Ari Silburt is a Staff Machine Learning Engineer in New York with 11 years of experience blending ML production engineering and astrophysics research. He has built large-scale ML systems at Google for real-time bidding and ad ROI optimization, earlier applied NLP and full-stack ML at Bloomberg, and founded the HERMES hybrid N-body integrator used in planetary dynamics research. Ari’s background as a PhD astrophysicist and open-source contributor gives him deep expertise in numerical simulation, probabilistic modeling, and scientific ML—skills he’s translated into high-throughput, low-latency production services. He’s equally comfortable training convolutional and gradient-boosted models as he is contributing low-level integrators in C, a mix that surfaces in both research-grade simulations and production ML pipelines. Notably, his work on HERMES and REBOUND demonstrates an ability to solve edge-case physics problems (close encounters, collisions) that most ML engineers never encounter.
11 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy (PhD) Astronomy and Astrophysics, Doctor of Philosophy (PhD) Astronomy and Astrophysics at University of Toronto
Bachelor's degree Physics Minors in Mathematics and Astronomy, Bachelor's degree Physics Minors in Mathematics and Astronomy at Mount Allison University
Contributions:34 commits, 39 PRs, 27 comments in 9 months
Contributions summary:Ari implemented the Hermes integrator, a hybrid integration scheme designed to handle close encounters and other celestial mechanics simulations, as indicated by the code changes in `src/integrator_hermes.c`. Their work involved adding the Hermes integrator, including the necessary code, data structures, and helper functions for its correct implementation, as well as modifying other associated files to accomodate these new features. The user contributed by making critical changes related to solar switch factor, and the correct scaling of energy from collisions.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.