• Build a random forest and xgboost model to identify and segregate customers into bucketsto determine their potential to adopt solar rooftops.• Used PU learning to handle unbalanced datasets and increased the efficiency of the model by 20%

• Worked in a team to analyze data from a cement power plant to find the exact parts responsible for the failure by using correlation and the Granger causality test.• Found optimum lag between variables in the power plant in different time frames by making models using regression analysis and calculating RMSE and MAE.• Build a machine learning model to predict the accuracy of secondary air temperature and predict failures in power plants by using an Artificial Neural Network and increase the accuracy to 90%.• Used k-means algorithm to classify dataset into clusters and predict new data points into clusters to easily identify new failures at might happen in the future.

• Designed a dashboard on Tableau to get insights about a company and did Businessintelligence to help make decisions about product sales for a client.• Cleaned up a large dataset and did data modeling to make the schema for it and import it intoPower BI to make dashboards to help clients get an insight into their company and make important decisions.• Designed a data warehouse by converting data from different sources into one and analyzed it to make business decisions by making queries on it using SQL.

• Reviewed, studied and analyzed multitudes of energy audits for Rajasthan Power Discoms tounderstand the company's financial position.• Conducted a comparative study; designed and proposed an implementation strategy to utilize solar energy to cut-down the costs.• Based on the conducted analysis, developed a linear model to restructure the billing system for power distribution companies.