-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

This project presents a web-based online shopping platform designed to improve user experience through advanced product discovery and seamless transactions. Built using the MERN stack (MongoDB, Express.js, React.js, and Node.js), the platform includes features such as product filtering by price and category, search functionality, personalized product recommendations using content-based filtering, and secure payment processing. These features help users easily find relevant products and complete purchases efficiently. The project demonstrates how modern web technologies can create a scalable, user-friendly, and engaging e-commerce platform that enhances customer satisfaction and product exploration.

The Online Library Management System (OLMS) is a web-based platform developed to modernize and simplify library operations in educational institutions. Built using HTML, CSS, JavaScript, and PHP, the system enables users to search, reserve, and check out books online through a user-friendly interface. It also provides administrative functionalities such as book cataloging, user management, and real-time book availability updates. By improving accessibility, efficiency, and digital resource management, the OLMS serves as a scalable solution suitable for academic libraries, public libraries, and other knowledge-based organizations.

The Nepali Sign Language Recognition System is a real-time machine learning application developed to reduce the communication gap between speaking and non-speaking or non-hearing communities. Unlike many existing systems that focus on globally recognized sign languages, this project specifically targets Nepali Sign Language. Using MediaPipe for hand landmark detection and a Feedforward Neural Network (FNN) model for gesture classification, the system accurately recognizes hand gestures and converts detected text into speech using Google’s Text-to-Speech API. The model achieved a validation accuracy of 97.43% across 60 gesture classes, demonstrating reliable performance. This system has practical applications in teaching, learning, and promoting inclusivity for the deaf and non-speaking community in Nepal.

The Courier Management System is a logistics platform developed to improve delivery operations through automation and intelligent route management. The system addresses common challenges such as delayed deliveries, inefficient routing, and limited delivery visibility by integrating features like automated order processing, real-time tracking, courier assignment, and secure payment handling. Using Dijkstra’s Algorithm, the system calculates the shortest and most efficient delivery routes, helping assign nearby couriers based on real-time distance and availability. Built with open-source technologies, the platform offers a user-friendly interface for managing deliveries efficiently while reducing travel time and operational costs. The project demonstrates a scalable and data-driven approach to modern courier and logistics management.

This project presents a comprehensive ride-sharing platform designed to connect riders and drivers through a secure, efficient, and user-friendly system. The platform integrates features such as ride coordination, user and driver verification, real-time communication, flexible ride scheduling, and review systems to enhance trust and safety. By utilizing the K-Nearest Neighbors (KNN) algorithm along with a dynamic pricing model, the system improves ride matching and maintains a balanced experience for both riders and drivers. The platform supports both immediate and future ride bookings, offering a scalable and reliable solution for modern urban transportation services.

The Secure Real-Time Chat Application is a privacy-focused messaging platform developed to ensure secure and reliable communication in the digital age. The system integrates advanced cryptographic techniques, including AES encryption for message confidentiality, RSA encryption for secure key exchange, and HMAC for message authentication and integrity verification. Alongside strong security mechanisms, the platform offers practical communication features such as channel-based messaging, direct messaging, file sharing, audio messaging, customizable profiles, status indicators, and audio/video calling. Designed for both personal and professional use, the application provides a secure and user-friendly environment for real-time communication while maintaining data privacy and authenticity

EduForge is a modern Learning Management System (LMS) developed to simplify online course delivery and educational management. Built using the MERN stack, TypeScript, and Next.js, the platform provides educators with a secure and efficient way to manage courses while offering students an organized and interactive learning experience. Key features include course management, admin dashboards, secure authentication through NextAuth, DRM-protected video streaming using Videocipher, Redis caching for performance optimization, and Razorpay payment integration. Designed for scalability, reliability, and usability, EduForge serves as an effective educational platform for individuals and small educational institutions seeking a streamlined online learning solution.

EasyHome is an online home services platform developed using Django, HTML, CSS, and JavaScript to simplify access to essential services such as plumbing, electrical repairs, and cleaning. The platform addresses inefficiencies in traditional service booking methods by offering a user-friendly system that connects customers with verified service providers in a fast and reliable manner. It uses recommendation techniques such as K-Nearest Neighbors (KNN) and content-based filtering to match users with the most suitable professionals. With added features like secure payment integration and a streamlined interface, EasyHome enhances convenience, transparency, and service quality for urban households.

The Personalized Healthcare System is an intelligent health management platform designed to predict possible diseases based on user-reported symptoms. It uses the Support Vector Classifier (SVC) machine learning algorithm to classify symptoms and recommend potential medical conditions along with precautions, medications, diet plans, and exercise suggestions. The system also helps users locate nearby hospitals relevant to their predicted condition using geolocation features, ensuring timely access to healthcare. Additional features include medication scheduling with email reminders and a community blog for sharing health insights. Overall, the platform enhances proactive healthcare by combining disease prediction, personalized recommendations, and accessible medical support.

The Food Ordering Web Application is a full-stack platform designed to simplify online food ordering for customers and restaurant owners. Built using Spring Boot for the backend, React for the frontend, and MySQL for data storage, the system provides a smooth and responsive user experience. Customers can register, browse restaurants, explore menus, manage their cart, and securely complete payments using Stripe. On the backend, restaurant data, categories, and food items are managed through REST APIs, with Postman used for administrative operations. The application emphasizes scalability, performance, and real-time interaction, ensuring an efficient and user-friendly food ordering experience.

The Online Bike Rental System is a web-based platform designed to improve traditional bike rental services by addressing issues such as manual booking, limited accessibility, and inefficient inventory management. It provides a streamlined interface where users can search, book, and manage bike rentals, while administrators can oversee inventory and reservations efficiently. To enhance user experience, the system incorporates a cosine similarity algorithm that analyzes user browsing behavior and recommends bikes with similar features to previously viewed options. This personalized recommendation approach improves accuracy in suggestions, increases user satisfaction, and optimizes the overall booking process. The project demonstrates how data-driven algorithms can enhance efficiency and personalization in rental services.

LearningMate is a modern Learning Management System (LMS) designed to simplify course management and improve student engagement through a seamless digital learning experience. Built using Next.js, React, Prisma, and MongoDB, the platform ensures scalability, performance, and secure data handling. It integrates Stripe for secure payments, Mux for high-quality video streaming, and robust authentication for user access control. The system also includes personalized course recommendations and engagement insights to make learning more interactive and effective. With a responsive UI powered by Tailwind CSS, LearningMate provides an efficient, all-in-one solution for educators and institutions seeking a scalable and user-friendly e-learning platform.

The College Recommendation System is a web-based platform designed to simplify the college selection process in Nepal, where students often struggle with fragmented and incomplete institutional information. It provides a unified system that aggregates accurate college data and offers personalized recommendations based on student preferences and qualifications. The system uses web scraping and hybrid recommendation algorithms to suggest suitable colleges, helping students make informed decisions. It also includes features such as college reviews, admission applications, and online payment integration, making the entire process more efficient and streamlined. Overall, the platform enhances transparency, accessibility, and user experience for both students and educational institutions.

This project focuses on automated tennis player and ball tracking to support performance analysis and training. It uses computer vision techniques to detect players and the tennis ball, enabling measurement of key metrics such as player speed, ball shot speed, and shot count. YOLO (You Only Look Once) is used for real-time object detection, while Convolutional Neural Networks (CNNs) help extract court key points for accurate tracking and motion analysis. The system processes video input to provide detailed insights into player movement and shot dynamics. This solution can assist coaches, analysts, and players in improving strategies and refining performance through data-driven analysis.

The School Management System (SMS) is a centralized digital platform designed to streamline academic and administrative operations within educational institutions. It integrates multiple modules such as attendance tracking, class and batch management, front office operations, transportation, hostel, inventory, notices, and live meetings into a unified system. The platform enhances efficiency by reducing manual workload and improving data accessibility for students, teachers, and administrative staff. It also incorporates role-based access control, biometric attendance, and analytics for better decision-making. Additionally, collaborative filtering is used to provide personalized recommendations in areas like course selection and resource allocation. Developed using the Spiral Model, the system emphasizes iterative improvement, high performance, security, and scalability.

The Civic Issue Management System (CIMS) is a web-based platform designed to improve the reporting and resolution of urban civic problems such as potholes, waste management issues, and streetlight failures. It enables citizens to report issues using geotagged photos and detailed descriptions, making complaint registration more accurate and transparent. City officials can track, prioritize, and manage these reports efficiently through an interactive dashboard with real-time updates and trend analysis. The system also includes automated notifications to improve communication between citizens and authorities. Overall, CIMS enhances transparency, efficiency, and civic engagement, contributing to smarter and more responsive urban governance.

This project presents an advanced travel recommendation system designed to overcome the limitations of traditional keyword-based approaches. It uses semantic analysis, GraphRAG (Graph Reasoning with Augmented Graphs), and large language models (LLMs) to generate personalized and context-aware travel suggestions. The system is built using Python and integrates the Llama-3.2-1b model via Ollama for entity extraction and query processing. Travel data is cleaned, vectorized using the Nomic embedding model, and stored in a Neo4j graph database to enable intelligent multi-agent reasoning, including query refinement and summarization. The system improves recommendation accuracy, highlights lesser-known destinations, and supports sustainable tourism through AI-driven insights.

This project introduces a freelance service platform designed to enhance the user experience by streamlining the process of connecting freelancers with clients and offering personalized service recommendations. Using a content-based filtering algorithm powered by cosine similarity, the platform matches clients with freelancers based on individual preferences, project history, and expertise. With a focus on simplicity and security, the application features a user-friendly interface, secure payment processing, real-time chat, and project management tools to facilitate clear communication and efficient collaboration. The platform also includes a rating and review system to foster trust and transparency within the freelance community. By integrating these features, this project aims to create a seamless freelance ecosystem that empowers freelancers and clients alike, improving engagement, satisfaction, and the effectiveness of project partnerships.

This project focuses on addressing the issue of frequent hotel booking cancellations, which often lead to operational disruptions and revenue loss in the hospitality industry. It proposes a predictive system that analyzes historical booking data to forecast the likelihood of cancellations using machine learning techniques, specifically Decision Tree and Random Forest algorithms. The system is implemented as a web-based application with separate interfaces for customers and administrators, enabling users to make bookings while allowing hotel managers to monitor cancellation risks and optimize room allocation. Experimental results show that the model effectively identifies cancellation patterns, helping hotels take proactive measures to reduce uncertainty and improve financial performance.

This project presents a web-based language learning platform designed to improve learner engagement and efficiency through personalization and gamification. It addresses common challenges in language learning, such as lack of feedback, difficulty identifying weaknesses, and low motivation. The system tracks user performance, categorizes mistakes, and provides tailored practice recommendations to focus on individual learning gaps. Built with Next.js for frontend and server-side rendering, PostgreSQL (via Drizzle ORM) for data management, and ElevenLabs for text-to-speech pronunciation support, the platform delivers an interactive learning experience. Gamification features such as points, hearts, and a leaderboard further enhance motivation and encourage consistent practice, making the system suitable for both individual learners and educational institutions.

This project introduces a centralized system for monitoring and optimizing the management of horticultural resources in the agricultural sector. It addresses the imbalance between supply and demand, which often leads to financial losses for farmers and higher costs for consumers. The system supports local governments in tracking resources through real-time data visualization and predictive analytics. It employs Linear Regression for forecasting and K-Means clustering for resource categorization, enabling better planning and distribution. By integrating data-driven techniques, the platform improves decision-making, reduces waste, and enhances overall market efficiency. The system demonstrates how predictive modeling and clustering can significantly improve agricultural resource management.