I lead the data analysis and modeling of the first ever high resolution infrared spectra of two ultra-cool brown dwarfs, in collaboration with a team from Smith College that leads a joint project that focuses on atmospheric modeling. I use the pipeline I developed in graduate school for this task. I also apply new algorithms I developed to achieve sub-pixel frame alignment for the telescope data, as well as advanced statistical models to extract insights from the spectra.

Led the development of the CHARIS-pyKLIP Post-Processing Pipeline. The pipeline is a major component of pyKLIP, a state-of-the-art open-source python library for analyzing imaging data of exoplanets that produced some of the most influential discoveries in the exoplanet field.Led the development and integration of image registration, spectral-photometric calibration, distortion correction, and a novel Expectation Maximization Principle Component Analysis algorithm algorithm for the Coronagraphic High Angular Resolution Imaging Spectrograph (CHARIS) as part of the pipeline.Developed the algorithm to fit for accurate on-sky positions of blended sources, using Monte Carlo simulations. Measured dynamical masses for two brown dwarfs that are the most precise and accurate mass measurements ever at the time of publication for all imaged brown dwarfs.Modeled brown dwarf evolution and provided important new insights on the physics of planet formation and cooling.Led the analysis of the first ever simultaneous spectral-polarimetry imaging data of a proto-planetary disk in the HD 34700 system.

Worked on GALEX Deep Imaging Survey and zCOSMOS Spectroscopic Survey to detect galaxies using cross-correlation.

Worked with sparse signal processing and improved an algorithm that distinguishes different spectral energy distributions (SED) in multi-band images.

Measured the mass of the Coma Cluster using machine learning. Applied a deep learning model that was trained on N-body simulation data to estimate the mass of the Coma Cluster.