Hi, I'm Siya P Kurandwad

Associated with RV College Of Engineering

This project presents a robust and extensible system for detecting malicious activities in network traffic. It employs an algorithmic approach, combining dynamic graph modeling, subgraph isomorphism for pattern matching (both baseline and incremental), and a GCN for advanced classification. The system processes network flow data to identify and analyze suspicious behaviors.

Developed a Graph Neural Network (GNN) based network intrusion detection system. Utilized causal sampling techniques to detect and classify cyber threats with improved efficiency and accuracy.Trained 4 attck-specific models on attacks - Phishing,Malware, IOT network and DDOS.A global model was trained using Fusion Classifier to learn from the 4 attck-specific models for generalized prediction and broad range detection.

Neuro Track is an AI-powered system for early-stage Alzheimer's disease prediction and patient monitoring. Leveraging GNN model, the system analyzes patient data such as genetic markers, PPI interactions and brain scans (MRI/CT), to provide early insights into the likelihood of developing Alzheimer's. Trained th GCN model on several important layers like gene co-ecpression, PPI network and neuroimaging.

Associated with ACCELERATE RVCE

Built a web-based tool to fetch live METAR/TAF aviation data, parse flight plans, analyze flight risks based on weather and route conditions, and visualize them using interactive maps and alert summaries. Participated in Honeywell Hackathon to work on this problem statement.

Associated with Google Solutions Challenge & Google Developer Groups RVCE

Developed a secure digital voting platform leveraging blockchain for vote integrity, biometric/Election ID verification for authentication, and AI for fraud detection, ensuring transparency and accessibility in elections.

Re Cloudify enables users to repurpose old laptops as remote storage devices, creating a personal cloud. Using a client-server model with socket programming, data can be seamlessly stored and accessed over a network. This decentralized approach offers cost-effective, private, and expandable storage without relying on third-party cloud providers.

A smart emergency response application that integrates real-time ambulance location tracking, hospital bed availability, and emergency medical services (EMS) alerts. Designed to streamline communication between ambulance teams, hospitals, and emergency responders by reducing response times and optimized routing, it aims to reduce delays in emergency care.

Implemented a decentralized weather forecasting model using edge AI and federated learning. The system adapts to local data while preserving privacy, improving prediction accuracy without centralized data collection. Implemented a global model to learn from local models using Federated Learning by sharing model weight and aggreagtes and preserving data privacy and also improving accuracy.

Associated with RVCE WiC Insider Circle

Developing and maintaining the official website for Women in Cloud COE RVCE dedicated to promoting and showcasing initiatives for women in cloud technology. The site serves as an informational platform that highlights community stories, upcoming events, and collaborations in the cloud ecosystem.

Associated with EECS @ IISc, Bengaluru

Worked as a Frontend Developer and built the WIPOD Car dashboard for the WIPOD (WIPRO Autonomous Vehicle). Worked mainly on the Diagnostics screens and the user centric screens like opening screens and entertainment features.

IoT-based smart water management system for real-time monitoring and efficient usage tracking. The system integrates sensors and microcontrollers to collect water consumption data, which is visualized through a Next.js-based dashboard. Features include automated alerts, usage analytics, and conservation recommendations, enabling data-driven water usage decisions.

Created an automated refill system for lead-acid batteries that monitors electrolyte levels and refills distilled water as needed, reducing manual maintenance and enhancing battery life.