I am an environmental computing researcher with training in quantitative economics. My work examines how machine learning, geospatial systems, remote sensing, and field-generated data can be used to understand environmental systems and support real-world decision-making.

My current research focuses on the reliability of environmental AI, particularly how sensing conditions, preprocessing decisions, human data collection, and implementation choices influence what computational systems ultimately observe and report.

Website · CV · LinkedIn · ORCID

I am currently working with Dr. Kala Fleming and Dr. Elizabeth Ondula on three interrelated projects spanning environmental computer vision, field-grounded AI, and geospatial decision systems.

Studying how the origin of a sliding-window inference grid affects individual tree-crown detections while tile size, overlap, and model configuration remain fixed. The work examines reproducibility and implementation sensitivity in environmental object detection.

Investigating how field observations, structured annotations, drone imagery, and physical interventions can reduce ambiguity in environmental computer-vision workflows and produce more reliable inputs for machine-learning systems.

Developing spatial data systems that integrate environmental, infrastructure, and community datasets to support green stormwater infrastructure analysis and real-world decision-making.

_{These projects are ongoing.}

A configurable decision-support system combining federal labor-market data, institutional data, scoring logic, and an interactive R Shiny interface to evaluate academic programs across demand, institutional, and financial dimensions.

Python R Shiny YAML Plotly BLS OEWS IPEDS O*NET

Repository: gabrielpriante/MQE-Capstone

Python R SQL ArcGIS Pro GeoPandas OpenDroneMap DeepForest scikit-learn R Shiny Git