Gravitational and High Energy Astrophysics

Anuj Kankani

PhD Candidate in Physics · West Virginia University · Advisor: Sean T. McWilliams

Using analytical and numerical methods, I study the gravitational and electromagnetic radiation from black holes and the plasma around them.

Anuj Kankani at a blackboard covered in equations with a colleague, who is writing. Photo: Maria O’Leary / Institute for Advanced Study Anuj Kankani presenting at a conference, beside a slide titled Temperature Anisotropies Along the Jet Boundary of M87*. Photo: James Beattie Anuj Kankani at the Institute for Advanced Study, wearing a Prospects in Theoretical Physics conference badge. Anuj Kankani with a colleague beside the Einstein statue at the Institute for Advanced Study. Photo: Maria O’Leary / Institute for Advanced Study Anuj Kankani standing on a rock outcrop against a blue sky. Illustration: a figure seen from behind stands in the rain with arms outstretched along a LIGO beam tube.

About me

I'm a PhD candidate in physics at West Virginia University, working with Sean T. McWilliams on gravitational and high energy astrophysics.

My research uses numerical and analytical methods to study gravitational waves, black holes on hyperbolic orbits, kinetic plasma effects around black holes, and more. This involves using general relativistic particle-in-cell simulations (Entity), numerical relativity simulations (Einstein Toolkit), and analytical black hole perturbation theory!

I expect to complete my PhD in May 2027.

Position
PhD Candidate, PhysicsWest Virginia University
Advisor
Sean T. McWilliams
Focus
Waveform modeling · NR · GRPIC

Publications

Papers & preprints

2026submitted · PRL

The Direct Wave is Not a Meaningful Test of Horizon Properties

A. Kankani, S. T. McWilliams

Using numerical relativity strain data, we show that the direct wave's frequency does not track the remnant's horizon frequency or surface gravity. Using it to test Hawking's area law (as done for GW250114) may lead to false violations.

2026submitted · PRD

Direct Waves in Black-Hole Binary Mergers: Insights from the Backwards One Body Model

A. Kankani, S. T. McWilliams

We explain why the BOB model is so accurate near merger. Using the Pöschl–Teller potential, we show how BOB's amplitude profile can be recovered through the QNM poles. We use rational filters to show BOB also naturally captures the non-QNM "direct wave". We show that the direct wave is largely uncorrelated with the horizon, even at high spin.

2026submitted · JOSS

gwBOB: A Python Package for Analytical Merger–Ringdown Gravitational Waveforms

A. Kankani, A. Morales, S. Dasgupta, S. T. McWilliams

An open source and user friendly Python package implementing the Backwards One Body model, with utilities for generating various flavors of BOB and comparing them directly against numerical relativity.

2025submitted · PRD

BOB the (Waveform) Builder: Optimizing Analytical Black-Hole Binary Merger Waveforms

A. Kankani, S. T. McWilliams

A comprehensive analysis of the fully analytical BOB model of the dominant (2,2) mode for non-precessing systems. We show it reaches accuracy comparable to highly calibrated effective-one-body and NR surrogate models. Furthermore, we compare BOB to a sum of QNMs and show how it outperforms QNMs, on a per-free-parameter basis, near the peak.

2025Phys. Rev. D 112, 124051

Modeling Relative Peak Times of Gravitational Wave Harmonics

A. Kankani, S. T. McWilliams

We use two semi-analytical methods, one from BOB's frequency evolution and one from a timelike geodesic in the remnant spacetime, to predict when each harmonic mode peaks, up to l = 8. We show our methods can outperform state of the art models for the largest timing differences.

2024Phys. Rev. D 110, 064033

Testing the Boundary-to-Bound Correspondence With Numerical Relativity

A. Kankani, S. T. McWilliams

Using numerical relativity simulations of bound and unbound black-hole encounters, we test the boundary-to-bound correspondence beyond the perturbative regime. We find that the relationships break down in full GR.

2021Phys. Rev. E 104, 035302

Optimization and Benchmarking of the Thermal Cycling Algorithm

A. Barzegar, A. Kankani, S. Mandrà, H. G. Katzgraber

We tune and benchmark the thermal cycling optimization algorithm, showing it competes closely with parallel tempering using isoenergetic cluster moves while far outperforming simpler heuristics like simulated annealing.

Software

Open-source tools I build and contribute to

gwBOB Primary Developer

A user-friendly, open-source Python package implementing the Backwards One Body waveform model, with tools for analysis and comparison against numerical relativity.

github.com/AnujKankani/BackwardsOneBody ↗

Entity Contributor

Kokkos-based General Relativistic Particle-in-Cell code for plasma astrophysics.

github.com/entity-toolkit/entity ↗

Einstein Toolkit Contributor

A widely adopted codebase for numerical relativity.

einsteintoolkit.org ↗

NRPy Contributor

Python symbolic code-generation toolkit for relativistic astrophysics.

github.com/nrpy/nrpy ↗

LisaWave-JAX Co-Developer

JAX-based LISA data-analysis tools for massive black holes and galactic binaries. In development.

Awards & recent talks

2021 – 2024

Ruby Distinguished Doctoral Fellow

Awarded to outstanding doctoral students in STEM fields at West Virginia University.

2024 – 2025

NASA WV Space Grant Graduate Fellowship

For the proposal "Modeling Black Hole Mergers with a Dynamical Background Spacetime."

2025

Mohindar Seehra Research Award

Departmental award recognizing outstanding research efforts in physics.

2026 · SCEECS — Plasmas Around Black Holes & Neutron Stars

Driving Temperature Anisotropies in the Jets of Black Holes

2026 · 16th International LISA Symposium

BOB the (Waveform) Builder + Direct Waves

2026 · Center for Gravitational Waves & Cosmology Symposium

Modeling Extreme Gravity: From Relativistic Plasmas to Gravitational Waves

2026 · APS Global Physics Summit

BOB the (Waveform) Builder: Optimizing Analytical Merger Waveforms

2025 · APS Global Physics Summit

Merger-Ringdown Modeling of Gravitational Waves Beyond the Dominant Mode

BOB

The Backwards One Body model, a quick explainer

Most merger models are highly reliant on NR information. BOB is a physically motivated, minimally tuned, and highly accurate model for the merger-ringdown waveform.

How BOB gets its two equations — perturbed null geodesics at the light ring give the amplitude; the News gives the frequency open video ↗

pip install gwBOB docs ↗

Explainers

A few of my recent papers

More videos will be added soon!

BOB the (Waveform) Builder

A comprehensive analysis of the Backwards One Body model against NR, state of the art waveform models and a sum of QNMs.

arXiv:2510.25012 ↗

Direct Waves in BBH Mergers: Insights from BOB

In this paper we show how BOB is related to a sum of QNMs. We then apply rational filters to show BOB also models the non-QNM direct wave componoent of the radiation.

arXiv:2603.15474 ↗

The Direct Wave Is Not a Horizon Test

The prompt emission after merger does not track the remnant’s horizon frequency or surface gravity can report violations that never happened. Working with numerical-relativity strain data, we show the direct wave is not a reliable probe of the remnant horizon.

arXiv:2607.02380 ↗

Widgets

Interactive widgets

Interactive widgets to explore spin weighted (s=-2) spherical harmonics and geodesic orbits. More widgets will be added soon!

Spin-weighted spherical harmonics (s = −2) — the angular modes every waveform is built from open the spin-weighted spherical harmonic visualizer ↗
Geodesic orbits around a black hole — how matter and light move in curved spacetime open the geodesic orbit visualizer ↗