Kulvir Singh

Engineer by profession

I am a software engineer and I love to build cool stuff.

About

I have been enthusiactic about technology from very young age, Hence here I am pursuing engineering from Thapar Institute of Engineering and Technology. Here, I also had the oppurtunity to co-author a research paper in the field of Computer Vision.

Other than that Web has been a great interest of mine. I have worked on various web applications built mostly using ReactJS, NextJS and Tailwind CSS. This tech stack allowed me to discover good of both client side and server side worlds. Although creating seamless and scalable UI's has been my motivation but the more I work on backend side, the more I am attracted to it. Therefore I am currently learning DevOps delving deep into the world of servers.

When I'm not in front of my PC coding, I'm usually watching some anime or probably some movies. Also, I like to read some books mostly classic fiction.

Projects

  • Dicsword | Dicsord Clone

    Developed a Discord clone where you can create servers and different channels within server and also manage server roles for each member. Realtime chat was enabled with websocket with long polling as a fallback method.

    • Next.js
    • Tailwind
    • Socket.io
    • PostgreSQL
    Dicsword
  • OA Helper

    An AI application that provide students answers for their online test questions. Answers are available in two formats: C++ Code or text based. Each user is allowed 25 free requests only.

    • Next.js
    • OpenAI API
    • Prisma
    • Stripe
    OA Helper website
  • Booklog | Personal bookshelf

    Web app specially built for book readers to keep track of their reading progress and discover new books. Users can create different Book shelves and share it with other users. Google books API was used vast collection of books. Google OAuth2.0 was used for secure and fast login process.

    • React.js
    • Node.js
    • AWS DynamoDB
    • Chakra UI
    Booklog homepage

Other Works

  • Implemented a Autoencoder neural network to denoise images. MNIST dataset was used for training of the model. They work by compressing the input into a latent-space representation, and then reconstructing the output from this representation

    • PyTorch
    • Python