Built a basic proof-of-concept pipeline that uses raw voice data samples to detect cognitive stress or decline indicators using NLP + audio feature extraction. Working on speech intelligence module, processed anonymized voice clips to extract patterns that might indicate early cognitive impairment.

AI-powered application that generates highly specific and evidence-based recommendations for pet care using Groq's LLM. Features include personalized diet plans, exercise routines, grooming schedules, emergency care guides, and training tips tailored to a pet's species, breed, age, weight, and health conditions. Includes PDF report generation and health tracking over time.

Successfully created a sophisticated chatbot that delivers accurate and contextually appropriate responses based on the Bhagavad Gita. Demonstrated the effectiveness of combining retrieval systems with large language models in providing informative and relevant answers

Performed exploratory data analysis (EDA) and visualized data and created regression models like linear regression model, gradient boost regressor, decision tree regressor and random forest regressor to predict healthcare costs and correlation between costs and various reasons/factors