Dipto Mukherjee

Logo

Computational Astrophysicist

View My GitHub Profile

About me

I am a PhD student in computational astrophysics at Carnegie Mellon University, having graduated summa cum laude in physics with minors in economics and computer science from Allegheny College in 2019.

My research focuses on developing numerical methods for gravitational dynamics, particularly in simulating massive black hole binary mergers and intermediate mass ratio inspirals in dark matter spikes. My collaborators and I created a collisional N-body code, Taichi, which utilizes an optimized fast multipole method (FMM) force solver. Additionally, I work on novel integration and timestep methods to preserve the Hamiltonian’s symplectic and time-symmetric properties. My diverse background includes solar system dynamics, stellar dynamics in young massive clusters, and combining hydrodynamics with gravitational dynamics for clusters.

I am proficient in Fortran (77, 90), C/C++, and Python, with experience in parallel computing using OpenMP, CUDA, and MPI. My interests in computer science extend to high-performance computing (HPC), GPU computing, and machine learning. I have applied Generative Adversarial Networks (GANs) to simulate galaxy images and used symbolic regression for new force softening techniques in cosmological simulations.

Beyond academics, I served as president of the Astronomy Club at Allegheny College and co-host Astronomy on Tap, Pittsburgh, bringing astronomy to the public. I enjoy reading, hiking, and traveling to new places.

Research | Projects | Talks and posters

Publication list | Orcid | Google Scholar