Steven Berguin
Principle Software Engineer at Archer
Atlanta, Georgia, United States
Summary
👤
Senior🎓
Top SchoolSteven Berguin is a Principal Software Engineer and operations research specialist with eight years of applied experience bridging optimization, simulation, and software development for defense and aerospace customers. Based in Atlanta, he has led mixed-integer and multidisciplinary optimization efforts—from coordinating airborne laser intercepts to aircraft maintenance scheduling—turning complex math into actionable decision-support tools using Python, automatic differentiation, and interactive Jupyter interfaces. His work repeatedly narrowed multi-year roadblocks into months-long successes and seeded sponsored transitions, including a $150K AFRL contract and production-ready capabilities. Steven combines deep aerospace research (PhD, Georgia Tech) with hands-on systems engineering at Raytheon and GTRI, and he’s practiced devops for legacy scientific code to achieve order-of-magnitude runtime improvements. Notably, he has a pattern of reducing problem dimensionality and building visual, AI-augmented tools that make physics-based tradeoffs accessible to non-experts.
8 years of coding experience
10 years of employment as a software developer
Bachelor's Degree Aerospace Engineering, Bachelor's Degree Aerospace Engineering at The University of Texas at AustinDoctor of Philosophy (Ph.D.) Aerospace Engineering, Doctor of Philosophy (Ph.D.) Aerospace Engineering at Georgia Institute of TechnologyGraduate Certificate Applied Trajectory Optimization, Graduate Certificate Applied Trajectory Optimization at Naval Postgraduate School
French, English
Github Skills (34)
conda10
sampling10
forge10
density-estimation10
derivative10
predictive-modeling10
surrogate10
toolbox9
mixture-of-experts9
modeling9
machine-learning8
gradient8
conda-forge8
neural-network5
tensorflow5
Programming languages (4)
Github contributions (5)
shb84/JENN
Apr 2018 - Jul 2020
Jacobian-Enhanced Neural Networks (JENN) are fully connected multi-layer perceptrons, whose training process was modified to account for gradient information. Specifically, the parameters are learned by minimizing the Least Squares Estimator (LSE), modified to minimize prediction error of both response values and partial derivatives.
Contributions:1 release, 19 commits, 30 PRs in 2 years 3 months
deep-learningneural-networkgradienttensorflowgenn
shb84/ipysensitivityprofiler
Jan 2024 - Jan 2025
A Library of Jupyter widgets to visualize models and assess predictive qualities
Contributions:1 PR, 51 pushes, 4 branches in 11 months
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.