Designing
AI people
can trust.

I'm Steph Campagnoli, an AI product and systems designer. I find the problems people have learned to work around, like memory that invents things or work that was never really saved, and design them into behavior people can rely on.

I design the behavior.
AI writes the code.

  1. 1

    Define

    Name the problem people work around and what it costs them.

  2. 2

    Specify

    Write the behavior, the boundaries and how failure should look.

  3. 3

    Direct

    Guide AI coding agents through the build.

  4. 4

    Review

    An independent reviewer has to pass every release.

  5. 5

    Accept

    Only what I can open, use and verify myself.

Three questions
I ask first.

What matters here?

Before features: the thing people can't afford to lose. Their words, their decisions, their say.

What repeats?

The friction someone describes twice is the requirement nobody wrote down.

What is still missing?

The gap between what a system claims and what you can open and check.

From the
studio chair
to systems
design.

For seven years I ran my own studio. The real work was discovery: hearing what someone actually needed, noticing friction before it turned into a goodbye, and turning expertise into choices people could make with confidence.

In 2026 I needed an AI system that didn't exist, so I designed it. StephOS taught me that trust in AI is a design problem, and that's the work I want to keep doing, on a team building AI people rely on.

150+in-depth consultations a year
7years running my own studio
3published case studies
Read my résumé

The system behind
the work.

StephOS is my AI assistant and recorder, built around one question: how do you make AI memory trustworthy? Original words are kept as evidence, nothing becomes approved without the person, and memory only reaches what it has been allowed to.

A still from the living portfolio: gold trails of light connecting points that stand for ideas, decisions and evidence
Coming soon

Explore how my
ideas connect.

A living map of the thinking behind StephOS: 39 real ideas, decisions and pieces of evidence, and the trails between them. Each connection lights up where two ideas meet. Hover or tap to see what's still uncertain.

Let's build AI that
people rely on.

I'm looking for my next role in AI product and systems design: memory, agents, human-in-the-loop. Remote, or hybrid near Savannah. Write to me at hello@stephos.studio.