Self-taught Data Scientist with hands-on experience building end-to-end ML solutions in
classification, regression, clustering, anomaly detection, price optimization, and time series
forecasting. Completed 20+ project…
Self-taught Data Scientist with hands-on experience building end-to-end ML solutions in
classification, regression, clustering, anomaly detection, price optimization, and time series
forecasting. Completed 20+ projects with strong focus on EDA, feature engineering, model
tuning, explainability. Skilled at handling large-scale datasets (1M–1.7M+ rows) and integrating
ML models with Flask for real-world use.
Technical Skills
Programming: Python, SQL, Java , JavaScript, PHP
Machine Learning: Classification, Regression, Clustering, Anomaly Detection, Time Series
Forecasting, Feature Engineering, Hyperparameter Tuning (Grid/Randomized/Optuna), SMOTE,
Scaling, Outlier Handling, SHAP
Frameworks & Libraries: Scikit-learn, Pandas, NumPy, Matplotlib, Seaborn, Prophet,
Statsmodels, Joblib, Scipy
Analysis: EDA (Exploratory Data Analysis), Correlation, VIF, ANOVA, t-test, Statistical Modeling
Deployment & Tools: Flask, Jinja, HTML/CSS, Git/GitHub, Jupyter Notebook, VS Code
Key Projects
ChurnShield – Customer Churn Prediction Web App
Flask, Scikit-learn, SQLite, HTML, CSS, JS | GitHub
• Developed a Flask web app for real-time churn prediction with authentication, admin
dashboard, and CSV export.
• Integrated Random Forest pipeline with explainability via feature importance.
• Implemented retention strategy generator based on model influence.
IEEE Fraud Detection – Transaction Classification
XGBoost, Feature Reduction, SHAP | GitHub
• Processed ~1M+ Kaggle IEEE-CIS records with Spearman correlation reduction (|ρ| >
0.9) and skew-aware preprocessing.
• Trained multiple ML models; tuned XGBoost achieving ROC-AUC 0.95, Accuracy 0.98, F1
0.66 for imbalanced fraud class.
Time Series Forecasting – Supermarket Retail Prices
LightGBM, Lag/Rolling/Calendar Features, Prophet | GitHub
• Forecasted SKU-level supermarket prices (1.7M+ rows) using LightGBM and Prophet.
• Applied TimeSeriesSplit CV and log-transformed target; evaluated vi
Retail Price Optimization – Competition-Aware ML Workflow Random Forest Regression, Feature Engineering, Model Explainability • Built full ML workflow integrating historical sales, competitor pricing, and demand features. • Developed Random Forest regression models for dynamic pricing; tuned hyperparameters for optimal performance. • Explained model predictions and pricing decisions using SHAP for actionable business insights.
Isolation Forest, Local Outlier Factor, KMeans, DBSCAN, Z-Score, IQR Developed pipelines combining model-based, cluster-based, and statistical anomaly detection on fraud and retail datasets. • Evaluated using confusion matrix, ROC-AUC, precision/recall; visualized anomalies with PCA and scatterplots.
Processed ~1M+ Kaggle IEEE-CIS records with Spearman correlation reduction (|ρ| > 0.9) and skew-aware preprocessing. Trained multiple ML models; tuned XGBoost achieving ROC-AUC 0.95, Accuracy 0.98, F1 0.66 for imbalanced fraud class.XGBoost, Feature Reduction, SHAP
The AI-first Recruitment Operating System that unifies and automates the entire hiring lifecycle for modern enterprises worldwide.
Dr Job FZ LLC. 2026 © All Rights Reserved
