Varad Pawar

Developed and implemented an ARIMA model to analyze time series NO2 emission data, successfully forecasting 2020 emission levels and demonstrating a significant decrease in NO2 levels during the COVID-19 lockdown. Conducted stationarity tests (ADF test), model selection using AIC/BIC, and time-series decomposition to extract key patterns from 150K records.

Analyzed student performance data across 15+ states, processing over 1 million student records using Python (Pandas, NumPy) and R to provide actionable insights. Designed scalable ETL pipelines in PySpark to automate student-level data aggregation and statistical analysis for district & state-level reporting. Implemented robust data validation checks using Python and SQL to identify missing values

Led lab practicums for statistics course, overseeing 100 students in hands-on application of statistical concepts using SPSS software. Facilitated hands-on exploration of descriptive statistics, probability distributions, correlation, regression, Bayes theorem, estimation, hypothesis testing, and nonparametric methods, resulting in a 90% improvement in students' data interpretation skills.

Deployed and implemented over 10 quantitative credit and risk strategies using FICO Decision Management Platform (DMP) and FICO Blaze Advisor, leveraging rule engine logic, Python, and SQL for data-driven decisioning, resulting in a 20% reduction in manual underwriting errors. Migrated credit risk models and rule strategies from Blaze Advisor to FICO DMP, supporting Prime Mortgage and Unsecured Le

Grade: 3.76/4

Activities and societies: Coursework: ML, NLP, Visualization, Regression, Time Series, Big Data, Predictive Modeling, DB Engineering, Statistical Inference.

Pursued a Master of Science in Applied Data Science, focusing on machine learning, natural language processing, and big data. Gained expertise in predictive modeling, database engineering, and statistical inference.