Louis Scalise

I build intelligent systems that do real work.

AI implementation and systems engineering

I find the constraint in an operation, then build a system designed to prove what works.

Method

I don't start with AI.

I start with the problem, the system around it, the assumptions we're making, and the outcome we're trying to produce. Then I build what we need to test whether we're right.

01 / 06 · Problem

What outcome are we trying to produce, and how does it happen today?

02 / 06 · System

Map the people, process, technology, data, timing and ownership around it.

03 / 06 · Hypothesis

Changing X should produce Y, because of Z.

04 / 06 · Build

Something real enough to test the riskiest assumption.

05 / 06 · Evidence

Compare what actually changed with what we expected.

06 / 06 · Learn

Understand why, find the next constraint, and go again.

Fig. 01

A workflow shows the sequence.

This happens, then this, then this. It's usually where we start when we try to understand how work gets done.

Fig. 02

A system shows what the sequence depends on.

The people doing the work. The information they need. The technology connecting it. The timing, ownership and decisions between each step.

Fig. 03

The failures usually live between them.

A record falls between systems. Information arrives too late. Ownership changes without anyone noticing. Every part can be working exactly as designed while the system still fails.

Work.

Work · 01

Business Diagnostic

Finding where an operation stops matching the process everyone thinks is happening. Follow one lead through a real pipeline, from raw export to a finding that has to earn its confidence.

BuiltSynthetic data
Work · 02

Dealer Flywheel

Operational intelligence and technology implementation for automotive dealerships. Built to find where performance breaks between systems, people, information, timing and ownership, then help implement and measure the fix.

In development
Work · 03

AI Recruiting System

An assistant runs first contact and screening. Rules hold the hard requirements, every recommendation keeps a decision trace, and a recruiter decides anything uncertain.

PrototypeSynthetic data
Work · 04

Find the Constraint

Automate any step of a pipeline and watch what happens to the output.

Experiment

I'm Louis Scalise.

I spent more than 15 years in sales, marketing, finance and business operations before studying AI engineering. Today I combine that experience with hands-on systems development.

I'm interested in difficult problems at the intersection of business, technology and AI.

More about me →

Currently.

Building
Dealer Flywheel: a system for finding operational failures across people, software, timing and ownership.
Testing
Business Diagnostic: whether operational event data can produce findings a manager can independently verify.
Exploring
AI evaluation, retrieval, agent state, system boundaries and failure handling.
[email protected]