Luke Meyers
Machine Learning Engineer at SyntheticGestalt
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolLuke Meyers is a Machine Learning Engineer with 11 years of cross-disciplinary experience building research-driven systems that encode rich mathematical structure to tackle ill-defined and high-impact problems. He has shipped production ML pipelines and prototypes across startups and open-source projects, from supply-chain and space-weather probabilistic models to text-to-CAD and distributed continual pretraining of LLMs. Luke combines deep numerical instincts—evident in contributions to the Rust nalgebra library (QZ decomposition, generalized eigenvalues, property testing)—with practical engineering skills in Clojure, Go, Python and scalable NPU-aware training. He’s comfortable navigating vague problem statements, designing ablation studies and validating novel model-merging and permutation-based methods to reduce inference variance. Based in Old Toronto, he pairs a mechanical engineering foundation with a creative, mathematically curious approach that treats model design as both an engineering and an expressive endeavor.
11 years of coding experience
10 years of employment as a software developer
Bachelor of Applied Science - BASc , Mechanical Engineering, Bachelor of Applied Science - BASc , Mechanical Engineering at University of Windsor
Stackoverflow
541reputation
14kreached
1answer
27questions
Github Skills (20)
algorithm10
algorithms10
matrix10
algebra10
data-structure10
data-structures10
rust10
linear-algebra10
testing9
serialization7
data-serialization7
lisp6
computational-geometry6
casting6
common-lisp6
Programming languages (11)
Github contributions (5)
dimforge/nalgebra
Jan 2022 - Apr 2022
Linear algebra library for Rust.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:10 reviews, 62 commits, 2 PRs in 2 months
Contributions summary:Luke primarily worked on the `nalgebra` library, a linear algebra library for Rust. Their contributions focused on implementing and testing the QZ decomposition algorithm, a core linear algebra function. They addressed serialization issues and made comments to improve the function of the decomposition. Further contributions included developing and refactoring tests related to generalized eigenvalues and matrices.
linear-algebrarustmatrixvectoralgebra
metric-space/trl
Apr 2023 - Mar 2025
Train transformer language models with reinforcement learning.
Contributions:131 pushes, 5 branches in 1 year 11 months
language-modelreinforcement-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.