Harish V Email and Phone Number

As a Research Assistant, I contributed to a project aimed at leveraging Machine Learning and artificial intelligence to address key challenges in healthcare analytics. The project involved developing predictive models to analyze large datasets for identifying trends in patient outcomes, disease progression, and treatment effectiveness. Using Python and R, I applied machine learning techniques, such as classification and clustering, to enhance data-driven insights. Additionally, I contributed to statistical modeling and data cleaning processes, ensuring data quality for model training. The project supported advancements in personalized healthcare recommendations, operational efficiency, and overall patient care quality. I also collaborated with faculty and other research team members to present findings, prepare research papers, and contribute to grant proposals.●Developed a deep learning model using CNNs in TensorFlow to classify medical images with high accuracy. ●This model, trained on over 10,000 images, achieved a classification accuracy of 92%, aiding in early diagnosis for healthcare practitioners.●Utilized machine learning to develop a predictive risk assessment tool, identifying patients with potential health risks based on demographic and clinical data. ●Conducted sentiment analysis on healthcare feedback forms using NLP techniques in Python.●Preprocessed data using tokenization, stemming, and lemmatization, and used sentiment classifiers to gauge patient satisfaction, with insights helping improve hospital services.●Collaborated on deploying machine learning models on research projects for secure, large-scale data analysis. ●Implemented Docker and Kubernetes for containerized model deployment, ensuring reproducibility and ease of scaling models in production environments.●Presented methodologies and results at academic conferences, contributing to the broader AI research community.

In this project, I was key in designing and implementing advanced data-driven solutions to address logistics challenges for CMA CGM. Leveraging machine learning, predictive analytics, and real-time data processing, I developed tools for optimizing supply chain efficiency, reducing operational costs, and enhancing decision-making in route planning and inventory management. This involved building predictive models to forecast demand, optimize shipping routes, and manage inventory, while also deploying AI-powered solutions for real-time cargo tracking and route adjustments. ●Developed machine learning models to predict demand fluctuations across shipping routes and optimize container space allocation. ●Implemented AI algorithms that analyzed historical shipment data, weather patterns, and port delays to recommend optimized shipping routes. ●By leveraging data from over 100 global shipping lanes, this initiative reduced transit times by 15% and minimized fuel costs, supporting CMA CGM's commitment to sustainability.●Led the creation of an advanced data pipeline for real-time cargo tracking, utilizing Kafka and Spark streaming to process high-frequency GPS and sensor data from vessels. ●Enabled near real-time visibility into cargo status, improving customer satisfaction and operational efficiency by providing accurate ETAs and reducing response times for potential delays.●Designed a forecasting tool that utilized statistical and machine learning models (ARIMA, Prophet) to predict inventory needs at various ports. ●Integrated data from past shipping volumes and seasonal trends, reducing stockout instances by 30% and supporting smoother logistics flow during peak seasons.●This streamlined decision-making by presenting complex data insights in an easily interpretable format, aiding in strategy formulation and operational adjustments.●Standardized data across systems, enabling accurate predictive analytics and creating a foundation for further AI-driven insights.