Miles Cranmer
Assistant Professor at University of Cambridge
Cambridge, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolMiles Cranmer is an Assistant Professor at the University of Cambridge who develops machine learning methods tailored to the physical sciences, blending deep learning with symbolic and physics-informed approaches. With an 11-year research and engineering track record spanning Princeton, the Simons Foundation, and a DeepMind internship, he moves fluidly between production-grade code and novel scientific models. His open-source contributions include performance and parallelization work in core scientific projects like NumPy and high-performance symbolic regression tools (PySR, SymbolicRegression.jl), showing a rare mix of numerical library optimization and algorithmic innovation. He often focuses on pragmatic maintainability—updating demos, tests, and cross-backend support—so others can reproduce and extend his work. Based in Cambridge, UK, he combines astrophysics training (PhD) with hands-on software engineering to extract interpretable laws from complex learned models.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - Ph.D., Astrophysical Sciences, Doctor of Philosophy - Ph.D., Astrophysical Sciences at Princeton UniversityHigh School, International Baccalaureate, High School, International Baccalaureate at Cameron Heights Collegiate InstituteBachelor of Science - BS, Honours Physics, Bachelor of Science - BS, Honours Physics at McGill University
Stackoverflow
865reputation
37kreached
10answers
7questions
Github Skills (44)
algorithm10
unit-testing10
code-optimization10
algorithms10
pytorch10
pytest10
back-end-development10
python10
multithreading10
testing10
machine-learning10
data-structure10
refactor10
numpy10
distributed-computing10
Programming languages (26)
Github contributions (5)
MilesCranmer/PySR
Sep 2020 - Jan 2023
High-Performance Symbolic Regression in Python and Julia
Role in this project:
Back-end Developer
Back-end DeveloperContributions:87 releases, 229 reviews, 1633 commits in 2 years 4 months
Contributions summary:Miles primarily focused on adding parallelization features to the Eureqa.jl codebase, a high-performance symbolic regression library. Their contributions involved modifying existing functions to incorporate multi-threading and distributed computing approaches using technologies such as `Distributed` and `pmap`. They also addressed a bug regarding local variable redefinition and improved code robustness by implementing checks for potential domain errors, improving the efficiency and performance of the symbolic regression algorithms.
regressionpythonnumpysymbolic-regressionmachine-learning
MilesCranmer/SymbolicRegression.jl
Jan 2021 - Jan 2023
Distributed High-Performance Symbolic Regression in Julia
Role in this project:Back-end Developer
Back-end DeveloperContributions:29 releases, 122 reviews, 1454 commits in 2 years
Contributions summary:Miles's contributions focused on refactoring the Julia codebase, converting it to a proper Julia namespace and preparing it for optimization. The commits involved modifying core files, including hyperparameter, operator, and dataset definitions, to support the new design. Furthermore, the user added functionality for more efficient expression evaluation and optimization, along with additional tests for the code.
regressionsymbolic-computationsymbolic-regressionscimlmachine-learning
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.