Rishabh Soni Email & Phone Number

Bengaluru, Ka, In

Bengaluru, Karnataka, India

Gen AI for Business

Bengaluru, Karnataka, India

Hands-on leader managing a team of data scientists and software engineers - Building ML opportunity identification platform which has led to multi million dollar revenue impact. These models recommend transaction opportunities spanning multiple IB products/regions e.g. M&A Sell-Side, ECM/DCM, Private Markets (IPO/M&A) etc- Build ESG analytics toolkit for client advisory e.g. carbon emissions profiling, investor analytics, cost of carbon, activism analytics etc- Recommendation engine to show relevant thematic content to clients- PDF parsing and table extraction from allocation documents (relationship lending/DCM)- Redesigned agency models for CCAR reporting- Build a PSI RFP automation tool using custom RAG for Large Language Models - Analytics for various client situations such as M&A, Activism etcTech Stack: Python, AWS, Tableau

Bengaluru Area, India

- Implemented a classification model to predict whether the FX available in a fixed income portfolio should be converted to portfolio base currency to avoid currency risk. Models implemented: Logistic Regression, Random Forest, LightGBM- Worked on statistical quantification of market herding bias in portfolios. Models used: Mixed effects model, Linear Regression- Forecast offer spread for corporate bonds. - Working on optimal trade execution strategy for high income trades using liquidity measure. - Built a classification model to predict the hard to borrow stocks for sec-lending market. Models implemented: Logistic Regression, Random Forest, XGBoost- Identification of impact of etf flows on stock characteristics such as returns, liquidity after controlling for systematic factors such as momentum, value etc. Models implemented: Quantile regression, Robust Regression- Helped in building a default model for banks (for counter-party) using features suggested in Basel Norms. Model used: Balanced RF, Balanced Logistic RegressionOther Initiatives:- Predict the outflows flows in portfolios using market factors, customer trends etc- Build a missing value imputation recommendation guide for data scientistsWinner of Eureka Award for innovationTech Stack: Python, R, SQL, Tableau, Snowflake, AWS

Gurgaon, India

- Behavioral Finance: Developed algorithms to detect trading patterns/investment styles using trade diaries for fundamental equity Portfolio Managers- Developing algorithms for operational and liquidity risk profiling. Building and enhancing factor risk models. Designed and implemented new exception framework which is used for regulatory reporting. - Portfolio constraints classification for private investor business. Design new methodology to calculate tracking error- Developed a recommendation algorithm for investment management documents using word embeddings to enhance the operational process- Anomaly detection algorithm in expenses to forecast operational riskTech Stack: R, Python, SQL, PySpark

Noida Area, India

R&D initiatives- Developed signal processing and deep learning capability on EEG data to predict seizure. (Paper presented at international healthcare conference - IIM Ahmadabad).- Developed claim fraud algorithm by association mining and clustering to identify wrong diagnoses, procedure bundling. This helped reduce fraud claims by ~5%- Enhanced discovery capability of intranet search portal using NLP- Predict claims with high risk of appeals in M&R space. This was replicated for other LOBsProject Delivery- Built a dynamic network optimization algorithm to determine next best provider to meet CMS requirement of having atleast 90% member coverage for all specialties- Predict ER visit for asthma patients using internal claims data supplemented by air pollutant, weather data etc (Paper presented in internal UHG conference)- Predict providers with high chances of having an incorrect record. Model successfully utilized for campaigns with ~60% hit rateTech Stack: R, Python, SQL, PySpark, Hive

Hyderabad Area, India

- Built a model for patient allocation in trials based on various constraints such as cost and time. Created a risk score algorithm to assess the performance of investigators in a clinical trialTech Stack: R, Python, SQL

Gurgaon, India

- Developed promotion effectiveness algorithm for an Indian retailer. Built a dashboard to evaluate overall and promotion level impact- Analyzed transaction data for a Thai Retail Chain, identified B2B and B2C transactions, and created recommendations by region which were delivered in form of Shiny App- Performed segmentation for an Indonesian Telecom to understand customers’ needs/demands. Market basket and correlation analysis was used to bundle various packets for cross-sellTech Stack: R, Python, SQL, Alteryx

Greater New York City Area

- Led the initiative to bring credit card submissions (50 TB) into HDFS. Performed UAT to ensure data sanctity. This assessment led to projects such as small merchant risk scanner& location based targeting- Developed and implemented the Merchant Exposure Calculation (PIT) algorithm. Managed the development of Exposure Management Capability. Interacted with global business partners to build wireframe for the tool- Worked directly with senior leadership to size two major initiatives: Transaction amount/Portfolio based Pricing and small merchant portfolio revamp using external data- Enhance marketing strategy of a top credit card issuer by predicting customers’ channel affinity. Developed offer level customer segmentation using predictive modeling and assessed campaign performanceTech Stack: SAS, SQL, R