Jianfei Chen
Machine Learning Engineer at Tsinghua University
Beijing, United States
Summary
🤩
Rockstar🎓
Top SchoolJianfei Chen is a Machine Learning Engineer with 14 years of experience specializing in scalable machine learning and Bayesian inference. Based in Beijing, he combines deep probabilistic modeling expertise with practical engineering skills to build and optimize MCMC and HMC-based samplers for real-world systems. His open-source work on the well-regarded zhusuan library includes adding NUTS utilities, Hamiltonian integrators, and adaptive stepsize/variance mechanisms, reflecting a focus on robust diagnostics and sampling performance. Jianfei is comfortable bridging research and production, turning advanced Bayesian methods into reliable tooling and diagnostics like effective sample size estimators. Colleagues rely on him for sharpening probabilistic pipelines and making sophisticated inference workflows operational at scale.
13 years of coding experience
Tsinghua University
Github Skills (13)
mcmc10
generative-model10
bayesian10
probabilistic-programming10
deep-learning10
tensorflow10
mc10
python10
bayesian-inference10
numpy9
graphical-models9
algorithms8
algorithm8
Programming languages (3)
Github contributions (5)
thu-ml/zhusuan
Aug 2016 - Nov 2017
A probabilistic programming library for Bayesian deep learning, generative models, based on Tensorflow
Role in this project:
ML Engineer
ML EngineerContributions:20 commits, 4 PRs, 3 pushes in 1 year 3 months
Contributions summary:Jianfei implemented and refined core components for Bayesian deep learning and probabilistic programming within the `zhusuan` library. Their contributions included the addition of utility functions, such as those for the NUTS algorithm, and the development of diagnostic tools for evaluating model performance, specifically the effective sample size. Furthermore, the user added integrators and a Hamiltonian class, which suggests they were involved in improving the MCMC sampling capabilities of the library. They also added and refined the implementation of adapters, providing features for stepsize and variance adaptation within the HMC framework.
information-theorydeep-learningbayesian-inferencegraphical-modelsmachine-learning
cjf00000/ScaCTM
Feb 2014 - Mar 2015
Contributions:34 commits, 7 pushes, 1 comment in 1 year
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.