Aswin P T

Built a comprehensive platform for booking appointments, tracking pets, and managing medical records. Integrated AI-based disease prediction and CRUD operations for seamless pet care management.

Designed a deep learning model to predict RNA secondary structures from sequences using CNN and RNN architectures. Achieved good training accuracy and F1-score on validation using dot-bracket notation.

Developed a TensorFlow-based LSTM model to forecast 7-day PM2.5 levels in major Indian cities. Applied advanced data preprocessing techniques and conducted comprehensive exploratory data analysis.

Built an offline Manglish (Malayalam-English) WhatsApp chatbot using locally hosted Llama-3 model. Features real-time messaging, personality engine, and automated chat flow with emoji-rich responses.

Implemented an adaptive confidence modeling approach to enhance SHADER-based 3D human body shape estimation from monocular images. Improved accuracy and robustness in diverse scenarios.

A comprehensive web-based interactive fuzzy logic system built with Flask and Python. This toolbox provides complete support for fuzzy set operations, membership functions, relations, implications, and full fuzzy inference workflows.