Breaking Games Externship via ExternMaster's student in Data Science with early-career experience and an externship in deal sourcing and startup analysis at Mangusta Capital.
Projects and analyses focused on sourcing opportunities and evaluating startups, combining data science methods with market and deal research from my Mangusta Capital externship.
Analyzed e-commerce transaction data using Python to identify sales, customer, product, and marketing trends. Performed data cleaning, EDA, feature engineering, and correlation analysis to evaluate business performance. Identified a 0.0104 correlation between advertising spend and revenue, supporting data-driven marketing decisions.
Analyzed 10,000 bank customer records to identify key factors influencing customer churn. Performed data preprocessing, EDA, categorical encoding, feature engineering, and SMOTE for class balancing. Built and evaluated Logistic Regression, Random Forest, and XGBoost models, achieving 82% accuracy and 0.84 ROC-AUC with Random Forest.
Master's student in Data Science with early-career experience and an externship in deal sourcing and startup analysis at Mangusta Capital.
I am Varikuppalabharath Kumar, a graduate student pursuing a masters degree in Data Science. I started my career and am exploring a career change while gaining practical experience through a Mangusta Capital Deal Sourcing & Startup Analysis externship.
Externships
SQL & Database Architecture Externship with Breaking Games
In progress
Experience
Micro-Internship Intern, AI Consultant
Kyndryl (via Springpod) · July 2026
Education
Gisma University of Applied Sciences
Master of Science in Management —Current Grade: 1.9
South centrail Railway Degree college
Bachelor of science Overall Percentage: 76.3
Skills
Analyzed e-commerce transaction data using Python to identify sales, customer, product, and marketing trends. Performed data cleaning, EDA, feature engineering, and correlation analysis to evaluate business performance. Identified a 0.0104 correlation between advertising spend and revenue, supporting data-driven marketing decisions.
Analyzed 10,000 bank customer records to identify key factors influencing customer churn. Performed data preprocessing, EDA, categorical encoding, feature engineering, and SMOTE for class balancing. Built and evaluated Logistic Regression, Random Forest, and XGBoost models, achieving 82% accuracy and 0.84 ROC-AUC with Random Forest.
✅ Verified by Extern · ⏱️ In progress
Turn six messy CSVs into an analytics database and a Q4 dashboard that drives real business decisions. SQL + Claude.
The externship consolidated six raw CSV data sources into a structured analytics database and produced a Q4 dashboard. The work surveyed the available sales, product, and transaction files, defined the necessary schema and transformations, and implemented SQL models to normalize and join the datasets for analysis.
I turned unstructured CSV exports into a cleaned, joined analytics dataset and a dashboard-ready schema. Below are the concrete steps I completed to map the data, design the schema, and prepare queries for the Q4 dashboard.
