A collection of projects showcasing my work in machine learning, data science, and full-stack development.
Solving the EEG Inverse Problem with deep learning methods for enhanced brain activity localization
A deep learning system for estimating brain activity locations from scalp EEG signals using a hybrid CNN-Transformer architecture with efficient synthetic data generation.
Enhancing PatchTST with selective multi-scale patching to capture rich temporal dynamics
A Transformer-based time series forecasting framework that extends PatchTST by processing data at multiple temporal resolutions in parallel and fusing scale-specific predictions through a learned mechanism.
Accurate, scalable bus ETA forecasting using spatio-temporal deep learning
A real-time bus arrival time prediction system for New York City that leverages large-scale spatio-temporal data from MTA buses and state-of-the-art Transformer-based forecasting models to improve public transportation reliability.
A full-stack platform for seamless doctor-patient appointment scheduling
A web-based appointment management system that enables patients to book medical appointments online and allows doctors to manage schedules, availability, and patient interactions through a unified platform.
Data-driven match outcome prediction using historical IPL analytics
A machine learning-based analytics system that explores historical Indian Premier League (IPL) data and predicts match winners for the 2025 season using supervised learning models and interactive visualizations.
A comprehensive system for employee, leave, and payroll management
A full-stack Human Resource Management System developed to digitize and streamline core HR operations such as employee management, leave handling, salary reporting, and access control.
A compact FPGA-based processor designed using VHDL
A custom-designed 4-bit Nano Processor developed using VHDL and Xilinx Vivado, focusing on fundamental processor architecture, instruction execution, and machine-level programmability.
