This project involved building a CNN-based music genre classification system using the GTZAN dataset. Ruthvik processed over 1,000 audio files into Mel spectrograms with Librosa and optimized the convolutional neural network using TensorFlow. To improve generalization, he implemented multiple data augmentation techniques like noise injection and time shifting. The model achieved 92% test accuracy across 10 genres. A Flask interface was developed to enable real-time prediction, maintaining latency under 2 seconds per audio clip. This project showcases skills in audio data processing, CNN modeling, hyperparameter tuning, and full-stack ML deployment, making it a great example of applying AI in multimedia analysis.

In this machine learning project, Ruthvik developed and compared three classification models—Logistic Regression, K-Nearest Neighbors, and Random Forest—to predict the presence of heart disease from patient data. Using a dataset of 900+ records, he conducted feature engineering on 14 clinical indicators including cholesterol levels, resting blood pressure, and ECG results. Through k-fold cross-validation and hyperparameter tuning, the final model achieved 89% precision. He applied one-hot encoding and MinMaxScaler for preprocessing and visualized performance using ROC curves and confusion matrices with seaborn. This project demonstrates practical application of supervised learning for healthcare risk prediction and highlights his expertise in model evaluation and data preprocessing

This data analysis project focused on understanding global COVID-19 trends using data from 190+ countries. Ruthvik used Python (pandas, seaborn) to clean, normalize, and process over 50,000 daily records, applying time-series indexing and schema unification to ensure data consistency. He created visualizations of case surges, death rates, and recovery trends using exploratory data analysis (EDA) techniques. The project also featured live API-powered dashboards capable of filtering data by date and region with sub-3 second latency. The insights helped reveal correlations between case spikes and regional policy changes. This work highlights Ruthvik’s skills in data cleaning, EDA, real-time analytics, and dashboard development.