Erik Kastman
Staff Machine Learning Engineer at LinkSquares
Viroqua, Wisconsin, United States
Summary
🤩
Rockstar🎓
Top SchoolErik Kastman is a Staff Machine Learning Engineer with 14 years of experience building predictive models, ML platforms, and production software across cognitive neuroscience, microbial ecology, ag-tech and legal tech. He blends data science and full‑stack engineering, having led lab informatics at Harvard and served as CTO at an ag-tech startup before moving into platform and staff ML roles at LinkSquares. A pragmatic contributor to prominent neuroimaging open‑source projects (nipype, nibabel, psychopy), he has improved distributed workflow execution, DICOM handling, and testing infrastructure—work that quietly underpins many research pipelines. Based in rural Wisconsin, he pairs deep scientific domain knowledge with production-focused DevOps and QA instincts, making him adept at translating complex research workflows into reliable, scalable systems.
14 years of coding experience
14 years of employment as a software developer
BS, Biology, Philosophy, BS, Biology, Philosophy at University of Wisconsin-Madison
Stackoverflow
528reputation
23kreached
6answers
7questions
Github Skills (24)
data-formats10
python10
data-science10
testing10
pydicom10
workflow-engine10
data-format10
controls10
neuroimaging10
lsf10
emulation10
psychopy10
test-automation10
bash9
brain-imaging8
Programming languages (13)
Github contributions (5)
nipy/nibabel
Feb 2015 - Mar 2016
Python package to access a cacophony of neuro-imaging file formats
Role in this project:
Back-end Developer / Data Scientist
Back-end Developer / Data ScientistContributions:31 commits, 1 PR, 35 comments in 1 year 1 month
Contributions summary:Erik contributed to the `nibabel` repository by implementing and testing a NIfTI1 DICOM extension. Their work involved integrating the pydicom library to handle DICOM data within the NIfTI format. The user also addressed compatibility issues with Python 3, enhanced testing procedures, and refactored code for better performance. Furthermore, the user added documentation to the extension and tested a series of scenarios.
pythontckimagingtrkstreamlines
psychopy/psychopy
Feb 2012 - Jun 2017
For running psychology and neuroscience experiments
Role in this project:QA Engineer / Test Automation Engineer
QA Engineer / Test Automation EngineerContributions:34 commits, 4 PRs, 25 comments in 5 years 4 months
Contributions summary:Erik's contributions focused on enhancing the testing capabilities of the psychopy library. They added new tests for the emulator component, ensuring its correct functionality. They also implemented checks for data file handling, including collision detection and renaming strategies during saving operations. Furthermore, they introduced a wait timeout option to the emulator, improving test control and reliability.
neurosciencepythonexperiment-controlcognitive-scienceopenscience
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.