Sudershan Sridhar
Research Assistant at Globus.org- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
-
Tamil Native or bilingual proficiency
-
Hindi Professional working proficiency
-
English Full professional proficiency
Topline Score
Bio
Credentials
-
Convolutional Neural Networks
CourseraMay, 2021- Oct, 2024 -
Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization
CourseraApr, 2021- Oct, 2024 -
Neural Networks and Deep Learning
CourseraApr, 2021- Oct, 2024 -
Structuring Machine Learning Projects
CourseraApr, 2021- Oct, 2024
Experience
-
Globus.org
-
United States
-
Software Development
-
1 - 100 Employee
-
Research Assistant
-
Jun 2023 - Present
-
-
-
University of Chicago
-
United States
-
Higher Education
-
700 & Above Employee
-
Teaching Assistant
-
Mar 2023 - Jun 2023
Teaching Assistant for Cloud Computing course Teaching Assistant for Cloud Computing course
-
-
-
Harris School of Public Policy at the University of Chicago
-
United States
-
Higher Education
-
100 - 200 Employee
-
Research Assistant
-
Sep 2022 - Jun 2023
1. Responsible for developing an energy e-commerce web application that monitors user interactions across different UIs administered through A/B testing. The captured mouse movements and navigation patterns are passed to a backend microservice to analyze behaviors. 2. Building a full stack application using ReactJs, NodeJs, and MongoDB for the website and Ruby on Rails for the admin panel. Set the vision to develop machine learning models to interpret the data. 1. Responsible for developing an energy e-commerce web application that monitors user interactions across different UIs administered through A/B testing. The captured mouse movements and navigation patterns are passed to a backend microservice to analyze behaviors. 2. Building a full stack application using ReactJs, NodeJs, and MongoDB for the website and Ruby on Rails for the admin panel. Set the vision to develop machine learning models to interpret the data.
-
-
-
GuardianLink
-
Internet Publishing
-
100 - 200 Employee
-
Back End Developer
-
Jun 2021 - Jun 2022
Responsible for building decentralized applications (dApp) for several use cases like the Non-Fungible Token (NFT) marketplace (beyondlife.club) and the world’s first metaverse cricket platform (jump.trade) Responsible for building decentralized applications (dApp) for several use cases like the Non-Fungible Token (NFT) marketplace (beyondlife.club) and the world’s first metaverse cricket platform (jump.trade)
-
-
-
District 120 Toastmasters (Tamilnadu, India)
-
India
-
Non-profit Organizations
-
1 - 100 Employee
-
Public Relations Manager
-
Jun 2020 - May 2021
Served as the PR Manager for 30 clubs (Division A of District 120) across the state of Tamil Nadu, India. Served as the PR Manager for 30 clubs (Division A of District 120) across the state of Tamil Nadu, India.
-
-
-
Vellore Institute of Technology
-
India
-
Education Administration Programs
-
700 & Above Employee
-
Student Researcher
-
Feb 2021 - Apr 2021
Developed a voice recognition model utilizing computer vision that comprehends sentences using visual cues. Researched meticulously and delivered enhancement to the preprocessing pipeline using BlazeFace and Silence detection model for better face detection and video segmentation. Accuracy bumped up by 20% more than similar models. Work pending submission to CRC Press. Developed a voice recognition model utilizing computer vision that comprehends sentences using visual cues. Researched meticulously and delivered enhancement to the preprocessing pipeline using BlazeFace and Silence detection model for better face detection and video segmentation. Accuracy bumped up by 20% more than similar models. Work pending submission to CRC Press.
-
-
-
-
Student Researcher
-
Aug 2020 - Sep 2020
Proposed and implemented a new encoding system for multilingual databases/corpora that reduced the corpus storage size by 14% and managed to decode with 95% accuracy. The space required for corpus storage is reduced by training the algorithm to encode batches of alphabets from multiple languages with the same character code. Consecutively, the algorithm learns to decode the characters with the help of language metadata. Proposed and implemented a new encoding system for multilingual databases/corpora that reduced the corpus storage size by 14% and managed to decode with 95% accuracy. The space required for corpus storage is reduced by training the algorithm to encode batches of alphabets from multiple languages with the same character code. Consecutively, the algorithm learns to decode the characters with the help of language metadata.
-
-
Education
-
University of Chicago
Masters, Computer Science -
Vellore Institute of Technology
Bachelor's degree, Computer Science