Institute of Technology.
Master of Financial Engineering
Algorithmic Trading Strategies Specialization
Bachelor of Technology
Metallurgy, Materials Science & Microeconomics
PayPal
Enterprise Data Lake / Unified Enterprise Customer
Built 3-state Hidden Markov Model with HAR-RV-J forecasting decomposing realized variance into diffusion/jump components. Integrated Lee-Mykland jump detection and Hawkes flash crash modeling. Optimized VWAP/TWAP execution sizing with regime-dependent CVaR constraints, achieving Sortino 2.41 (vs Sharpe 1.57) on 252-day regime-switching synthetic data.
Developed Mean-Variance Optimization framework with Two-Fund Separation Theorem to construct minimum-variance synthetic hedge baskets, isolating idiosyncratic risk from concentrated equity positions. Implemented closed-form Lagrangian solver for efficient frontier computation. Achieved beta neutrality with 30–40% volatility reduction relative to unhedged position.
Implemented Black-Scholes, Heston, and SABR models in C++ using Monte Carlo, PDE (Crank-Nicolson), and binomial/trinomial tree solvers for equity options. Applied Factory and Strategy design patterns for extensible Greeks computation (Delta, Gamma, Vega) across pricing methods.
Engineered MACD-RSI rule-based strategy and Random Forest classifier on technical features (EMA, RSI, PPO, MACD, CMO) to forecast price direction. Optimized entry/exit via out-of-sample Sharpe ratio and momentum thresholds (δ = 2–15%) across 10 large-cap stocks.
Ranked 6th out of 150+ teams with Sharpe ratio of 2.74 over 12-week period
2025Ranked 1st in Data Science Club NIT Trichy competition — multiclass classification using CNN ensemble
2021Thoughts on cool things.
A definitive breakdown of the hardware bottlenecks, memory architectures, and framework optimizations required to scale machine learning to millions of users.
Exploring Hidden Markov Models for volatility regime detection and their applications in adaptive trading strategies. We examine transition probabilities, state persistence, and real-world implementation challenges.
Why Conditional Value-at-Risk (CVaR) provides superior tail risk measurement compared to VaR. Practical implementation for portfolio optimization and position sizing.
Implementing Lee-Mykland jump detection methodology. How to distinguish between continuous price diffusion and discontinuous jumps in tick data.
Complete walkthrough of the Black-Litterman model for portfolio optimization. Incorporating investor views with market equilibrium returns.
Beyond technical indicators: creating predictive features from price data, volume patterns, and order book dynamics for ML models.
Mathematical framework for Volume-Weighted Average Price execution. Balancing market impact, timing risk, and implementation shortfall.
Books that shaped my understanding of quantitative finance, trading, and systematic strategies.
Essential framework for ML in finance — fractional differentiation, meta-labeling, and bet sizing strategies.
Comprehensive guide to market microstructure, execution algorithms, and direct market access.
The bible of quantitative portfolio management — information ratio, transfer coefficient, and alpha forecasting.
Deep dive into options, volatility trading, and risk management from a practitioner's perspective.
Demystifying quantitative and algorithmic trading strategies — from alpha generation to execution.
Rigorous foundation in continuous-time models, Brownian motion, and derivatives pricing theory.
Practical strategies for algorithmic trading — mean reversion, momentum, and statistical arbitrage.
Modern perspective on limit order books, optimal execution, and high-frequency market making.