From the trading desk
to the data desk.

A beginner-to-intermediate SQL project: database setup, data cleaning, exploratory data analysis, and business-question queries — top customers, category-wise sales, monthly trends, and shift-based order patterns.

An intermediate SQL project modeling a library system — branches, employees, members, books, and issue/return tracking. Covers CRUD operations, CTAS, stored procedures, and advanced queries like overdue-fine calculation and branch performance reporting.

SQL analysis of Netflix's content catalog — content-type distribution, most common ratings, top countries by content volume, genre breakdown, top directors and actors, and keyword-based content categorization.

‹ ›

An end-to-end Python + SQL pipeline on Walmart sales data — cleaned and transformed with Pandas, loaded into MySQL and PostgreSQL via SQLAlchemy, then queried with complex SQL to surface revenue trends, top categories, and profit margins by branch.

A dashboard built from my own trade logs — win rate, drawdown, P&L distribution, and risk-adjusted return.

‹ ›

An end-to-end credit risk pipeline on the German Credit Dataset (1,000 customers) — EDA, categorical encoding, an Extra Trees classifier predicting loan default probability (~77% accuracy, ROC-AUC ~0.80), and a Streamlit app for real-time risk scoring.

A Power BI dashboard analyzing Facebook and Instagram ad campaigns — platform-specific views, core KPIs (impressions, reach, CTR, CPC, conversions), and demographic/engagement breakdowns to guide ad spend and targeting decisions.

Completed a forensic-technology data analysis simulation — built a Tableau dashboard and used Excel to classify data and draw business conclusions.