Wengong Jin
Assistant Professor at Northeastern University
Cambridge, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolWengong Jin is an assistant professor and computational scientist with 11 years of experience at the intersection of machine learning and molecular design, now based in Cambridge, MA. After a postdoc at the Broad Institute and internships at Google Brain and Microsoft Research Asia, he completed PhD work in computer science at MIT and brings a strong research-to-code mindset. His contributions to the ICML 2018 Junction Tree VAE for molecular graph generation show hands-on development of core model architectures and training pipelines for generative chemistry. He blends academic rigor with practical engineering—frequently modifying model and data-processing codebases—to push property prediction and molecule generation forward. Colleagues value him for translating cutting-edge research into reproducible, usable implementations that accelerate drug discovery and computational chemistry.
11 years of coding experience
3 years of employment as a software developer
PhD Candidate Computer Science, PhD Candidate Computer Science at Massachusetts Institute of TechnologyComputer Science, Computer Science at Shanghai Jiao Tong University
Chinese, English
Github Skills (8)
variational-autoencoder10
pytorch10
machine-learning10
deep-learning10
python10
modeling10
data-science9
bioinformatics8
Programming languages (2)
Github contributions (5)
wengong-jin/icml18-jtnn
Feb 2018 - Apr 2021
Junction Tree Variational Autoencoder for Molecular Graph Generation (ICML 2018)
Role in this project:
ML Engineer
ML EngineerContributions:44 commits, 60 pushes, 2 branches in 3 years 2 months
Contributions summary:Wengong made several updates to the code base, mainly focused on the core machine learning aspects of the project. These changes involved modifying key files related to the model architecture and training processes, including `jtnn_vae.py`, `jtprop_vae.py`, `vae_train.py` and `mol_tree.py`. The commits suggest ongoing development and refinement of the variational autoencoder model for molecular graph generation and property prediction. These changes also suggest work on both model and data processing steps.
autoencoderjunctionmoleculargraph-neural-networksvariational-autoencoder
wengong-jin/multiobj-rationale
Jul 2020 - Sep 2022
Multi-Objective Molecule Generation using Interpretable Substructures (ICML 2020)
Contributions:12 commits, 1 PR, 16 pushes in 2 years 3 months
molecule-generationicml-2020moleculeicmlinterpretable
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.