Jay's Projects

Some cool things I worked on

Barcode scanning Attendance Tracker

2024 August

OpenCV • Python • Pandas • Openpyxl

  • Designed and developed a barcode scanning sign-in/out system using Python, OpenCV, and Pyzbar to track attendance for a 35–40 member robotics team.
  • Automated attendance logging into Excel spreadsheets with Pandas and OpenPyXL, including sign-in, sign-out, and hours calculation per student.
  • Built a time-tracking module to calculate session durations and accumulated hours per student.
  • Implemented real-time performance monitoring (CPU/memory usage) with psutil for system reliability during continuous use.
  • Enhanced user experience with audio feedback on successful or failed scans and real-time webcam display of scanned results.
  • Delivered a low-cost, efficient alternative to manual attendance tracking, reducing errors and improving student accountability.
View Repo →

Real-time American Sign Language (ASL) Alphabet Translator

2024 December

OpenCV • TensorFlow • MediaPipe • pyttsx3

  • Built a real-time ASL translation system using Python, OpenCV, and cvzone to detect hand gestures and classify them into ASL letters.
  • Integrated a Keras deep learning model to achieve accurate classification of alphabet signs from live webcam input.
  • Implemented image preprocessing techniques (cropping, resizing, normalization) to improve model performance under varying lighting and angles.
  • Added text-to-speech functionality (pyttsx3) to convert recognized signs into spoken language for seamless communication.
View Repo →

Employee-Management API (Backend)

2025 August

Spring Boot • Java • Spring Data JPA • Docker • PostgreSQL

  • Developed a RESTful backend service using Java, Spring Boot, and PostgreSQL to manage employee data with CRUD functions.
  • Containerized the PostgreSQL database with Docker, simplifying setup and improving deployment consistency.
  • Integrated Spring Data JPA/Hibernate for seamless persistence, automatic schema updates, and efficient query handling.
  • Addressed the challenge of unreliable manual employee record management by providing a scalable, automated solution with reliable database-backed APIs.
  • Implemented a layered architecture (Controller → Service → Repository) with DTO mapping to ensure clean separation of concerns and maintainable code.
View Repo →
Separator