Skip to content

Portfolio Projects

Data systems, ML workflows, and research tools.

A curated set of projects showing how I work with messy data, geospatial analysis, OCR, machine learning, and reproducible research workflows.

GeospatialCompleted

April 2026

Solar Energy Site Suitability Analysis

Geospatial decision support for renewable energy planning

A reproducible geospatial suitability model for identifying candidate utility-scale solar locations across Colorado using environmental, infrastructure, and land-use datasets.

Why it matters: Shows applied geospatial modeling, multi-source data integration, and decision-support communication for energy planning.
  • Integrated land use, elevation, protected lands, solar resource, and transmission data.
  • Applied spatial joins, buffering, coordinate transformations, and multi-criteria scoring.
  • Generated maps and visual outputs for energy planning workflows.

Evidence

Suitability scoring workflow
Map outputs
Documented GIS data sources
PythonGeoPandasShapelyPyProjQGISSpatial Analysis
Links coming after repository cleanup
Research ToolsCompleted

April 2026

Historical Economic Geology Education Atlas

Spatiotemporal mapping of academic programs

A geospatial research dataset and visualization workflow showing how economic geology education expanded across U.S. universities over time.

Why it matters: Connects archival data processing with spatial analysis, showing research judgment and technical implementation.
  • Geocoded institutional records and organized historical academic program data.
  • Combined OCR, information extraction, and spatial analytics workflows.
  • Built interactive visualizations for historical and educational research.

Evidence

Geocoded institution dataset
Spatiotemporal visualizations
Reproducible notebooks
PythonGeoPandasOCRGeocodingData Visualization
Links coming after repository cleanup
Data MiningIn Progress

June 2026

University Course Catalog Mining Pipeline

OCR and data extraction for archival research

An end-to-end Python pipeline for discovering, processing, validating, and analyzing historical university course catalogs at research scale.

Why it matters: Demonstrates research software engineering: automation, validation, documentation, and structured dataset creation from messy records.
  • Automated catalog discovery, OCR, information extraction, and validation.
  • Built structured datasets from large collections of archival records.
  • Documented reproducible workflows for downstream research analysis.

Evidence

Before/after OCR examples
Validation reports
Structured CSV or SQL outputs
PythonSQLWeb ScrapingOCRPDF ProcessingETL
Links coming after repository cleanup
AIPlanned

Portfolio roadmap

AI Research Paper Explorer

Planned full-stack AI research assistant

A planned tool for searching, summarizing, and comparing research papers using PDF parsing, embeddings, and source-grounded question answering.

Why it matters: A future project intended to show full-stack AI application development, NLP, retrieval, and evaluation.
  • Planned features include PDF upload, semantic search, and citation-backed summaries.
  • The finished version should include screenshots, a demo link, an architecture diagram, and evaluation examples.
  • Designed to show applied AI, research tooling, and full-stack product thinking.

Evidence

Architecture diagram
Sample paper corpus
Grounded answer evaluation
PythonFastAPIEmbeddingsVector SearchReact
Links coming after repository cleanup