Peter Sharpe
Senior Scientific Machine Learning Software Engineer - Physics at NVIDIA
Cambridge, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolPeter Sharpe is a Senior Scientific Machine Learning Software Engineer at NVIDIA who blends deep aerodynamics and PDE-driven physics expertise with production-grade ML engineering to model complex physical systems like aerodynamics, weather, and heat transfer. With a PhD from MIT AeroAstro and nine years of experience across academia, industry, and government-funded fellowships, he specializes in physics-aware ML architectures that enforce symmetries, conservation laws, and discretization/units invariance to improve generalization for real-world engineering problems. He is the creator and active maintainer of influential open-source projects such as AeroSandbox, NeuralFoil, and TorchMesh, translating research ideas into practical Python and GPU-accelerated tools used for aircraft design and simulation. His consulting work across aerospace and fusion companies demonstrates a knack for turning first-principles modeling into deployable optimization tools that span from low-level CFD/mesh work to system-level cost and mission analysis. Notably, he pushes ML for PDEs beyond black-box approaches by explicitly encoding physics and global information flow into model architectures, a perspective he argues is essential for industrial-scale SciML. Based in Cambridge, MA, he combines academic rigor with hands-on software delivery and a track record of open-source impact.
9 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy - PhD Computational Science and Engineering (Dept. of Aeronautics and Astronautics), Doctor of Philosophy - PhD Computational Science and Engineering (Dept. of Aeronautics and Astronautics) at Massachusetts Institute of TechnologyBachelor of Science (B.S.) Mechanical Engineering; Minors in Aerospace Engineering Robotics and Mechatronics, Bachelor of Science (B.S.) Mechanical Engineering; Minors in Aerospace Engineering Robotics and Mechatronics at Washington University in St. Louis
English, Spanish, Italian, python
Github Skills (8)
hardware-designs10
aircraft10
develop10
system-design10
python10
aerodynamics10
aerospace10
data-analysis9
Programming languages (12)
Github contributions (5)
peterdsharpe/AeroSandbox
May 2019 - Jan 2023
Aircraft design optimization made fast through computational graph transformations (e.g., automatic differentiation). Composable analysis tools for aerodynamics, propulsion, structures, trajectory design, and much more.
Role in this project:
Aerospace Engineer & Software Developer
Aerospace Engineer & Software DeveloperContributions:17 releases, 7 reviews, 2976 commits in 3 years 9 months
Contributions summary:Peter's commits focus on enhancing the aerodynamic capabilities of the "AeroSandbox" repository, an aircraft design optimization tool. Their work involved adding functionality for calculating the base pressure coefficient for fuselages, modeling transonic wave drag, and implementing methods for airfoil polar generation. They made several improvements, including the introduction of a new plotting function and improvements to existing tools, showcasing a focus on improving the accuracy and usability of the code.
pythonautomatic-differentiationsimulationaircraftmdao
peterdsharpe/better-biobot-boston
Jan 2022 - Feb 2023
Contributions:623 commits, 14 pushes in 1 year 1 month
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.