Data enthusiast with strong background in statistical analysis, machine learning, and data visualization. Skilled in Python, R, SQL, and various data processing tools, with focus on delivering actionable insights. Known for collaborative approach, adaptability, and consistently achieving impactful results in dynamic environments. Recognized for problem-solving abilities and innovative thinking in leveraging data to drive business decisions.
• Conducted comprehensive item-level RFM and ABC-XYZ analysis for a manufacturing client to drive product segmentation and sales optimization.
• Developed automated data pipelines using Python and SQL for seamless data extraction, transformation, and loading (ETL), ensuring accurate and consistent forecasting inputs.
• Ensured full coverage of A-category items in predictive workflows to support high-value sales forecasting.
• Designed and deployed an interactive Shiny dashboard to visualize key sales metrics and inventory classifications, enabling real-time insights for business stakeholders.
• Applied core MLOps principles such as automation, reproducibility, and modular architecture to enhance maintainability and scalability of the solution
• Assisted in developing machine learning models for classifying X-ray sources, focusing on time-based feature extraction and analysis of light curve characteristics.
• Contributed to creating a classification system to enhance the understanding and analysis of compact celestial objects
Stock Dashboard using Finnhub API, Kafka & Superset
• Built a real-time data pipeline using Apache Kafka to stream live stock data from Finnhub API, enabling scalable and fault-tolerant data ingestion.
• Developed Kafka Producer-Consumer architecture for processing and storing real-time market data, forming the backbone for ML model input and retraining.
• Integrated Apache Superset for real-time visualization, supporting monitoring of stock trends, data quality, and potential data drift.
• Designed the system for modular deployment, supporting CI/CD pipelines and future integration with ML model inference and retraining workflows
BMTC Passenger Flow Analysis | Real-Time Data Pipeline & Predictive Analytics
• Designed and implemented a real-time data ingestion pipeline using Apache Kafka to stream passenger and bus movement data for BMTC operations.
• Developed data preprocessing and aggregation workflows, enabling downstream machine learning models to identify passenger flow patterns and demand forecasting.
• Established a robust PostgreSQL data storage layer to support scalable, reliable data persistence and accessibility for analytics.
• Created interactive Apache Superset dashboards for real-time visualization of route utilization, traffic peaks, and predictive insights, enhancing operational decision-making.
University Chatbot
• Developed a conversational AI chatbot using the Rasa framework, enabling intent classification and entity extraction for user queries.
• Integrated the Gemini API to generate dynamic, context-aware responses, enhancing natural conversation quality.
• Designed a JSON-based knowledge base to handle structured university information (establishment, campus, leadership, resources).
• Implemented NLP-driven dialogue flows to improve response relevance and user engagement.
• Built with modular architecture, supporting scalability
Stock Price Prediction App
• Built a real-time stock prediction app using Python and Streamlit, fetching live data from Yahoo Finance.
• Integrated data preprocessing, prediction logic, and time-based behavior (uses today's price post 4PM, else yesterday’s).
• Developed an interactive UI where users input a stock ticker to get next-day price predictions instantly.
• Modularized the codebase following MLOps best practices for scalability and maintenance.
• Enabled easy deployment and sharing without external servers or frameworks