Jun 2019 – Feb 2020 · 8 mos · Mumbai Metropolitan Region

Jul 2017 – May 2019 · 1 yr 10 mos · Mumbai Metropolitan Region

• Involved in the Methodolgy Development for the Economic and Regulatory Capital Models.
• Developed Quantitative Risk Models for Equity Risk Arbitrage, Life Insurance Trades and Litigation
• deals.
• Recalibrated scaling factors for higher quantiles by fitting tail distribution using Extreme Value Theory
• for both the models.
• Documented and performed in-depth analysis for both the models in accord with the SR 11-7 guide-
• lines which comprised of but not limited to sensitivity analysis, outcome analysis(backtesting), and
• benchmarking.
• Enhanced the pre-existing models methodologies for better estimation of the risk.
• Fully implemented the excel based risk models in R programming language for faster computation and
• analysis.
• Implemented enhanced monte-carlo based solution for Credit Spread, Migration Risk and Default Risk
• Overlap problem.
• Researched on impact of overlapping returns on estimation of the risk at higher quantiles.
• Worked on various statistical methods like Generalized Method of Moments(GMM) for estimating the
• parameters and Expectation-Maximization for solving missing data problem.

Apr 2014 – Jun 2017 · 3 yrs 2 mos · Mumbai Area, India

• Machine Learning Based Trading Algorithms
• Developed pattern recognition model using SVM to determine RSI values which spot the market trends
• Implemented the deep-learning neural network to model the major indices using technical indica-
• tors.
• Developed models based on Naive-Bayes Classifier and decision trees to determine the trending
• stocks in Indian and US capital markets.
• Bar-Breakout
• Developed a Intra-Day Momentum Trading Strategy from widely used technical analysis library
• TA-LIB in C++ on Interactive Broker(IB) Unix platform.
• Implemented Socket Programming to create a hub of large no. individual investors connect to a
• single IB C++ API.
• Order Execution Optimization
• Optimized the execution of large volume of trades in short time window of 10 minutes with
• minimum impact on stock price.
• Developed a highly profitable statistical algorithm based on the five-day rolling average to decide
• the portfolio positions.
• Performed risk analysis and backtesting for various trading strategies using pyfolio python library.
• Deployed machine learning algorithms for trading in tensor-flow Python environment.

Sep 2013 – Mar 2014 · 6 mos · Mumbai Metropolitan Region

• Studied the higher-moments of particle multiplicity to locate the Critical Point for the first order
• transition between hadron matter and quark-gluon plasma.
• Carried out the charge-dependent Particle Correlation measurements to study the local parity violation.
• Performed
• Sphericity Analysis for the events producing Helium and Anti-Helium in Au-Au collisions
• at √ s = 200GeV using the STAR detector.

Feb 2010 – Aug 2013 · 3 yrs 6 mos · Mumbai Metropolitan Region

• Photon-Hadron Classification by Multi-Variate Analysis
• Analysed different variables in Particle Multiplicty Detector in ALICE most appropriate for the
• photon-hadron classification.
• Achieved classification efficieny of 78% using ROOT TMVA package for GEANT4 generated
• simulated data.
• Researched different variants of the Classical Backpropogation Algorithm used to solve the pattern
• recognition problem.
• Analyzed the performance of the Scaled Conjugate Gradient Backpropogation Algorithm to clas-
• sify the particles generated during high-energy collision in Large Hadron Collider(LHC).
• Machine learning C++ Software Development
• Developed a Multi-Layer Feed-Forward Neural Network Software in C++ which implements the
• standard Backpropogation Algorithm for particle classification.
• Re-Designed the pre-existing Support Vector Machine Software(LIBSVM) to meet the quality
• coding standards required by the ALICE ROOT Framework in LHC.

Aug 2009 – Feb 2010 · 6 mos · Mumbai Metropolitan Region

• Structured Product Development
• Implemented FX Range Accrual, Snowball and Digital Strip Options on KONDOR+ for a globally
• well reputed bank.
• Validated the pricing of the above instruments against the standard benchmark models available
• in Numerix Pricing Engine with an approximate error range of +/−0.2%, +/−0.8% and +/−1.5%
• respectively.
• Option Trading Strategies, Option Pricing using Black Scholes Model

Jan 2008 – Jan 2008 · 0 mo · Gainesville Metropolitan Area

• Developed the price determination models for the corn and wheat corps.
• Estimated the model using step-wise multiple regression in SPSS.

Jan 2007 – Jan 2007 · 0 mo · Mumbai Metropolitan Region

• Automated the treasury reports generation for various asset classes like Equities, Mutual Funds on Actuate Reporting System.