Applied AI Engineer · Denver

CONNORO’Dea

I build AI-native products end to end — architecture through ship. Forty-six shipped across seven industries in the last eight months.

Denver
01The log
Live from GitHub

Forty-six shipped,
seven industries.

Shipping log · 46 public repositories · 7 domains · 8 monthsSource: github.com/connorodea →
Video / Media
Infrastructure
Commerce
Real Estate
Health
Finance
Consumer
DecJanFebMarAprMayJunJul
Hover a mark — every one links to its repository
02Selected work
Eight of forty-six

The evidence.

01

Cutroom

An AI-native video editor that cuts from the transcript.

Video / Media
In buildRead →
02

Sightline

Tells Shopify merchants when AI recommends their store.

Commerce
In buildRead →
03

Juricratic

Litigation strategy modeled as a game, not a guess.

Legal
In buildRead →
04

APIDistributed

An API marketplace built for agents, not portals.

Infrastructure
In buildRead →
05

GeoStamp

Geotags photos for local SEO in seconds.

Commerce
LiveCode →
06

Lockrooms

Multi-sport film study as a subscription.

Video / Media
In buildPrivate
07

MetroDealReport

A publication for real-estate investors, not a listings feed.

Real Estate
In buildPrivate
08

ScreenRec Studio

A native macOS screen recorder and editor.

Video / Media
In buildCode →
03What I do

What I can build for you.

01

Zero to shipped

You have a product that does not exist yet. I take it from a conversation to something in front of users — architecture, build, deploy, and the unglamorous parts in between.

02

AI where it earns its place

Retrieval, agents, and model pipelines built into a product because they make it work better, not because they are on the roadmap. I have shipped enough of these to know which ones do not pay off.

03

Systems that outlive the demo

Real deploys, real monitoring, real data models. The difference between a prototype that impresses and software that runs is mostly work nobody sees.

04About

I am an engineer who ships.

Most of what I build starts as a problem I ran into myself, or watched somebody else lose a week to, and ends as software that is actually deployed and actually used.

The range is deliberate. Working across video, infrastructure, commerce, health, consumer, real estate, and finance means the patterns start to rhyme — a retrieval problem in litigation looks a lot like one in property data, and the solution carries over. That is where the speed comes from. Not shortcuts, transfer.

If you are building something and the hard part is that it has never been built before, that is the work I want.

Connor O'Dea
Based
Denver, Colorado
Focus
Applied AI, 0→1 product
Domains
Seven, and counting
Also
odeaworks.com, preecursor.com
© 2026 Connor O’Dea · Denver, CO