Shared baseline
Create a common language for what AI can do, where it fails, and what should stay human-owned.
Human in the Middle
Practical AI consultation and training for small and midsize companies
Streamline workflows, reduce tedium, and free up human energy for the work that matters.
Human in the Middle helps companies use AI in a way that is practical, controlled, and genuinely useful. The goal is not blind automation. The goal is better workflows, stronger review habits, and more energy for the judgment-heavy work that should stay human.
My background in design, UX, and systems thinking helps me build AI workflows, automations, and learning formats that are not just powerful on paper, but understandable and usable in practice.
Companies can bring me tool questions, workflow questions, pipeline decisions, or team adoption friction. I can help establish the right AI-supported workflows first, then train staff to use them well.
Problem
What usually goes wrong
AI adoption often starts in the wrong place. Tools arrive before standards. Curiosity outruns judgment. Managers want ROI, teams want clarity, and nobody agrees on what should stay human-owned.
- Teams use AI in scattered, inconsistent ways.
- Prompting tricks get mistaken for real workflow design.
- Review quality stays weak while output volume rises.
- Leaders feel pressure to move fast without creating chaos.
- Governance becomes either too loose or too bureaucratic.
Promise
What this work is designed to do
The goal is not to make your company dependent on AI. The goal is to help your people work with it more intelligently.
Repeatable workflows
Turn experimentation into systems that can actually be used, reviewed, and improved over time.
Better energy use
Reduce avoidable drag so people have more capacity for judgment, relationships, creativity, and care.
How the offer works
The offer spans consultation, workflow design, live labs, team adoption, and more advanced builds. Some companies start with tool choice and process questions. Others start with a live lab. The important part is finding the right entry point for the actual work.
Foundation
Build a shared baseline, reduce fear, and create the first repeatable workflow wins.
Consultation
Clarify tool choices, workflow priorities, pipeline logic, and where AI should or should not sit inside the work.
Labs and Training
Teach teams through live, hands-on work so the new workflow becomes usable, reviewable, and easier to carry forward.
Leadership Rollout
Align decision-makers, create reusable leadership workflows, and build early operational proof of value.
Team Adoption
Turn scattered AI use into shared operating standards, review logic, and adoption discipline.
Specialist Builds
For smaller advanced groups ready for supervised workflow automation or agentic experimentation.
Framework
How Human in the Middle works
Judgment Before Automation is the philosophy. Human in the Middle is the executional framework inside it.
Judgment Before Automation explains the rules. Human in the Middle shows how those rules get built into real work.
Human in the Middle operationalizes that by defining ownership, checkpoints, escalation paths, and boundaries before speed becomes the only value left in the system.
The executional framework is also a UX question. If a workflow is confusing, brittle, or unpleasant to use, people will bypass it. Good AI adoption is not just about the model or the prompt. It is about the experience of the system around it.
Read the philosophyWhy Arthur
Why I do this work
I came to this work as a designer, but not only as a designer.
I have been self-employed for most of my life. That means I have spent years dealing not just with creative output, but with the practical realities around it: winning clients, caring for them well, marketing work clearly, prioritizing limited time, and leading from responsibility rather than hierarchy.
When AI started reshaping creative and knowledge work, I chose to study it deeply instead of standing at a distance from it. What I found was not a magic replacement for human skill, but a set of tools that becomes powerful only when paired with clear thinking, strong review, and good system design.
My UI/UX and design-systems background gives me a specific angle on this work. I naturally think about journeys, interfaces, mental models, friction points, trust, and the difference between a powerful system and a usable one.
I am not presenting myself as someone who has already solved everything. I am still exploring this territory myself. But I have learned enough through practice to help others explore it more clearly, more critically, and with better structures around them.
The point is not to automate people out of meaning. The point is to remove avoidable tedium so that teams have more energy left for the real work.
Next step
Ready to make AI useful?
If your company wants a clearer, more responsible way to work with AI, we can shape the right path together. That may start with a foundation lab, a leadership rollout, or a more focused team intervention. The important part is starting from your actual work, not from generic hype.