Nishitha Ramesh

Computer Science Engineer with expertise in Machine Learning and Intelligent Systems. Demonstrated ability in developing predictive models and optimizing data pipelines for effective data science solutions. Proficient in various ML libraries and visualization tools, driving impactful results. Seeking Data Scientist roles to implement practical ML solutions in real-world applications.

• Master Thesis - DFKI, Kaiserslautern, Investigating the Effects of Non-IID Data on Energy Consumption in Federated Learning for Object Detection, Created non-IID datasets using COCO., Built an energy monitoring package for ML pipeline., Developed a federated object detection framework using the Flower framework., Analyzed the effect of different hyperparameters, aggregation strategies, and non-iid datasets on the energy consumption of the yolov8 model in a federated setup using the custom built energy monitoring package.
• Visual Analytics Master Project, Analyzed patterns and created a dashboard to visualize illegal fishing activities., https://vast-challenge.github.io/2023/MC1.html
• Data Science Master Project, Used the CRISP-DM process to extract, clean, and preprocess COVID-19 data. Modeling using Multiple Linear Regression., Created an interactive dashboard for visualization.
• Post-Graduation Thesis, Capstone Project on life insurance sales, Developed and validated predictive models to forecast insurance agent bonuses., Performed data analysis, including univariate and bivariate analysis., Handled missing data, feature engineering and modeled it using a ML model.