My Projects

Built a multimodal deep learning framework aimed at enhancing student engagement and actionable insights in virtual classrooms. The system integrates Bi-LSTM, CNN, and Transformer-based architectures to effectively extract and analyze emotional cues—acoustic, semantic, and linguistic—from both voice and video data. Leveraged the RAVDESS and CREMA-D datasets.

Developed a comprehensive platform during a HackByte 3.0 at IIITDM Jabalpur that combines daily health monitoring (Google Fit sync, water and fasting trackers) with real-time disease risk prediction using Random Forest models. Designed intuitive UI dashboards and Flask APIs to visualize health trends, flag risks, and recommend doctor visits based on combined vitals and diagnostic results.

Built a platform to streamline the university’s document approval process by enabling online submission, tracking, and feedback—cutting approval time from days to hours and eliminating repeated physical visits.

The Rajasthan Club Website, built with React and Tailwind CSS, showcases Rajasthan's culture, heritage, and events. This responsive, user-friendly platform offers an immersive experience, connecting enthusiasts to traditions, tourist spots, and community activities for our college club, blending creativity and functionality seamlessly.

Engineered a Bi-LSTM model to detect early Parkinson’s disease from speech, achieving 93% average accuracy with strong F1-score and precision on benchmarked datasets(with 195 entries).

we're redefining financial advisory services with cutting-edge generative AI technology. Our mission is to democratize personalized financial advice, provide real-time insights, ensure transparency, and scale efficiently to meet diverse user needs.

An Attendance Tracker using Facial Recognition automates attendance marking by identifying individuals through facial features. It uses OpenCV (cv2) for real-time face detection and recognition, comparing detected faces with a pre-registered database. Recognized individuals are logged into a database with details like name, date, and time. The system features a tkinter GUI for user interaction, allowing attendance tracking, record viewing, and adding new faces.

A Fintech Anomaly Detection & Privacy Protection System! Our mission was to elevate security and privacy in fintech.