Stefan Radev
Assistant Professor at Rensselaer Polytechnic Institute
City of Troy, New York, United States
Summary
👤
Senior🎓
Top SchoolStefan Radev is an Assistant Professor at Rensselaer Polytechnic Institute and founder of BayesOps, where he develops neural-network-based tools that make advanced statistical inference accessible to scientists and engineers. With nine years of experience spanning doctoral research and postdoctoral roles at Heidelberg University, he specializes in simulation-based inference and is the core developer of the open-source BayesFlow framework. His background uniquely blends computer science and psychology, enabling him to translate probabilistic theory into user-friendly software and practical analysis pipelines across domains from EEG to motion capture. Stefan has repeatedly shipped research-grade tooling—GUI applications, MEAN-stack data collection apps, and C++/Python packages—demonstrating a rare combination of deep methodological insight and production engineering. He’s driven by computational philosophy and a passion for explainability, often collaborating with explainable ML groups to make complex models interpretable. Notably, his work emphasizes reproducible research practices that help non-specialists make better data-driven decisions.
9 years of coding experience
8 years of employment as a software developer
Bachelor of Science - B.Sc., Computer Science, Bachelor of Science - B.Sc., Computer Science at Heidelberg University
Github Skills (37)
bayesian-statistics10
python10
abc10
bayesian10
deep-learning10
generative-ai10
legacy-code10
pytorch8
data-science7
machine-learning7
inference7
psychology7
cuda7
transformer6
gpt6
Programming languages (7)
Github contributions (5)
stefanradev93/cINN
Jun 2019 - Dec 2020
Contains legacy code and model examples for the paper "BayesFlow: Learning complex stochastic models with invertible neural networks"
Contributions:102 commits, 45 pushes in 1 year 6 months
neural-network
stefanradev93/fast-dm-gui
Oct 2017 - Mar 2018
A python-based graphical user interface for Windows users of fast-dm - a diffusion model analysis tool.
Contributions:13 commits, 10 pushes, 1 branch in 4 months
graphical-user-interfacepythonwindowscognitive-modelsdiffusion-model
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.