I’m Cort — a software engineer with 25+ years building systems for Fortune 500
companies and startups across finance, energy, healthcare, and marketing.
For most of that time the work was conventional engineering: design the system, build it,
test it, ship it, keep it running. When AI models got good enough to do real work, I
started by putting one at the center of an application instead of bolting it on: a model
scoring rental photos in Apartment Finder, another writing one summary per real-world event
in Corticorp News. Then the projects outgrew a single model. Finance became 44 simulated
fund managers, each with its own discipline and its own published reasons, and keeping two
sites honest every day became the job of a staff: a Board that reads the code, a Newsroom
that reads the data, and me deciding what ships.
That is what CortiCorp builds now. When this site says “we”, it means me and
the organizations I direct, not a large human staff. And none of it works without the old
disciplines: data models, tests, audits, deploy paths and someone accountable for every
change. The agents do the work. The engineering is what makes the work worth trusting.
AI organizations
Systems where Claude models hold defined roles — writing,
scoring, auditing and reviewing each other’s work — with a human decision
point and a public record of what each one said.
Data Engineering
Scraping, cleaning, enriching, and visualizing data from messy real-world sources.
Turning unstructured listings, photos, and text into searchable, scored, actionable
dashboards.
Audio & DSP
Real-time audio synthesis in C++ with JUCE. Oscillator design, filter modeling,
physical-string simulation, zero-allocation audio threads, and shipping a
full VST3 plugin.
Full Stack
End-to-end: Python backends, interactive web dashboards, REST APIs, cloud
deployment, and desktop applications. Whatever the project needs to ship.
