• Developed and implemented state-of-the-art machine learning models for healthcare fraud detection using AWS SageMaker. Leveraged SageMaker's comprehensive suite of built-in algorithms and advanced features to train highly accurate models on large-scale healthcare datasets, including supervised learning techniques such as gradient boosting and deep learning architectures like neural networks• Developed and deployed a Django-based web application, empowering the Quality Assurance department with custom analysis reports for faster processing of customer complaints. Implemented an XGBoost model in AWS Sagemaker, optimizing analysis efficiency.• Engineered a real-time fraud detection system by deploying machine learning models as SageMaker endpoints. Integrated SageMaker endpoints seamlessly with the healthcare provider's existing infrastructure to continuously monitor incoming insurance claims data in real-time. Utilized SageMaker's scalable and reliable deployment capabilities to ensure low-latency inference, enabling timely detection and prevention of fraudulent activities

• Conducted research and analysis of more than 42 factors affecting business travel stress and performed feature engineering and factor analysis to identify the most significant factors for stress prediction.• Developed and implemented a hybrid model architecture using multiple artificial intelligence or machine learning techniques such as classification using ensemble methods like random forests and logistic regression to predict a business trip's and a traveler's stress levels based on approximately 5 million flight records.

• Led end-to-end software development lifecycle (SDLC) processes, driving requirements gathering, system analysis, design, development, testing, and deployment phases.• Engineered Python modules to seamlessly extract and load asset data from the MySQL source database, optimizing data retrieval efficiency.• Established a dedicated MYSQL database server to support web apps, enabling daily progress reporting and enhancing overall system performance.

• Developed and implemented a machine learning library on Apache Apex, a unified batch and stream processing platform for extensive data analysis based on Hadoop• Integrated MLLib, a machine learning library of Spark, into Apache Apex's engine, to enable running Spark's libraries on Apache Apex's streaming platform.• Implemented Spark's Lazy Evaluation technique in Java for Apache Apex to enhance the performance of the platform.• Designed and developed an adapter that enables running Spark's libraries on Apache Apex's engine to facilitate the integration of Spark's libraries with Apache Apex.