JITHIN M

• Developed an autonomous navigation system using reinforcement learning, leveraging ROS, and Gazebo.
• Built multiple chatbots for sentiment analysis, question-answering, and text summarization using Hugging Face Transformers, fine-tuning pre-trained models like BERT, RoBERTa, and DistilBERT, deployed via a user-friendly Streamlit interface.
• Developed specialized dashboards for Jaguar Land Rover (JLR), leveraging BigQuery, Tableau, and Python to empower Subject Matter Experts (SMEs) in monitoring and addressing system deviations effectively.
• Designed a navigation page in Tableau Online to streamline access to multiple dashboards, enhancing user experience, and efficiency.
• Developed multiple machine learning models for MHEV (Mild Hybrid Electric Vehicle) DC-DC fault prediction, leveraging one-class SVM, and various binary classification techniques. The models were designed to predict potential DC-DC converter failures, a crucial component in electric and hybrid vehicles, by analyzing operational data and identifying early signs of failure.

Amazon Customer Review Sentiment Analysis.

• Developed a sentiment analysis system for Amazon customer reviews by first web scraping reviews using BeautifulSoup to extract relevant data from product pages.
• The scraped data was then converted into a pandas DataFrame for efficient processing and analysis. Leveraged state-of-the-art BERT and DistilBERT models to classify the sentiments of customer reviews into positive, negative, or neutral categories.
• The project involved fine-tuning the pre-trained models to enhance performance on sentiment classification, enabling businesses to gain valuable insights into customer opinions and feedback at scale.