John V.
Kenny

Municipal governments sit on decades of as-built drawings, handwritten service records, and construction documents that are trapped in paper or unstructured PDFs — invisible to the GIS systems staff rely on every day.

Building a suite of computer vision and AI pipelines that extract structured data from these documents at scale — from vision language models that read handwritten water service cards to automated as-built ingestion systems that pull metadata, organize files, and feed results directly into ArcGIS Pro catalogs.

Decades of paper became a living, queryable system — from scanned drawing, to georeferenced GIS feature, to natural-language answer. For the municipalities it serves, that means higher-quality infrastructure data with traceable lineage back to the source record — and with it, smoother day-to-day operations and the decision-readiness that comes from trusting the data underneath you. Building it stretched me across the full stack along the way: computer vision, model training, cloud architecture, agentic AI, and RAG.

Emergency dispatch lives or dies on accurate GIS — addresses that geocode, routes that work. When the data is wrong it causes real-world failures, and catching those problems stays reactive and manual until something breaks during a call.

I built a scripted Python toolbox that runs a series of seven automated health checks against the dispatch datasets we deliver. Before data ever goes out the door to a dispatching client, the toolbox systematically scans it for the structural problems that break geocoding and routing before they cause a failure in the field — making data-quality assurance a repeatable internal step in our delivery process, not reactive firefighting after something breaks.

The dispatch data we hand off is verified sound before it reaches production. For the dispatch centers and fire districts that depend on it, that means fewer failed address lookups and routing errors when a call comes in — caught by our own consistent QA process instead of discovered mid-emergency.

Public crime data is often locked in formats no resident can use — and publishing it openly raises real privacy concerns about pinpointing incidents to a single household.

I built a Python pipeline that ingests incident data, anonymizes each record to the 100-block level to protect privacy, and auto-publishes a clean public dashboard to ArcGIS Online — replacing a manual update process with a repeatable, scheduled one.

What took 60–90 minutes of manual work per update now runs in about 5 — freeing department staff from repetitive publishing while giving residents a transparent, privacy-respecting, always-current view of local incidents. For the municipality, that’s both reclaimed staff capacity and stronger public trust.

Thousands of municipal address records needed verification before large-scale system migrations, with no efficient way to catch invalid or partially correct entries.

Built an internal validation tool that checks records against the official USPS database, with an interactive dashboard, map visualization, and pause/resume for multi-day jobs.

By validating records against the official USPS database before a migration, the tool protects the address data that municipal billing, permitting, and dispatch all rely on — catching errors at the source instead of after they reach production, and sparing staff hours of downstream cleanup.

The City of Lake Forest needed residents and visitors to understand downtown parking at a glance — including “how far is this lot from where I'm going?” — and off-the-shelf ArcGIS Experience Builder couldn't deliver that experience.

I designed, built, and deployed a custom React + ArcGIS app: every CBD lot mapped with color-coded restrictions and detailed profiles, on-demand walk-time routing from any origin, and a mobile-first interface — all driven by a single JSON config that makes it portable to other communities.

MGP's first custom-coded GIS web application — delivering functionality off-the-shelf tools couldn't, and now a reusable template already carried forward to a second municipality. A first-of-its-kind standard for the firm.