Miles Stoudenmire
Research Scientist at Flatiron Institute, Center for Computational Quantum Physics
New York, New York, United States
Summary
👤
Senior🎓
Top SchoolMiles Stoudenmire is a research scientist with 14 years of experience applying and developing tensor network algorithms across condensed matter physics, quantum chemistry, and machine learning. Based at the Flatiron Institute’s Center for Computational Quantum Physics, he co-leads development of the widely used ITensor C++ library, contributing core low-level improvements for efficient tensor operations and memory management. He proposed a novel framework that uses tensor networks to parameterize ML models with training-time scaling linear in dataset size and constant-time inference, offering new avenues to probe model interpretability and generalization. With a PhD in physics and a track record spanning academia and research institutes, he combines deep theoretical insight with production-grade software engineering on open-source scientific tools.
14 years of coding experience
4 years of employment as a software developer
BS, Math, Physics, BS, Math, Physics at Georgia Institute of TechnologyPhD, Physics, PhD, Physics at UC Santa Barbara
Github Skills (11)
neural-network10
tensor10
c-language10
tensorflow10
cprogramming-language10
template-metaprogramming10
linear-algebra10
performance-optimization9
numerical-analysis9
refactoring8
refactor8
Programming languages (8)
Github contributions (5)
ITensor/ITensor
Aug 2011 - Mar 2022
A C++ library for efficient tensor network calculations
Role in this project:
Back-end Developer
Back-end DeveloperContributions:30 releases, 10 reviews, 3667 commits in 10 years 9 months
Contributions summary:Miles contributed to the C++ library for efficient tensor network calculations by updating the contract code, implementing doTask(Contract) for a new Dense storage design, generalizing ncprod and performing various upgrades in the storage. Their work focused on core functionalities of the library, including low-level memory management and the efficient computation of tensor operations. The changes involve mixed data types, and are therefore part of the library's core functionality.
physicstensor-decompositionchemistrycppc-library
ITensor/ITensorWebsite
Feb 2022 - Mar 2025
New ITensor website (https://itensor.org)
Contributions:13 PRs, 452 pushes, 11 branches in 3 years 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.