Looking for a job

Prachi Masurkar

@prachimasurkar

Aspiring Data Scientist building machine-learning fraud detection pipelines and Power BI dashboards for financial analytics.

India

Message

I built an end-to-end financial fraud detection pipeline that processed credit card transaction records to identify high-risk anomalies. I used a Random Forest model and designed a Power BI dashboard with executive KPIs, risk trends, and PCA-based outlier analysis.

In my churn prediction project, I prepared customer data, addressed class imbalance with SMOTE, and evaluated classification models to support retention strategies. I also developed a Power BI financial forecasting dashboard with DAX measures and What-If Parameters for scenario planning.

Experience

Work history, roles, and key accomplishments

AI-Powered Financial Fraud Detection System & Analytics Dashboard

Project

Jun 2026 - Oct 2026 (4 months)

Developed an end-to-end financial fraud detection pipeline using Python, processing 10,000+ credit card transaction records to identify high-risk anomalies. Engineered machine learning classification algorithms and data preprocessing techniques to compute real-time fraud probabilities.

Python scikit learn Power BI Pandas Git Github Random Forest StandardScaler PCA

Financial Forecasting & Scenario Analysis Dashboard

Project

Jul 2026 - Sep 2026 (2 months)

Developed an end-to-end financial forecasting model using Power BI and DAX to analyze historical trends and project future revenue. Designed a multi-page interactive dashboard featuring a Star Schema data model and implemented What-If Parameters for real-time scenario planning.

Power BI DAX Star Schemas What If Parameters Financial Data Analytics Data Modeling

Telco Customer Churn Prediction & Retention Analysis

Project

May 2026 - Sep 2026 (4 months)

Developed an end-to-end machine learning classification model to predict customer churn, enabling telecom operators to proactively identify high-risk accounts. Handled missing values, engineered key account features, and applied SMOTE to resolve class imbalance issues.