Joshua (Stapley) Montoya

I’m a data-focused engineer with a background in systems engineering, handling data workflows and integration across complex systems. I have a strong interest in geospatial data, urban development, and the use of data to understand and visualize real-world systems.

I hold a Master of Computer Science from the University of Illinois Urbana-Champaign and currently work as a systems engineer at L3Harris Technologies, supporting reliability and logistics data and analysis for large-scale aerospace programs.

Currently open to work in Los Angeles, CA.

Email  /  Resume  /  Github  /  LinkedIn

profile photo

Selected Data Projects

NYC Building Permits
Joshua Montoya
2025
project code

Daily batch pipeline ingesting NYC building permits data from the Socrata Open Data API into BigQuery, enabling geospatial analysis and visualization of construction activity across the city. Orchestrated ingestion and transformation workflows using Airflow. Provisioned cloud infrastructure with Terraform (GCP). Stored raw data in GCS and modeled structured datasets in BigQuery. Built dashboards in Looker to explore spatial and temporal trends.

Plant Monitoring IoT Device
Joshua Montoya
2022
implementation details / project code / short demo

Battery-powered IoT system for collecting environmental sensor data and transmitting it via Bluetooth LE to an iOS application. Designed and implemented embedded system using Arduino and Python. Built mobile interface in Swift to display time-series sensor data. Focused on low-power operation and reliable data transmission.

Forest Species Segmentation in Aerial Imagery of Wasatch Mountains Using 3D Convolutional Neural Network
Joshua Montoya
2023
implementation details / zoomable segmented map / project code

Machine learning pipeline for classifying tree species from aerial imagery of the Wasatch Range using a 3D convolutional neural network. Created labeled training data from NAIP imagery. Trained CNN model on hyperspectral image cubes. Generated geospatial prediction maps and built a web interface for visualization. Achieved up to 84% accuracy on unseen test data.


source for webpage design