Dongkwan Kim
Postdoctoral Researcher at Texas A&M University
College Station, Texas, United States
Summary
👤
Senior🎓
Top SchoolDongkwan Kim is a postdoctoral researcher at Texas A&M University with a decade of experience at the intersection of machine learning and computational chemistry/biology. He develops foundational multimodal models that integrate graphs and text to enable reasoning over complex relational and hierarchical structures in biological and chemical domains. During his Ph.D. at KAIST he advanced graph representation learning by modeling pairwise and higher-order interactions, including subgraph- and k-hop-based methods. He has hands-on experience contributing to PyTorch Geometric—improving pooling, negative sampling, tests, and docs—bridging research ideas with robust open-source implementations. With a background combining computer science and a chemistry minor, he brings domain intuition to build models that accelerate scientific discovery.
10 years of coding experience
Master of Science - MS, Master of Science - MS at 한국과학기술원(KAIST)Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Korea Advanced Institute of Science and Technology
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (8)
pytorch10
machine-learning10
deep-learning10
pytorch-geometric10
graph-neural-network10
graph-convolutional-networks10
python10
test-automation9
Programming languages (7)
Github contributions (5)
pyg-team/pytorch_geometric
Jan 2019 - May 2022
Graph Neural Network Library for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:13 reviews, 51 commits, 19 PRs in 3 years 4 months
Contributions summary:Dongkwan primarily contributed to the development and improvement of the PyTorch Geometric library, focusing on graph neural network (GNN) related functionalities. Their work included implementing and testing new features for existing modules such as `global_sort_pool` and negative sampling methods. They also fixed typos, added documentation, and contributed to the examples and tests of different GNN models.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
todoaskit/dike
Nov 2017 - Apr 2018
Contributions:29 PRs, 80 pushes, 24 branches in 5 months
kaistsocial-computingcomputingjusticecrowdsourcing
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.