Projects

Things I've built and shipped.

A selection of technical projects spanning educational technology, scientific computing, and open-source infrastructure. For a full publication list, see my Google Scholar profile.

ROAR: Rapid Online Assessment of Reading

The Problem

Schools and researchers need reliable, scalable reading assessments that work across diverse devices, networks, and student populations, with strict privacy requirements and measurement validity constraints.

What I Own

Team of six engineers (and growing), technical roadmap, full-stack architecture, data pipeline design, production reliability, and cross-functional coordination with researchers, educators, and school districts.

Technical Choices

  • ▸ Web-based assessment delivery (browser-native, no installs)
  • ▸ Real-time data pipelines for research analytics
  • ▸ FERPA-compliant student data handling

Scale

  • ▸ 309 districts, 2,708 schools rostered
  • ▸ 160,000+ users who have completed assessments
  • ▸ 873,000+ assessment runs
EdTech Web Platform Data Pipelines Leadership FERPA

pyAFQ: Automated Fiber Quantification

Open-source Python pipeline for automated tractography and white-matter analysis of diffusion MRI data.

Role: Core developer. Designed the analysis pipeline, implemented statistical methods, and built quality-controlled data resources.

Impact: Published in Nature Methods and Scientific Data. Used by neuroimaging labs worldwide.

Python Neuroimaging Open Source Data Science

Cloudknot

A Python library that lets researchers run existing code on AWS Batch with minimal configuration. Eliminates the boilerplate of containerization, IAM setup, and job queue management.

Role: Creator and maintainer. Designed the API, implemented AWS integrations, and published at SciPy 2018.

Impact: Enabled researchers without cloud expertise to scale computations to hundreds of parallel jobs.

Python AWS Cloud Computing Open Source

Groupyr: Sparse Group Lasso

A scikit-learn-compatible Python library implementing the sparse group lasso for high-dimensional supervised learning with grouped features.

Role: Lead developer. Designed the API, implemented optimization algorithms, and published in JOSS.

Impact: Published in Journal of Open Source Software. Used for feature selection in neuroimaging studies (PLOS Computational Biology).

Python Machine Learning Statistics Open Source

AFQ-Browser

A browser-based tool for interactive visualization and analysis of diffusion MRI tractometry data.

Role: Core developer. Built interactive web visualizations for white-matter tract profiles.

Impact: Published in Nature Communications. Enabled collaborative, reproducible analysis of neuroimaging data.

JavaScript Visualization Neuroimaging Open Source

Selected Publications

Full list on Google Scholar or in my CV.

  • 2022 · Richie-Halford A, Cieslak M, Ai L, et al. An analysis-ready and quality controlled resource for pediatric brain white-matter research. Scientific Data. doi
  • 2021 · Kruper J, Yeatman JD, Richie-Halford A, et al. Evaluating the reliability of human brain white matter tractometry. Aperture Neuro. doi
  • 2021 · Richie-Halford A, Yeatman J, Simon N, Rokem A. Multidimensional analysis and detection of informative features in human brain white matter. PLOS Computational Biology. doi
  • 2021 · Cieslak M, et al. (incl. Richie-Halford A). QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data. Nature Methods. doi
  • 2020 · Richie-Halford A, Drut J, Bulgac A. Emergence of a pseudogap in the BCS-BEC crossover. Physical Review Letters. doi
  • 2018 · Richie-Halford A, Rokem A. Cloudknot: A Python Library to Run your Existing Code on AWS Batch. SciPy 2018. paper
  • 2018 · Yeatman JD, Richie-Halford A, Smith JK, Keshavan A, Rokem A. A browser-based tool for visualization and analysis of diffusion MRI data. Nature Communications. doi