My Projects

An intelligent computer vision system that analyzes football match footage to detect players, identify teams, and track passes in real-time. The system uses YOLOv8n for accurate player and ball detection, then applies color-based clustering to distinguish between teams.

The project demonstrates advanced machine learning techniques including deep learning for object detection, K-means clustering for jersey color extraction, and custom heuristics for identifying referees and goalkeepers. It processes video frames, annotates them with bounding boxes, and outputs analyzed clips for further review.

An intelligent customer support automation system for "Abby Sports," featuring both a web-based chatbot and automated email responses. The system uses a 6-page knowledge base as its single source of truth, ensuring consistent and accurate responses across all channels.

Built with n8n for workflow automation and powered by Google Gemini AI, the system automatically filters business-relevant inquiries and responds with policy-compliant answers. Email responses are formatted in clean HTML for professional presentation, while the chatbot interface can be embedded on any website via webhook.

An interactive business intelligence dashboard built in Power BI to analyze retail performance across multiple dimensions. The dashboard provides executives and managers with real-time insights into profit trends, sales by category, and product performance.

Features dynamic filtering capabilities, drill-down analysis, and key performance indicators that update automatically. The clean, intuitive layout makes complex data accessible to non-technical stakeholders, enabling data-driven decision making.

A comprehensive data analysis project focused on Q4 sales performance, combining Python for data cleaning and transformation with Power BI for executive-level visualization. The project demonstrates end-to-end data analytics workflow from raw data to actionable insights.

Used Pandas for data manipulation, applied outlier detection rules, and created regional and categorical visualizations with Seaborn. The final deliverable was a concise Power BI summary page designed for stakeholder presentations.

A Python-based backend system for an e-commerce platform, implementing core services including authentication, shopping cart management, and product catalog. The project showcases object-oriented programming principles and clean code practices.

Built with proper input validation, comprehensive docstrings, and unit tests for core functions. Includes local run instructions with sample data, making it easy for other developers to understand and extend the codebase.