Data Science Portfolio

Throughout my journey in Data Science and Analytics, I have applied my skills to a variety of real-world projects, gaining hands-on experience in data extraction, processing, visualization, and predictive modeling. These projects span multiple domains, including retail, HR analytics, computer vision, and sports data analysis.


Key areas of expertise demonstrated across these projects include:

- Data Manipulation & Analysis: Pandas, NumPy, Python – transforming and preparing structured and unstructured datasets for modeling and visualization.

- Machine Learning & Predictive Modeling: Scikit-learn, XGBoost, LightGBM, SARIMA, PCA, K-Means – building and evaluating models for regression, classification, clustering, and time series forecasting.

- Feature Engineering & Data Wrangling: creating new variables, aggregations, and temporal features to improve model performance.

- Data Visualization & BI: Power BI, Tableau, Matplotlib, Seaborn – designing interactive dashboards and visualizations to extract actionable insights for stakeholders.

- Computer Vision & Image Processing: YOLOv8, OpenCV, Roboflow – developing real-time object detection pipelines for industrial and retail use cases.

- Text & Document Processing: PDFPlumber, PyPDF2, Regex, OCR – automating extraction of data from unstructured documents, including large-scale PDF processing.

- Deployment & Automation: Flask, Railway, PostgreSQL – delivering production-ready ML models and APIs for real-time predictions.

- Performance Optimization: Multiprocessing and efficient data handling techniques for large datasets.


These experiences have enabled me to handle complex datasets, design end-to-end ML pipelines, automate ETL workflows, and deliver predictive analytics solutions that support data-driven decision-making.