top of page

Projects

Online Class Bot
Python | Selenium

  • A python script to attend online classes on Microsoft Teams on your behalf.

  • The script runs automatically upon starting the PC, captures details about your schedule for the day, and starts joining your lectures based on the Microsoft Teams' calendar. The script is terminated after all the lectures for the day have ended.

what-is-digital-school.jpg

FileShare
ReactJS | WebRTC | Firebase

  • A P2P file-sharing web application for receiving and transferring files on a serverless network.

  • Provides the user with the liberty of using a server along with the benefit of faster and safer sharing of files among one’s peers.

  • The application also minimally demonstrates the conceptual working of a distributed P2P network, which is a common network architecture in blockchains and cryptocurrency networks.

image001.jpg
  • Conducted a research on above mentioned topic while interning at the Indian Space Research Organisation (ISRO).

  • Applied several data science modules and techniques using python to find, filter, spatially and temporally collocate, and analyse data generated by the SCATSAT-1, a scatterometer satellite launched by ISRO. 

  • Further applied several basic machine learning models and algorithms, and compared the outputs to rank the algorithms from worst to best.

  • Moreover, the natural parameters (such as the wind speed, ocean current, etc) that could potentially be causing the bias were also pinned down.

  • First draft of the research paper is in the works. 

SCATSat1_Auto4.jpeg
  • Prepared a linear regression machine learning model to predict the calories burnt by a person by taking their physical aspects into account. The dataset used was taken from Kaggle.

  • Compared the results and mean absolute error of the model with previous experiments executed on the same dataset. 

  • Linear regression was chosen to minimize the computational power of the model, so as to implement it into smaller and more economical devices.

cal-2.png
  • Used machine learning algorithms to detect abnormality in the spine. The dataset used was taken from Kaggle.

  • Analysed and compared multiple machine learning models to find the one with the best accuracy.

download (4).png
bottom of page