Atharva Singh

Master's in Computer Science (Data Science) from Leiden University with expertise in Python, SQL, and cloud platforms like AWS and Google Cloud. Specializes in transforming complex data into actionable insights through advanced analytics, data engineering, and AI solutions. Passionate about solving real-world problems with technology, excelling in dynamic, creative settings, and delivering meaningful outcomes as a data expert.

Data Scientist

• Improved RMSE by 90% using XGBoost for Walmart store forecasting.
• Enhanced API end-user access by developing RESTful APIs for web applications.

• Financial Health Analysis of European IndustriesConducted a comprehensive comparative analysis of post-COVID and post-Brexit financial resilience across European industries. Evaluated key financial ratios, tax structures, cost of capital, and operational KPIs to identify recovery trends and vulnerabilities.
• Real-Time Face Mask Detection SystemBuilt and deployed a live video classification pipeline in Python to detect face mask usage. Demonstrated real-time data ingestion, transformation, and classification capabilities using OpenCV and pre-trained models.
• Bankruptcy Prediction ModelingApplied Logistic Regression and Random Forest algorithms to forecast corporate bankruptcy using Taiwanese economic data (1999–2009). Optimized model accuracy through feature selection and hyperparameter tuning.
• Indian Vehicle Detection Using YOLODeveloped a YOLO-based object detection model to identify the five most frequently occurring vehicle types on Indian roads. Trained on custom datasets to enhance detection accuracy in diverse real-world conditions.
• Hospital Chatbot with GUIDesigned a patient-assistance chatbot with a graphical interface to help users access medical reports and resolve queries. Focused on accessibility and ease of use for both patients and hospital staff.
• Movie Revenue PredictionBuilt a regression-based model to predict box office success using production budget and metadata. Highlighted ROI estimation in pre-release planning for film production houses.
• Online News Popularity AnalysisAnalyzed over 39,000 news articles to identify content features influencing popularity. Applied statistical methods and exploratory data analysis to derive insights into reader engagement metrics.