Jonas Kulhanek
Research Scientist at Apple
Prague, Prague, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolJonas Kulhanek is a research scientist and PhD-trained computer scientist specializing in implicit 3D scene representations and neural rendering, with research supervised by Torsten Sattler and Marc Pollefeys as part of the ELLIS program. With 11 years of industry and research experience spanning deep reinforcement learning, neural scene representation, and large-scale GPU training, he has worked at institutions including CTU, ETH Zürich, Meta Reality Labs, Google, and now Apple. He contributes to prominent open-source projects such as nerfstudio—improving build processes, data pipelines, and Torch 2.0 support—bridging research prototypes and production-ready tooling. His background combines practical software engineering (full-stack .NET/NodeJS work and DevOps) with cutting-edge ML research, enabling reproducible, deployable systems for 3D perception. Notably, he has built open 3D simulators for DRL training and collaborated with leaders like Tomáš Mikolov on cross-disciplinary projects, reflecting a knack for applying theoretical ideas to working systems. Based in Prague, he brings a blend of academic rigor and hands-on engineering to problems at the intersection of computer vision, robotics, and ML infrastructure.
11 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Czech Technical University in PragueBachelor’s Degree, Information Technology, Bachelor’s Degree, Information Technology at Faculty of Electrical Engineering, Czech Technical University in PragueComputer Science, Computer Science at Sungkyunkwan UniversityMaster's degree, Artificial Intelligence, Master's degree, Artificial Intelligence at Charles University
English, Czech, French
Stackoverflow
197reputation
13kreached
15answers
1question
Github Skills (16)
pytorch10
machine-learning10
python10
3d-reconstruction10
data-processing9
cli9
command-line9
command-line-interface9
devops8
git7
statistics6
overleaf6
sharelatex6
openmp6
latex6
Programming languages (10)
Github contributions (5)
nerfstudio-project/nerfstudio
Feb 2023 - Jan 2025
A collaboration friendly studio for NeRFs
Role in this project:
Back-end Developer & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:128 reviews, 97 PRs, 188 pushes in 1 year 11 months
Contributions summary:Jonas made several contributions focused on improving the build process, method registration, and overall functionality of the NeRFstudio project. They implemented a system to allow method configurations to be discovered from external packages. Additionally, they addressed issues with the `ns-process-data` script, improving image processing and COLMAP integration. Furthermore, they fixed a bug in the viewer, improved the command line interface, and added support for Torch 2.0.
pytorchphotogrammetrydeep-learningcomputer-vision3d-reconstruction
jkulhanek/viewformer
Mar 2022 - Dec 2022
ViewFormer: NeRF-free Neural Rendering from Few Images Using Transformers
Contributions:16 commits, 46 pushes, 15 comments in 8 months
pytorchtransformersdeep-learningneural-renderingrendering
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.