Siddhant Laddha Email & Phone Number

Chicago, Il, Us

Boston, Massachusetts, United States

Bengaluru, Karnataka, India

1. Created and backtested a currencies hedging pipeline, creating a dynamic basket of synthetic products with correlated risk profiles to reduce long tail risk by 0.5mm/bp on average2. Worked on a deep hedging pipeline using reinforcement learning for long-term financial derivatives for variable annuities for European markets reducing downside risk by 30%3. Implemented liquidity, sensitivity and volatility analysis suggesting real-time trades to the equities desk for index baskets, to reduce exposure by $10mm on a weekly basis4. Improved a pricing engine batching for correlated products, across asset classes such as commodities, rates and FX to reduce recalculation and improved pricing time by 15%5. Developed a distributed pipeline to determine cross asset illiquid risk held by the firm, offsetting within firm risk and reducing Tier-1 cash holding requirement by 8%

Bengaluru, Karnataka, India

1. Implemented an identification and remediation strategy for inconsistent PnL vectors, leveraging K- means clustering, hamming distance & lasso regression, reducing risk-weighted average capital for each desk by ~10%2. Devised correlation metrics, k-means grouping and hamming distance to backfill data for new products such as single-name IPO stocks, bitcoin futures, illiquid currencies to improve pricing engine performance

Banglore

Enhanced Auto-ML capabilities which democratizes the ML model development process, empowering users, no matter their expertise, to identify an end-to-end machine learning pipeline for any target metric. 1. Created tools for generating algorithm agnostic visual explanations such as SHAP and micro-segment analysis, enhancing modeler interpretability to validate model robustness2. Migrated modeling capabilities and data hosting from an individual server based solution to a cloud-based solution, bridging architecture gaps and refactoring code, saving $300K yearly on hosting3. Ideated a pipeline leveraging NLP, spacy and regex to convert natural language to SQL with 95% accuracy4. Developed & tested a API implementation to classify prospects and display personalised ads with 70% accuracy while reducing modelling time by 15%, maintaining accuracy through LightGBM implementation5. Worked on feature engineering and dimensionality reduction, parsing through hundred of features using adjusted mutual information and information, improving model efficiency by 25%

India

1. Tested and simulated over 100 trading alphas over metrics sharpe ratio and coverage to determine robustness2. Developed delay-0 alphas for USA, Europe and Asia, collating data from 5 + providers for prediction pipeline3. Created an investment strategy received a total of 1.5 weight in selected portfolios

Gurgaon, India

1. Facilitated supply chain analysis and logistics network design for a cold chain tracking system for medical supply-chain ecosystem using Hyperledger Fabric involving 3 projects for temperature sensitive medicines and vaccines

Master'S Degree, Financial Mathematics

Improve pricing of catastrophe bond pricing using probability of event and expected loss 1.Created a pipeline to generate a prediction.

Biotechnology, Bioinfomatics

Activities and Societies: Business Club,Debating Scoiety