Research

Pulsating stars, star clusters and machine learning

I apply computational and machine-learning methods to stellar photometric and spectroscopic data, using model-based interpolation and full-spectrum analysis to study RR Lyrae variables and characterise stars in globular clusters.

Helix Nebula · NASA, ESA, C.R. O'Dell (Vanderbilt University), M. Meixner, P. McCullough & G. Bacon (STScI)

Themes

Four connected questions

Each theme links to the papers behind it.

RR Lyrae variability

RR Lyrae stars are old, pulsating stars whose light curves encode their physical properties. I work on light-curve modelling, photometric metallicities and stellar-parameter estimation for RR Lyrae variables, in the field and in globular clusters.

Machine learning & model interpolation

I develop artificial-neural-network interpolators for grids of stellar pulsation models and spectra, so that observed light curves and spectra can be compared with model predictions across the grid and physical parameters recovered directly.

Globular clusters & stellar spectroscopy

I combine multiband photometry with full-spectrum fitting to characterise stars in globular clusters, including the RR Lyrae populations of M3 and M53 and the stars of NGC 6397.

Background

Thesis & observing

Ph.D. 2025
Automated Analysis of Stellar Photometric and Spectroscopic Astronomical Data, Astronomy and Astrophysics, University of DelhiSupervisors: Prof. H. P. Singh (University of Delhi) and Prof. Philippe Prugniel (University of Lyon). Presented in the K. D. Abhyankar Best Thesis Presentation category at ASI 2026.
Observing
Co-PI, 3.6 m Devasthal Optical Telescope (ARIES), two successive cyclesTime-series photometry and spectroscopy of RR Lyrae stars in the globular cluster M3 (NGC 5272).
Computing
Python, C, C++, Fortran, IDL, SQL/ADQL; TensorFlow and scikit-learnLarge-scale computation on the IUCAA Pegasus cluster.
Instruments
Celestron 11″ and 8″ telescopesInstallation, operation and astrophotography for teaching and outreach at UPES.
Software

Tools

Web app · 2023

ANN Interpolator

Predicts RRab light curves by neural-network interpolation of a grid of pulsation models.

Teaching · interactive

Simulation hub

Browser-based labs on orbits, blackbody radiation, the H–R diagram, optical depth and cosmology.

Teaching · quiz

Stellar Quest

A stellar structure and evolution challenge with a class leaderboard.

Presentations

Talks & posters

Invited talks and outreach lectures are listed on the Outreach page.

  • 2026Ph.D. thesis, “Automated Analysis of Stellar Photometric and Spectroscopic Astronomical Data” — oral, K. D. Abhyankar Best Thesis Presentation category, 44th ASI Annual Meeting, IIT Guwahati (15–19 May)
  • 2024Research presentation as resource person, “Probing Stars and Galaxies Using Innovative Data Science Tools”, Gauhati University with University of Delhi, IUCAA and NIF (4–6 Sep)
  • 2022“Application of Artificial Neural Networks in Generating RR Lyrae Light Curves” — Young Astronomers’ Meet (YAM 2022), ARIES, Nainital (9–13 Nov)
  • 2021Poster, “Spectral Interpolation Using Artificial Neural Networks (ANN)” — 39th ASI Annual Meeting (virtual, 18–23 Feb) · Abstract