🚀 Scientific Software Engineer | Python • HPC • Data Pipelines • ML

I build scientific software, data pipelines, and compute systems that transform complex datasets into reliable, reproducible workflows.

My work spans high-performance computing, GPU cloud infrastructure, neuroimaging, clinical prediction, astronomical data processing, and large-scale scientific data engineering across research and production environments.

My work sits at the intersection of research software engineering and production infrastructure, turning complex technical and scientific requirements into maintainable systems.

• 🧠 Psychiatric EHR analysis and clinical prediction pipelines
• 🤖 Machine learning workflows using LightGBM, calibration, cross-validation, and SHAP
• 🔬 Causal inference and causal discovery workflows
• 🗺️ GIS-linked EHR analysis and social determinants of health
• ⚡ Reproducible scientific computing and HPC workflows

Reusable Python/Bash wrappers enabling scalable execution of large data-processing workloads across HPC environments.

Tech: Python • Bash • SLURM • HPC
🔗 https://github.com/DCAN-Labs/SLURM_wrappers

Containerized deep learning application improving performance and accuracy for large-scale neuroimaging pipelines.

Tech: Python • Deep Learning • Containers • HPC
🔗 https://github.com/DCAN-Labs/BIBSnet

Python-based astronomical image-processing pipeline for detecting transiting exoplanets from time-domain FITS observations, including CCD calibration, astrometric processing, aperture/differential photometry, and light-curve generation.

Tech: Python • Astropy • Photutils • Astrometry.net • FITS • Scientific Computing

FAIR-compliant container-linking system enabling reproducible scientific pipelines across heterogeneous compute environments.

Tech: Python • Docker • Singularity
🔗 https://github.com/DCAN-Labs/CABINET

Enterprise-style automation platform integrating compliant web scraping, territory mapping, structured logging, and workflow alerting.

Tech: Python • APIs • ETL • Automation • Data Validation

Google Apps Script tools automating workflow tracking, dynamic link generation, and real-time updates.

Tech: JavaScript • Google Apps Script • Node.js

My work has contributed to 30+ scientific publications spanning astronomy, neuroimaging, psychiatry, and scientific software.

📚 Google Scholar

Primary: Python • Bash • Linux/Unix

Additional: R • JavaScript • MATLAB • Go • SQL

Pandas • NumPy • Statistical Modeling
Predictive Modeling • LightGBM • Calibration
Cross-Validation • SHAP • Causal Inference
Benchmarking • Data Validation

Kubernetes • SLURM • Docker • Singularity
AWS S3 • Ceph • Distributed Compute Systems

Grafana • Prometheus • Linux/Unix Systems

ETL Pipelines • API Integrations
Workflow Automation • CI/CD Concepts
GitHub Actions

✅ Provisioned and maintained thousands of NVIDIA GPU nodes in production Kubernetes clusters
✅ Built pipelines processing petabyte-scale datasets
✅ Achieved 600× performance improvements in imaging workflows
✅ Reduced runtimes 10–12× through HPC optimization
✅ Designed automation systems reducing manual operational overhead
✅ Led documentation & reproducibility initiatives across multi-team environments
✅ Contributed to 30+ scientific publications across astronomy, neuroimaging, psychiatry, and scientific software

• Scientific Software Engineering
• Research Software Engineering
• Scientific Data Engineering
• High-Performance Computing
• Research Computing
• Machine Learning for Scientific Applications
• Astronomy & Space Software
• Python Software Engineering

• 💼 LinkedIn: https://linkedin.com/in/audreyhoughton
• 📧 Email: audreymhoughton@gmail.com