Muhammad Murtaza Mir

Master's

Data Science

Rocket Central, LLC

Data Science intern

• Created 10+ SQL scripts for creating and modifying database objects like Tables, Views, and Stored Procedures. Developed 3 AWS Glue ETL scripts using Python, and PySpark to automate the data transfer between different buckets in AWS S3. • Successfully completed a business-critical project by extracting relevant data consisting of 5 million records. Performed data cleaning and preparation, data analysis, feature engineering, and hyper-parameter tuning on Random Forest for binary classification. Communicated the insights by creating visualizations using Power BI and Seaborn. • Worked in a team on an Unsupervised Machine Learning problem. Performed clustering analysis on the data, obtained from multiple sources consisting of 8.3 million records, by applying the K-Means algorithm and formulating relevant clusters leading to the identification of the customer base. Applied Elbow Curve and insights from the data to find the appropriate number of clusters.