Graduate research assistant at Louisiana State University​, Baton Rouge, USA

2018 - present

 

As a research assistant, my duties involve working on research projects with fellow graduate students, post-docs, and other scientists in the LSC collaboration along with mentoring younger graduate students. I also presented my work in the meetings and telecons to the wider working groups.

 

Major Projects

1) Reducing Slow Scattering noise in LIGO detectors

• Noise in the detector due to scattered light diminishes its sensitivity for the detection of gravitational waves. After finding the exact mechanism through which this noise couples to the detector, we implemented a technique that reduced the rate of noise by a huge factor.

• Identified better witness channels for scattering noise that helped catch more noise in the gravitational wave channel. The Exploratory Data Analysis also led to identifying two different populations of scattering noise in the data.

• Research paper accepted in CQG: Reducing Scattered Light in LIGO's Third Observing Run. S Soni et al arXiv link

​

2) Noise Classification with GravitySpy

• GravitySpy is a Deep Learning algorithm that uses a convolutional neural network (CNN) to classify noise based on its appearance in the time-frequency plane.

• I trained a new model to recognize two new classes of noise, namely "Fast Scattering" and "Low-frequency Blips." The identification of Fast Scattering led to an improved understanding of its origin in the gravitational wave detector.

• Discovering features in gravitational-wave data through detector characterization, citizen science, and machine learning. S Soni et al DCC link

​

3) Assistance to Detector Characterization Team at KAGRA, Japan

• Installed Conda environments in the computing clusters and trained the Detchar staff in their maintenance.

• Created a list of standard channels to be analyzed by data analysis tools.

• Assisted another team member in setting up a channel monitoring system.

• DCC link

​

4) Fast Scattering data analysis.

• Statistical data analysis of Fast Scattering noise led to the identification of 2 Hz and 4 Hz Fast scattering noise transients in the data.

• The 2 Hz and 4 Hz couple to different ground motion bands and possibly have different sources.

• LIGO-Virgo-KAGRA talk public link

​

5) Loud noise transients in LIGO Data

• Loud triggers are short-duration noisy events that result in a massive drop in the sensitivity of the LIGO detector.

• In this project, I answer the question "Is the absolute rate of loud triggers higher during Observing run 3 (O3), or is the higher rate a side-effect of improved sensitivity in O3?" The tougher question to answer "Where do these loud triggers originate?" is an ongoing investigation.

• DCC link

​

Graduate teaching assistant at Louisiana State University​, Baton Rouge, USA

​2016 - 2018

 

• During the first 2 years as a teaching assistant, I worked towards finishing my course work and passing the qualifying exam. My work also included teaching undergraduate Physics lab, grading lab reports, home works, and assisting undergraduate students in the tutoring center. During this time, I also worked with the Quantum Gravity research group at LSU and this helped me gain an understanding of cosmology, topology, and general relativity.

 

Mentor to Summer/Undergraduate students

​

• Mentored a summer research student in 2018 and an LSU physics undergraduate student in Summer 2020 on projects with aLIGO.

​

High School Maths/Physics Teacher at Jain Heritage School, Hyderabad, India

​2013 - 2016

 

• I taught Cambridge AS and A level Mathematics and Physics to high school students at a Cambridge affiliated school in Hyderabad, India. My work also included grading the home works and examinations.

​

Management Trainee at Vizag Steel Plant, India

2011 - 2012

• During my brief stint with the Vizag steel plant in Visakhapatnam, India, I worked as a trainee in the machine maintenance department of Steel Melt Shop.

​