Open-source tools and research software for exoplanet science, mission simulation, and scientific computing.
Led development of a Python library for loading, analyzing, and visualizing results from different exoplanet yield calculators. Enables standardized comparisons between mission concepts and observation strategies.
Created a high-performance Python library for interpolating coronagraph point spread functions using Fourier interpolation techniques. Built with JAX for JIT compilation.
Core developer of EXOSIMS, an open-source Python framework for simulating exoplanet direct imaging missions. The software is used to evaluate and optimize future space telescope designs and observation strategies. My contributions include performance optimizations, new scheduling algorithms, exozodiacal light modeling, and integration with modern coronagraph models.
Created a JAX-accelerated Python package for simulating realistic coronagraphic observations of exoplanetary systems. Includes realistic astrophysical scenes from ExoVista, Poisson noise, detector effects, spatially dependent coronagraph PSFs, and post-processing algorithms for generating science-grade synthetic data.
Shared utility functions and constants for my HWO simulation stack. JAX-native physical constants, unit conversions, and flux-conserving image transforms used across yippy, orbix, coronagraphoto, coronalyze, jaxedith, and hwosim.
Contributing JAX acceleration and yippy integration to pyEDITH, a Python exposure time calculator for HWO. Enables fast, differentiable exposure time calculations for mission yield optimization.
Developed an extremely optimized solver for Kepler's equation using JAX. Features automatic differentiation, vectorization, and JIT compilation, making it ideal for large-scale orbit propagation and optimization problems in mission design.
Developing a JAX-accelerated post-processing pipeline for coronagraphic observations. Includes PCA-based PSF subtraction, SNR map generation, aperture photometry, and detection statistics for evaluating exoplanet recovery in simulated and real direct imaging data.
JAX-native astrophysical scene modeling for HWO direct imaging. Composable Equinox modules for stars, planets, disks, and zodiacal light, with loaders that bridge high-fidelity tools like ExoVista. The same Scene flows through both coronagraphoto (image simulation) and jaxEDITH (exposure-time calculations), so the astrophysics stays consistent across all downstream science products.
JAX-native exposure-time-calculator kernel for HWO direct imaging. Returns scalar count rates, exposure times, and SNR predictions from a coronagraph + scene + observation triple. Built as the fast inner loop for yield calculations, sensitivity studies, and retrievals where the ETC has to be JIT-compatible, vmap-able, and differentiable end-to-end.
Shared hardware objects like primary aperture, throughput-affecting elements, coronagraph backend, and detector as composable Equinox modules. Goal is to have a single source of truth for HWO hardware configuration: both coronagraphoto and jaxEDITH import the same OpticalPath and component types, so changes propagate automatically across the simulation suite.
Matplotlib style sheets and utilities for HWO figures. Has the cyberpunk dark theme I use for talks, a publication-ready light theme for papers, and the Barbie colors Natasha Latouf uses because why not.
Built a comprehensive simulation framework that models the collection of radial velocity precursor data, manages orbit fitting, and optimizes direct imaging observation schedules using a constraint programming approach. Combines Python for high-level logic with C extensions for computational bottlenecks. Relies of forked versions of RadVel and RVSearch for performance improvements.
Core developer for the Roman Space Telescope's Imaging Mission Database. Designed and implemented the database schema that loads and analyzes all known exoplanets systems from the NASA Exoplanet Archive. The database informs target lists, observation schedules, and mission constraints.
Core developer of the official exposure time calculator for NASA's Roman Space Telescope Coronagraph Instrument. Python tool helping astronomers plan observations and optimize science yield.