Data Analyst | Python | SQL | Power BI | Machine Learning Enthusiast
I help businesses turn complex data into actionable insights, optimize revenue, and drive strategic decisions. Passionate about data storytelling, predictive modeling, and building dashboards that empower decision-makers.
- Experienced in analyzing large datasets to uncover trends, anomalies, and opportunities.
- Skilled in Python, SQL, Power BI, Tableau, and data visualization.
- Passionate about building predictive models for churn, revenue leakage, and business optimization.
- Always learning and exploring the latest in data analytics and AI.
| Programming & Tools | Data Analysis & Visualization | Machine Learning |
|---|---|---|
| Python, SQL, c+ | Pandas, NumPy, Matplotlib, Seaborn | Scikit-learn, XGBoost |
| Power BI, Tableau | Excel, Power Query, DAX | Regression, Classification, Clustering |
| Git, GitHub | Data Cleaning & ETL |
- Developed a comprehensive ESG-driven supplier risk analysis framework, integrating supply chain, ESG, and geopolitical datasets (60+ features) to identify high-risk suppliers and sectors.
- Built and deployed a Random Forest model (ROC-AUC: 0.89) to predict supplier risk, enabling proactive risk mitigation strategies.
- Designed interactive Power BI dashboards to visualize sector-level risk exposure, supplier dependencies, and geopolitical disruptions for executive decision-making.
- Delivered actionable insights such as high-risk concentration in the Energy sector and strong ESG–delivery delay correlations, supporting data-driven supply chain resilience.
- Built a Revenue Leakage Detection pipeline on a 4M+ row Telecom IoT CRM dataset using Python for data cleaning, feature engineering, and KPI creation.
- Developed Power BI dashboards with segmentation filters (age, region, revenue thresholds) to visualize leakage trends and top impacted users.
- Identified $133K revenue leakage (14k users) out of $17M total revenue, uncovering 14% high-usage low-revenue customers.
- Optimized raw data from 4M+ rows to 97k active users, improving dashboard performance and enabling weekly trend analysis.
- Predicted high-risk churn customers using Random Forest & XGBoost.
- Improved retention strategies by identifying actionable patterns.
- Visualized insights using Python & dashboards.
- Tools Used : Python , Mysql
- Developed an end-to-end Power BI dashboard for transaction analysis.
- Highlighted spending patterns, fraud detection, and KPIs.
- Automated reporting for business efficiency.
- Tools Used : Power BI
More projects can be found in my GitHub Repositories
- LinkedIn: Harsh Patil
- Email: harshrp2309@gmail.com
"Data is the new oil, and insights are the refinery."