Portfolio

Projects

A collection of my full-stack development projects showcasing expertise in modern web technologies, microservices architecture, and real-world problem-solving.

Travel Planning System

Source Code

A comprehensive travel planning platform with real-time data scraping from Yatra and Kayak.

Technologies Used

Next.js Prisma PostgreSQL Zustand Puppeteer BullMQ Redis Typescript Docker

Key Features

  • ▸ Travel Planner with scraping real data from Yatra and Kayak, focusing on tours, flights, and hotels.
  • ▸ Developed Admin routes for booking, dashboard for analytics, login, logout, scraping queue and trips.
  • ▸ Developed User routes for searching tours, flights and hotels, comparing prices and booking the tours.
  • ▸ Real-time Data Scraping solution, focusing on tours, flights and hotels asynchronously.
  • ▸ Queue to handle the scraping process with Caching and storing the scraped data in PostgreSQL.
  • ▸ Authentication with JWT, handled creating jobs for queue, Integrated Stripe for booking Trips.

Learn Pulse (Ed Tech Platform)

Source Code

A full-featured educational technology platform for instructors and students.

Technologies Used

React.js Node.js Express.js Redux Toolkit MongoDB Cloudinary Typescript Tailwind CSS

Key Features

  • ▸ Backend: Integrated secure sign-up, login, OTP verification, forgot Password with email and Dynamic URL.
  • ▸ Enabled instructors to manage courses and content, while allowing students to buy courses.
  • ▸ Student: Implemented pages for Courselist, Cart, Course, Content display and Account management.
  • ▸ Instructor: Implemented Course Builder, Management with view and edit and Dashboard.
  • ▸ Admin: Implemented Category, Coupon, User management and Dashboard (gender ratio, Top selling etc).

Expense Ease (Expense Tracker)

Source Code

A microservice-based expense tracking application with AI-powered data extraction.

Technologies Used

Spring Boot Hibernate React Native MySQL ChatMistralAI Flask Kong Kafka Docker

Key Features

  • ▸ Implemented microservice architecture including Authentication, User, LLM and Expense Services.
  • ▸ Auth Service: Handling user login, signup, JWT generation and filtering, and refresh token.
  • ▸ User Service: Managing User information and employed Kafka for consuming from auth service.
  • ▸ LLM Service: Processing user messages, using the Mistral API to extract data, and storing it in the DB.
  • ▸ Expense Service: CRUD on all expenses, including those from the LLM Service and user-added entries.
  • ▸ Used Kong API Gateway to secure and route requests via JWT validation before reaching the User Service.
  • ▸ Android: App with authentication, expense management, categorization and chart for graphical analysis.

Mobile Price Prediction using Machine Learning Models

Source Code

A machine learning project predicting mobile phone prices using regression and classification models.

Technologies Used

Python Pandas NumPy Scikit-learn Matplotlib Seaborn

Key Features

  • ▸ Developed a model to predict mobile phone prices using both regression and classification models.
  • ▸ Exploratory Data Analysis to understand data distribution and relationships between features.
  • ▸ Performed data preprocessing, including cleaning, encoding categorical features, and scaling features.
  • ▸ Applied model tuning and hyperparameter adjustments to improve prediction accuracy, identifying kNN Classifier and Linear Regression as the best performing models.
  • ▸ Trained and evaluated multiple models including KNN, Linear and Logistic Regression, achieving strong results in both regression (R² = 0.72) and classification (Accuracy = 81%).