Qingyu Qu
Research Assistant at NumFOCUS
Hangzhou City, Zhejiang, China
Summary
🤩
Rockstar🎓
Top SchoolQingyu Qu is a research-driven software engineer with six years of experience building high-performance numerical solvers and scientific computing tools, currently working on computational wave algorithms and RBF-based Hilbert transform methods at HKUST Guangzhou. He has a strong track record in the SciML ecosystem—contributing Runge–Kutta Nyström implementations and adaptive threading to the widely used OrdinaryDiffEq.jl—and has developed GPU-accelerated and collocation solvers for boundary value and optimal control problems. Qingyu blends academic rigor (Zhejiang University master’s, Shandong University bachelor’s) with hands-on open-source development, having implemented MIRK solvers and defect-control adaptivity during Google Summer of Code. Pragmatic and results-oriented, he prefers shipping code to talking about ideas and often learns by building concrete, performance-sensitive implementations that bridge research and production.
6 years of coding experience
2 years of employment as a software developer
Bachelor's degree Control Science and Engineering, Bachelor's degree Control Science and Engineering at Shandong UniversityMaster student Control Science and Engineering, Master student Control Science and Engineering at Zhejiang University
French, English, Spanish
Github Skills (9)
algorithm10
algorithms10
differential-equations10
adaption10
adaptation10
ordinary-differential-equations10
implement10
julia10
high-performance9
Programming languages (14)
Github contributions (5)
SciML/OrdinaryDiffEq.jl
Jan 2023 - Jan 2023
High performance ordinary differential equation (ODE) and differential-algebraic equation (DAE) solvers, including neural ordinary differential equations (neural ODEs) and scientific machine learning (SciML)
Role in this project:
Back-end Developer
Back-end DeveloperContributions:14 reviews, 26 commits, 23 PRs in 18 days
Contributions summary:Qingyu primarily contributed to the implementation of various Runge-Kutta Nyström (RKN) methods for solving ordinary differential equations, focusing on methods like DPRKN4, DPRKN5, DPRKN6FM, and ERKN7. Their work involved defining constant caches and implementing the `perform_step!` functions for these methods within the `ordinarydiffeq.jl` project. Additionally, the user added adaptive regression methods tests and fixed adaptive parameters within the codebase. The user also added more optional threading to the existing code.
adaptiveodesscientific-machine-learningdifferential-algebraicdifferential
ErikQQY/OrdinaryDiffEq.jl
Mar 2021 - Oct 2024
Contributions:36 pushes, 30 branches in 3 years 7 months
odesscientific-machine-learningdifferential-equation-solversdifferentialode
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.