About

I am an NLP researcher working on a wide range of topics in the field with some overlap with vision, speech and HCI. I have previously founded an esports org During 2018-2022. Apart from that, I have worked with meme pages and moderated a large region specifc subreddit gaining understanding of what goes on behind the scenes in social platforms and what issues preside over the platforms

From 2022-2024, I used to participate in shared tasks / Kaggle competitions and have finished 1st in over Half of those i took part in using limited resources competing against those who used over 100x. From 2024, I have been applying and obtaining grants on my own and started working on a few ideas.

Educational Journey

Masters in Machine Learning

University of Maryland , College Park
2022 Aug - 2024 May

Graduate Certificate in Data Science

University of Maryland , College Park

2022 Aug - 2023 May

Bachelors in Computer Science & Engineering

RVR & JC College of Engineering

2018 Aug - 2022 May

TBD

TBD

2025 Aug - TBD

Experience



Research Lead

1-800-Shared-Tasks

Aug 2024 - Present


Research Project Lead

Cohere4AI

Jul 2024 - Aug 2024


Research Assistant

RVR & JC

Jan 2024 - May 2024

LLM Post-Training, AI generated text Detection, AI Bias and Fairness. More info: Current Works


Research Assistant

VVIT

Aug 2023 - Jan 2024


Research Assistant

RVR & JC

Aug 2022 - Jan 2023

Built compute-efficient models for extracting data from self-reported health texts from various social platforms for further analysis.


ML & IoT Intern

Bolt IoT

Feb 2021 - Apr 2021

Built Time-series models for a smart humidity, lighting, and temperature management based on current and estimated usages and periodic trends


Founder

Black Ops Esports

Nov 2018 - Apr 2022

Leading a team of 30 : LLMs for Low-Resource Languages, Datasets and Benchmarks, LLM/VLM Applications.

Built models for detection of watermarks and texts/ text portions from Cohere’s LLMs over 23 Languages including human-AI co-authored texts

Built image segmentation models for identifying optical disk and cup regions from retina scans for providing accurate diagnosis of Glaucoma

Built models for bias detection in Telugu newspapers and articles for the task of identifying reliability each of the news source

Built Image time-series models to predict and analyze algorithms in battle-royale games to plan better strategies.


NLP Reserach Lead

Traversaal.ai

Nov 2024 - Present

Interested in collaborating / have any question regarding one of the previous papers ? fill the form below