Rajarshi Dutta

Rajarshi Dutta

Final Year Undergraduate

Indian Institute of Technology Kanpur

Biography

I am currently a Final Year Undergraduate at Indian Institute of Technology Kanpur. I am deeply interested in Deep learning,NLP, Optimization techniques and its applications to solve complex problems. I also had a decent exposure on other topics as well like Web Development, solving CTFs, Scripting, Cryptography, Robotics. Recently, I like to delve into the concepts behind workings of LLMs and other models. I will be found playing sudoku and watching anime during my spare time. My other hobies apparently includes travelling, photography.

I was also a Team Head at Aerial Robotics IITK which is group of students motivated towards building autonomous quadcopters. I was also somewhat associated with the Astronomy Club at IITK during the Inter IIT Tech Meet 10.0 and 11.0 :P.

I would love to start a random convo with others so feel free to contact me below!

Interests
  • Machine learning & Optimization
  • Control Systems
  • Autonomous Systems
Education

  • BTech in Materials Science and Engineering, 2020 - Present

    Indian Institute of Technology Kanpur

Work Experience

 
 
 
 
 
Univeristy of Toronto Scarborough
MITACS Summer Research Intern
Univeristy of Toronto Scarborough
May 2023 – August 2023 Toronto, Canada
Leveraging state of the art Graph Neural networks for efficient quantum chemical screening of inorganic and organic materials. Focus on implementing GCN and other Message passing networks, optimizing CGCNN for accurate bandgap and orbital energy prediction of wurzites and rocksalt based crystal structures. Also involved developing models for regression analysis on Morgan fingerprints of organic XTB dataset.
 
 
 
 
 
Indian Institute of Technology Kanpur
SURGE
Indian Institute of Technology Kanpur
May 2022 – July 2022 Kanpur, India
Implemented Contour Detection on TEM images for nanoparticle detection and creation of training set for YOLOv3 Used image segmentation methods like K-Means Clustering, GMMs for foreground isloation from grayscale images. The project involved utilizing AutoDetect-mnP framework which is based on an unsupervised clustering based approach.

Projects

*
All ML & Data Robotics Image Processing Astronomy Quantum
Orbital Dynamics Challege
Cosmic Innovation Challenge | Shastra'23 | Developed a real time calculator for inferring satellite coordinates and also deployed a web app
Code Slides Web
ISRO’s Chandrayaan Moon Mapping Challenge
Inter IIT 11.0 Problem Statement from ISRO | Developed AI/ML model for generating high-resolution map using Chandrayaan’s Orbiter’s imaging payloads
Code Atlas
Model Zoo
SNT Summer Project | Implemented several SOTA models in areas like Image classification and Action Recognition
Code
Range based Control law for trajectory generation using Unicycle Model
Under Graduate Project | Dr. Twinkle Tripathy | Implemented trajectory generation of a differentiable holonomic drive in ROS and Gazebo
Code
Quantum Algorithms
Stamatics Project 2022 |Studied concepts from Quantum information and implemented several quantum algorithms in Qiskit
Project Page
ISRO’s Web Based X-Ray Burst Identification
Inter IIT 10.0 Problem Statement from ISRO | Developed a end to end application and Python Package for detection and classficaion of Solar Flares
Code Presentation App

Contact

Up for a random conversation! Feel free to contact me here.