Hey, I'm Namith

A comprehensive property listing platform built with the MERN stack featuring separate buyer and seller portals. Implemented advanced search functionality with pincode-based filtering, secure image upload system using Multer, and real-time property updates. Features include user authentication with JWT, responsive UI with React hooks, MongoDB database for scalable data storage, and RESTful API architecture. Deployed on Vercel with optimized performance and SEO.

Enterprise-grade placement management system with role-based access control for administrators, students, and placement coordinators. Built with Spring Boot backend featuring microservices architecture, JPA for database operations, and Spring Security for authentication. React frontend with Redux for state management, Material-UI components, and interactive dashboards with Chart.js. Features company registration, drive scheduling, student applications, and comprehensive analytics. MySQL database with normalized schema and optimized queries.

Full-stack event services platform connecting customers with service providers. Angular frontend with TypeScript, RxJS for reactive programming, and Angular Material for UI components. Spring Boot REST API with PostgreSQL database, implementing geolocation-based search to find nearby service providers. Features include worker registration portal, service booking system with real-time availability, payment gateway integration, booking history, and rating system. Implements JWT authentication, input validation, and comprehensive error handling.

Machine learning-powered property valuation system using Linear Regression model trained on extensive real estate dataset. Flask backend serving ML model with Joblib serialization, featuring data preprocessing pipeline with StandardScaler and feature engineering. Users input property details (location, size, bedrooms, amenities) and receive instant price predictions with confidence intervals. Integrated with the real estate platform to provide sellers with market insights and profit estimates. Model achieved 87% accuracy through hyperparameter tuning and cross-validation. Deployed on Render with API documentation.

Backend-focused expense management system implementing full CRUD operations with category-based filtering and analytics. Designed modular REST API architecture with middleware-based validation and error handling. Implemented aggregation queries for monthly spending insights and optimized MongoDB indexing for performance. Built production-style routing structure and scalable schema design.

Developed a comprehensive deep learning model for classifying toxic comments in online platforms. The project involved building a robust NLP pipeline that included text preprocessing, tokenization, and sequence padding. Implemented an LSTM-based neural network to accurately classify comments as toxic or non-toxic. The model was trained and evaluated using a large dataset, achieving high accuracy and performance. Deployed the model using Streamlit, enabling real-time toxicity detection with an intuitive user interface. This system ensures scalable, automated moderation of online content.