Company · Oracis AI LLC

Sheet
A-600
Set
Practice
Status
Issued

A software engineering practice, built by engineers.

Oracis Engineering is the software engineering practice of Oracis AI LLC, a U.S.‑based company. We work with U.S. and international engineers, and we built our own AI‑assisted engineering infrastructure rather than taking one off the shelf.

A-600PracticeIssued

A compass, ruler and pencil on a blank drafting sheet.

How we're set up

Operator-led and deliberately focused. The people who make engineering decisions here have built and run production software themselves, and every engagement gets that attention directly, with no layer of account management between you and the work.

The engineering team

A senior full‑stack engineer leads every engagement and owns the merge. Full‑stack engineers build end to end; mid‑level engineers carry scoped work under senior review.

The engineering team →
Legal entity
Oracis AI LLC, doing business as Oracis Engineering
Track record
20,900+ automated tests, 16,800+ CI runs and 11,000+ commits across three codebases, including a production telehealth platform, out of 55+ repositories in all. How we counted
How we build
Engineers direct the work, and our own system runs bounded AI tasks under hard limits. See How We Work and Responsible AI.

How a change reaches production

How a change reaches production A bounded AI task, scoped and capped, goes to an independent review by a second seat, then to a senior engineer who owns the merge, and only then to production with tests, CI and deploy. Bounded AI task scoped, isolated, capped Independent review a second seat Senior engineer owns the merge Production tests, CI, deploy How a change reaches production A bounded AI task, scoped and capped, goes to an independent review by a second seat, then to a senior engineer who owns the merge, and only then to production with tests, CI and deploy. Bounded AI task scoped, isolated, capped Independent review a second seat Senior engineer owns the merge Production tests, CI, deploy
Fig. C-1Machines do the bounded work; a person owns the merge

Operating principles

01

Evidence over adjectives.

We describe what we checked and what we found. We don't publish claims we can't show you.

  1. 02

    Status, stated.

    Production, built or prototype. We label our own work the way we'd want a vendor to label theirs.

  2. 03

    Humans decide.

    AI does real work here, and a person is accountable for every decision that ships.

  3. 04

    Say no early.

    If we're not the right fit, or you don't need us yet, you'll hear it in the first call.

Talk to an engineer.

Tell us what you're building and what comes next.