I developed a machine learning pipeline as a Summer Research Intern with the Dept. of Mechanical Engineering, MNNIT Allahabad, using 152 tribology records to predict eight bearing performance parameters. I benchmarked FFNN, RBNN, and GRNN models; FFNN achieved an average test R of 0.991 and MAPE of 2.6%.
For DeliverySense, I designed a nine-table schema and a Python ETL pipeline to load 100K+ Olist orders into PostgreSQL. I analyzed delivery delays by region, product category, and seller-customer geography, then launched a Streamlit dashboard with root-cause views.
I also created an AI Tool Adoption Analytics Dashboard in Power BI, preparing 50K+ records and authoring 15+ DAX measures. My work spans data analysis, dashboards, SQL, and applied machine learning.
