Data Science Student Consultant

• Jul 2022 - Present

Risk Analytics for Health Non-Profit (Python, Tableau, LightGBM) • Predicted maltreatment risk score using tree-based classification by integrating investigation features for vulnerable individuals to prioritize complaints • Built a Tableau dashboard for investigation team to prioritize complaints and increase staffing efficiency, with a 65% increase in proficiency in investigation evaluation process Exploratory Analytics for Educational Assessment Provider (Python, SQL, Tableau, Lasso Regression) • Analyzed 7M candidate data to understand regional performance disparity and candidate persistency using Lasso Regression • Monitored 8 KPI’s for candidate performance through a Tableau dashboard; recommended a checkpoint system to improve test center support for candidates • Identified 1600 under-performing cities using clustering with matching techniques, informing causal impact of study resources on test scores for optimal resource allocation Data Analytics for Second Harvest Heartland (R, Linear Regression) • Identified impact of media channels on donations using Linear Regression model based on 2 years of data of earned media • Provided recommendations to client to focus resources on 6 out of over 100 media channels to increase earned media donations Show less

Business Analyst

• Apr 2020 - Apr 2022

• Managed and ingested raw telecom data as a part of the delivery team for the deployment of a Pan-European Network Performance Platform for Vodafone’s customer groups across Europe replacing 100s of separate network performance applications in use • Engineered data links from Google Cloud Storage, with gsutil command line, and Hadoop Distributed File System • Modeled the incoming data for various granularities using SQL nested queries in SparkSQL and Apache Hive • Scheduled and monitored jobs on Apache Yarn to produce processed data out in Google Cloud (CSV, AVRO, Elasticsearch) • Created visualizations and dashboards for data analysis using Kibana and provided reports and insights • Infrastructure Design for ELK Clusters processing 8 billion network performance events per day • Automated web tasks using Python scripts with Selenium, reducing manual work by 80% Show less

Intern

• Jan 2018 - Jun 2018

• Performed automated testing of conversational platform for enterprise application using advanced Selenium as a testing tool with Python • Performed automated testing of conversational platform for enterprise application using advanced Selenium as a testing tool with Python

Intern

• May 2017 - Jul 2017