Identified for the first time the applicability of positive and unlabelled machine learning techniques to study repeating fast radio bursts; flagged 66 (18 new) repeater candidates in the CHIME/FRB catalogPresented findings at the 2023 IEEE Massachusetts Institute of Technology URT Conference expanded paper published in the Monthly Notices of the Royal Astronomical SocietyOne of 20 projects to win the grand prize at the Indian IRIS National Science Fair in 2024 and be selected as part of Team India for the 2024 International Science & Engineering Fair in Los Angeles.Trained, optimized five PU classifiers; improved recall by 4.62% and outperformed past work on key PU metrics. Supervised by Dr. Vinesh Maguire Rajpaul from the University of Cambridge.Conference proceedings: ieeexplore.ieee.org/abstract/document/10535037)Full-length MNRAS paper: doi.org/10.1093/mnras/stae1972arXiv:2408.11436

Selected as the Head of Megabyte, TSRS Moulsari's computer society.Co-organized the school's first AI Day in 2022 and AI Week in 2023:- Organised 10 installations and activities to help 400+ students interactively engage with emerging AI technologies such as language models (2022) and voice cloning (2023).- Addressed the 1000+ student body on the responsible use and identification of AI-generated contentDesigned, co-developed, and launched club website (www.tsrsmegabyte.com) with Astro and TailwindCSS.Selected and mentored teams from grades 6-12 for 15+ inter-school competitions and conducted weekly sessions and workshops for 30+ society membersCaptained teams and won at four inter-school hackathons / competitive programming events

Organized inter-house puzzle-solving event (inspired by the MIT Mystery Hunt) in 2021, 2022, and 2023.Led organizing team in 2023 and increased participation to 350+ students.Developed online platform for the event with Django and TailwindCSS. Continuously served hundreds of simultaneous users a crash-free experience for 72+ hours every year while logging 40k+ cumulative answer submissions.

Selected to be a member of the Student Council as Secretary of Information and Communication Technology.Part of the 3-member team that led the organization of ShriTeq 2022:- Brought back the inter-school tech symposium after 4 years- Increased number of participants by 3x to 1000+, number of participating schools by 2x to 51 in the first international edition of the event- Co-organized 13 technology-related competitions for students across grades 6-12

Ideated and co-organized Bot Xchange, a first-of-its-kind algorithmic trading competition for high schoolers at ShriTeq 2022.- Co-developed market simulator code, time series machine learning models for stock price prediction, and tutorial notebooks (github.com/tsrsmict/bot-xchange-22)- Wrote webscraper to extract historic price data for 200+ stocks- Authored extensive documentation for the event (bit.ly/bot-xchange-starter-pack)

Developed four new features for a Python natural language processing webscraper that extracts market insights from 200k+ news articles/day; deployed to containerized Linux systems using Docker and AWS ECR; implemented interfaces for features with NodeJS, React, and AWS EC2- Implemented rapid offline extractive text summarization using TF-IDF for all news articles- Refactored algorithm to filter articles and identify buzzwords and stocks using natural language processing techniques and libraries such as spaCy, nltk- Enhanced weighting of articles and authors with machine learning regression algorithm to penalize biased language- Developed object-oriented program to asynchronously identify terms driving daily market sentiment for F500 stocks