Harness Design: the discipline that will redefine product design

Many companies with product design teams are trying to use artificial intelligence to speed up their work. Over the last few years, this work has increasingly become building interfaces for software: components, screens, flows, journeys. Experienced designers know design goes far beyond that, but this dense part, the visual construction, has become the biggest slice of the day to day.

Now, AI is structurally changing this scenario. I want to talk about what we are living in practice at Prolog, and about a term I believe will become part of the market's vocabulary: Harness Design.

The metaphor that changed the way I think

Everything became clearer when I read Why Your "AI-First" Strategy Is Probably Wrong, by Peter Pang. The central argument is simple and devastating: most companies use AI as an auxiliary tool, when the right move is to put it at the center of production, and redesign everything around it.

The designer is no longer the one who uses AI. AI produces, humans build the production line, feed it with context, and review what it generates.

This concept gained even more depth when Luiz Felipe (CTO of Prolog App) shared with the team OpenAI's article about Codex, Leveraging Codex in an agent-centric world, written by Ryan Lopopolo. It introduces the idea of the agent at the center, with humans orchestrating around it. This is not science fiction. It is the model we are trying to build now.

From Design System for humans to Design System for machines

Today I see many design teams trying to optimize their design systems for tools like Figma Make or Lovable, still under the conventional view of a designer building a prototype so a developer can later take it to the final front end. That view is conventional, just with a new tool.

What we are building now is different. We are no longer optimizing a library so a designer can look at it, choose, and know where to use each component, or so Figma Make can generate several ideas. We are creating a whole system designed so that a robot can consume it.

This means documenting with intent. It is not enough to name a button: you need to explain when to use it, in which context it makes sense, when it must not be used, and why. Each token, each variable, each component needs documentation robust enough for an LLM to make good decisions on its own.

The article Your Next Design System User Is an AI Agent, by Kartva, which Glauber Franco brought to the team and was, without exaggeration, gold for us, gave us a much clearer map of how to make this transition in practice. It confirmed what we were already intuiting and shed light on points that were still unclear about how to optimize a Design System for real LLM use.

What is Harness Design

Harness has two meanings that fit together perfectly here.

The first is harness (the gear): the structure that directs an animal's strength without limiting it. A harness channels the horse's power to where you need to go. Without it, AI gallops wherever it wants. With it, AI goes where the product needs it to go.

The second is the verb to harness: to take advantage of, to channel. As in "harness the power of the wind", turning raw energy into something useful and controlled.

Harness Design, then, is the discipline of designing documented design systems, with intent, so that LLMs can make good autonomous interface decisions. It is the designer moving from being a builder of screens to becoming the architect of the context the AI will consume.

The new role of the designer: three positions

In this model, the designer works in three distinct moments, and all of them are critical.

Before: They define the intents. They are the great librarian of the design system, creating the structure and documentation that allow the LLM to make good decisions. Design is designating: giving intent. And that is exactly the skill machines lack. Each software has its particularities, each business has its nuances. The designer needs to understand all of this and translate it into a language the machine can process. The focus is no longer only on form: the function of each element becomes central.

During: They lead the machine. Just as tech leads and managers removed blockers and secured resources for human teams, now we all need to learn to do this for LLMs: ensure enough context, remove blockers, and provide what the AI needs to produce well. This layer already has a name in engineering: Harness Engineering. What I am proposing is that it also exists in design.

After: They review and curate. The designer remains the guardian of the customer experience. They look at what the AI produced, check whether it makes sense, whether it delivers real value, and whether it is aligned with the needs of the business and of the user, including everything that lives outside the documentation, in people's heads.

And there is more: the designer can now also build the front end directly. Using Claude Code or similar tools, they create functional interfaces and hand this front end to a developer to refine: checking security gaps and structural problems, and making the connections with the back end. The handoff stops being a Figma full of ambiguities and becomes real code. The developer gets an even more technical and strategic role. Nobody loses ground; everyone changes position.

This is already happening on our team

I am not talking about theory. I am talking about what we are already building.

Gustavo Navarro is already building production lines with multi-agents and skills in Claude Code, a practical demonstration that this model is not the future, it is the present. Glauber Franco constantly brings new knowledge about AI applied to design, and he is the one who brought the Kartva article that greatly accelerated our view of Harness Design.

Together with Diogenes Vanzela, we are building a system with AI at the center for the upstream phases (problem definition and discovery), where PMs and designers work together. It is one of the most exciting fronts we are pushing right now. And Hegon Damazo is already getting concrete results using skills and Claude to optimize PM work: discovery, documentation, planning, and problem framing.

Many things are emerging at the same time, and what impresses me is how fast people are learning and sharing.

Design as an industrial discipline, back to its origins

There is something very coherent in this transformation when we go back to the origins. Design has always been about creating artifacts that can be produced at scale: scalable systems, cheaper processes, quality standards, faster delivery. The design system work is exactly that.

What changes is that the scale is no longer for other designers to use. It is for machines to produce.

Is there anxiety in all this? Yes.

There is fear, FOMO, and real uncertainty. LLM costs are still high and may rise further. Some AI companies are not yet profitable. Bubbles can form, and they can burst. But there is also courage to move forward, learn from the experiments, and understand how this new scenario can actually work.

The designer of the future will not disappear. They will become the architect of the intelligence that produces. And that, deep down, is a return to the roots of the discipline: deeply understanding the problem, defining intent with clarity, and making sure what reaches the customer really makes sense.

Harness Design is the name that today, for me, describes what is already happening, and that will become increasingly central to anyone who works with product.

Are you experimenting with something similar on your team? I really want to hear how other designers, engineers, and PMs are navigating this change.

Attachments

Portuguese Medium: https://johnmorais.medium.com/harness-design-a-disciplina-que-vai-redefinir-o-design-de-produto-64fd9f5723bf?postPublishedType=repub

English Medium: https://medium.com/@johnmorais/harness-design-the-discipline-that-will-redefine-product-design-24ee80c0e38a

https://youtu.be/rQfz3GINEaw

Concept or Reference	Role in the AI Ecosystem	Definition or Application
Harness Design	Discipline of designing documented design systems, with intent, for autonomous AI decisions. It is not limited to that: it involves processes, tools, people, and governance, everything a team needs to work AI First.	Channeling the power of AI through contexts and rules, turning the designer into a context architect.
Your Next Design System User Is an AI Agent	Transition from a Design System focused on humans to a model focused on AI agents.	Map to optimize component libraries and variables (tokens) for consumption and interpretation by LLMs.
Leveraging Codex in an agent-centric world	Introduces the model of the agent at the center of operations, with humans orchestrating around it.	Using Codex to create an environment where the machine executes tasks and the human acts as the orchestrator.
Multi-agent production lines	Automation of the integrated design and code production line through multiple agents.	Using skills in Claude Code to create workflows where AI agents collaborate with each other.
AI in Upstream (Discovery)	Structural support in defining problems and in the product discovery phase.	A system with AI at the center to help PMs and designers understand and define the initial strategy.
PM task optimization with Claude	Use of artificial intelligence to automate product management.	Application of skills to optimize discovery, documentation, planning, and problem framing.
Why Your "AI-First" Strategy Is Probably Wrong	Puts AI at the center of production, instead of treating it only as an auxiliary tool.	Redesign the product strategy with AI as the main engine, where humans build the production line.
Harness Engineering in Design	Role of technical leadership and removing blockers for the machine's production.	Ensure enough context and remove technical blockers so the LLM produces interfaces smoothly.