One of the ways I learn best is to understand things in a conceptual way and gain enough theoretical knowledge on the topic, then I have to try and build something. This page provides information about two current projects I am using to help me learn about AI engineering, DevOps, and cloud architecture. Each of these projects represent a solution to an identified problem from the field of Emergency Management.

DISCLAIMER: I can not take credit for every piece of code in these systems - I am still learning - and have substantially benefitted from various commercial genAI models. The genAI tools have supported my learning by walking me through the development using a process much like I would use in planning other projects: through a work breakdown structure and aligning resources to the work packages. The decisions are mine, based on feedback from the genAI tools to help me understand trade-offs among effeciency, security, privacy, and other key factors.

These projects are in progress; future site updates will include more detailed information and diagrams regarding architecture, trade-offs, and overall system design.

Chemical Incident Response Platform Simulation (ChIRPS)

This project is a web-based simulation and training platform designed to model complex chemical, biological, radiological, and nuclear (CBRN) emergency response scenarios. The system enables facilitators to run realistic, multi-agency incident exercises while capturing detailed, time-ordered decision-making and operational actions.

The platform is intended to support:

  • Emergency management training
  • Multi-agency coordination exercises
  • After Action Reports (AAR)
  • Decision analysis and replay
  • Auditability and compliance-oriented recordkeeping

ERES: Emergency Response & Evaluation System

In Emergency Management, Emergency Operations Plans (EOPs) are critical but cumbersome. Often spanning hundreds of pages, these PDFs are difficult to navigate under the high-stress conditions of an active incident.

ERES is a local-first, high-performance RAG (Retrieval-Augmented Generation) pipeline designed to ingest complex EOPs and provide instant, grounded answers to field responders.