• Contributed to internal Python libraries to facilitate index construction.• Crafted factor models using alternative datasets for factor investing and collaborated to enhance novelresearch in portfolio construction.• Analyzed historical market data employing diverse portfolio construction tools.• Optimized the index rebalancing workflow and authenticated rebalancing outcomes utilizing Python andSPEL.• Created an automated visualizer tool for pre-checking index data utilizing LLM.

• Discovered regions of interest (ROI) zones for renewable resource infrastructure through machinelearning approaches.• Identified lung diseases with a custom-built CNN architecture, achieving a 95% accuracy rate.• Predicted seismic risk areas by categorizing them into 5 zones using clustering techniques and assessedtheir magnitudes with regression methods.

• Investigated behavioral data to analyze post-error slowing response times in mice, examining theinfluence of external and internal conditions like temperature, lighting, attention span, contrast level,and session phase.• Conducted regression analysis resulting in a 95% Mean Squared Error (MSE) score attainment.

• Acquired hands-on experience utilizing Python for data analysis, SQL for database querying, andimplementing functionalities within the AWS cloud infrastructure.

• Developed a machine learning model using regression techniques to predict payment dates and classifiedthe system-generated invoices into different payment timeline groups based on the anticipated paymentdates.• Implemented an Invoice Bill Management Application using ReactJs and Java, enabling search, edit, anddelete functionalities within the system.• Seamlessly integrated the above ML model to enhance the application capabilities.

• Conducted exploratory data analysis (EDA) to examine the correlation between students’ test scoresand factors such as gender, ethnicity, parental education level, lunch, and test preparation courses.• Constructed a regression-based machine learning model to forecast the prices of pre-owned cars usingKaggle data, achieving a 92% R2 score.