Rastko Ciric
Scientist at Allen Institute
Mountain View, California, United States
Summary
👤
Senior🎓
Top SchoolRastko Ciric is a Stanford-trained scientist and software engineer with a decade of experience applying machine learning and data engineering to medical imaging and reproducible neuroscience. He has authored over 30 peer-reviewed papers with 8,000+ citations and has driven practical open-source contributions to flagship neuroimaging projects like fMRIPrep and Nipype, improving confound estimation and local-statistics interfaces used broadly by the community. His work spans research and production: from JAX-based differentiable imaging pipelines and graph U-Net architectures to engineering optimizations that cut I/O latency 10× for gigascale datasets. More recently he fine-tuned open-weight LLMs and vision-language models for web navigation agents, demonstrating a knack for translating deep research into production ML. Based in Mountain View, he blends rigorous scientific validation with pragmatic software development to build reproducible, scalable data-intensive systems.
11 years of coding experience
1 year of employment as a software developer
Bachelor of Arts - BA, Neuroscience, Bachelor of Arts - BA, Neuroscience at Pomona CollegeDoctor of Philosophy - PhD, Bioengineering, Doctor of Philosophy - PhD, Bioengineering at Stanford University
Github Skills (9)
nipype10
workflow-engine10
python10
neuroimaging10
brain-imaging10
data-analysis10
image-processing9
dataflow9
dataflow-programming9
Programming languages (5)
Github contributions (5)
nipreps/fmriprep
Jan 2019 - Apr 2020
fMRIPrep is a robust and easy-to-use pipeline for preprocessing of diverse fMRI data. The transparent workflow dispenses of manual intervention, thereby ensuring the reproducibility of the results.
Role in this project:
Backend Developer & Data Scientist
Backend Developer & Data ScientistContributions:52 commits, 6 PRs, 7 comments in 1 year 3 months
Contributions summary:Rastko primarily focused on enhancing the fmriprep pipeline, adding functionality for calculating and reporting confound regressors. Their contributions included implementing new CompCor decompositions, integrating RMS framewise displacement, and generating metadata for confound time series. They also refined the project's reporting features.
preprocessingfmrifmri-preprocessingbrain-imagingneuroimaging
nipy/nipype
Oct 2018 - Aug 2019
Workflows and interfaces for neuroimaging packages
Role in this project:Back-end Developer
Back-end DeveloperContributions:53 commits, 5 PRs, 12 comments in 9 months
Contributions summary:Rastko's contributions focused on enhancing the `nipype` library, which is used for neuroimaging workflows. Specifically, the user implemented new functionalities related to the `3dLocalstat` interface for computing local statistics and added a new `ReHo` (regional homogeneity) interface. Additionally, the user modified the `ROIStats` interface by expanding its functionalities and made improvements to the `Localstat` interface. The user's changes indicate a focus on enhancing the capabilities of the neuroimaging package and expanding its functionality.
neuroimagingpythonworkflow-enginebrain-imagingbig-data
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.