From Reactive Graphics to the Agent Era: Reflections on the Evolution of Design and Code
The Reactive Square (1993)

From Reactive Graphics to the Agent Era: Reflections on the Evolution of Design and Code

Looking back at my path through design and technology, I’m often struck by how much the landscape keeps shifting—and how much there is to learn at every turn. In the early 1990s, my curiosity led me to experiment with what I called reactive graphics: programmed visuals that responded in real time to user actions. I wanted to see if computers could display a sense of life, with interactions that felt fluid and engaging. Projects like “The Reactive Square,” which moved in response to sound, or “Flying Letters,” where the mouse controlled type as marionettes, were early attempts to show that computers could be expressive—not just utilitarian. [Link to a short film of the Reactive Books]

Article content
via CD-ROM Journal

At that time, design itself was in flux. Digital tools like Illustrator and Photoshop expanded what was possible, but also brought new constraints, sometimes limiting creativity to the boundaries set by the software. Many traditional designers were skeptical, seeing these tools as a threat to craft and discipline. I understood their reservations, but I was drawn to the potential of computation—not as a replacement for craft, but as another way to explore and make.

This tension led me to create Design by Numbers, a programming environment and book intended to make computation approachable for people who didn’t see themselves as technologists. My hope was to help others experience programming as a hands-on, visual act—not just abstract code, but something tangible you could see and manipulate. I’ve always felt that creative work benefits from understanding what’s beneath the surface, not just what’s on it.

Fast forward to 2025, and the field is changing again—faster than I ever expected. In my recent Design in Tech Report, I tried to capture a moment where artificial intelligence isn’t just a support tool, but a real partner in creative work. We’re now in what I call the “Agent Era,” where AI agents can complete complex tasks on their own, adapting and improving as they go. The cost and speed of experimentation have dropped dramatically, making it possible to test and iterate on ideas at a pace that was unthinkable before.

One of the biggest changes is the move from traditional user interfaces toward what’s now called “agent experience,” or AX. In many cases, users can now reach their goals directly, with AI agents working in the background to make things happen. This changes the designer’s role: instead of crafting every detail manually, we’re setting the intent, flow, and boundaries for AI-driven systems.

I’ve also noticed a trend toward more conversational, less formal coding approaches—often referred to as “vibe coding.” Developers and designers are treating code more like a dialogue, with generative AI tools making software creation more accessible and expressive. The impact and value of this approach are still being debated. It’s not always smooth or predictable, and the field is still learning where and how it’s most effective.

Even with these shifts, I don’t believe AI is replacing designers. If anything, it’s forcing us to focus on what only humans can provide: judgment, empathy, ethics, and the ability to ask the right questions. AI lets us scale and experiment in ways that weren’t possible before, but meaning, care, and resonance still come from human insight and intent.

If there’s a lesson in this journey—from reactive graphics, to Design by Numbers, to today’s agent-driven design—it’s that the tools and platforms will keep evolving, but our job as designers stays much the same. We’re here to guide, question, and shape the work, always trying to bring humanity and purpose to everything we make. —JM


The Reactive Square via CD-ROM Journal / https://cdrom.ca/art/2022/06/14/reactive-square.html

It's fascinating to reflect on the evolution of design and technology from programmed visuals to AI agents. The shift to prioritizing experiences for agents opens up countless possibilities for enhancing user engagement. We resonate with the idea that multidisciplinary thinking is essential in this landscape; combining insights from various fields truly leads to innovative solutions. How do you envision the balance between guiding AI and retaining creative intent in design moving forward? Your insights could spark an important conversation on navigating this new era.

Like
Reply

What a beautiful reflection, John. Your journey from reactive graphics to the agent era continues to inspire all of us who walk the line between code and creativity. The clarity and generosity with which you share your insights—both technical and human—remind us that progress isn’t just about speed or scale, but about meaning and intent. Congratulations on another milestone in shaping the future of design and technology. Proud and honor to call you a friend.

Embracing change is the one true constant in tech, and design is at its most powerful when it’s treated not as a finished product, but as an ongoing conversation. Especially in the Agent Era, the best work comes when we pair new tools with timeless traits: human judgment, empathy, and intent.

It’s a great time to be a designer. AI isn’t replacing creativity. It's removing the friction between ideas and execution.

Like
Reply

Been thinking a lot on this lately too as I work on some new projects — it feels like a momentous shift again to a new era of design, interaction and what it means to “create” when skills of all types are being democratized with AI. Exciting!!

To view or add a comment, sign in

More articles by John Maeda

  • Design in Tech Report 2026: From UX to AX

    This year’s Design in Tech Report is about a major shift for designers. We are moving from UX to AX: from user…

    29 Comments
  • AI Upskilling for Product Designers

    The E-P-I-A-S × SAE Framework I made this system after being asked for one. And I know that by the time I publish it…

    7 Comments
  • Design The Machine

    The real investment isn’t in systems that generate more output. It’s in systems that make output compound.

    7 Comments
  • Get Started with Agent Skills in GitHub Copilot CLI and OpenCode CLI in 2026 on MacOS, Part 1 of 2

    By the time you read this, everything may already have changed. But I thought you might want to get on the skills…

    4 Comments
  • The key to life is earning the right to always be told that, “You have potential.”

    In my 30s, I had the good fortune to work with AARP and to have the opportunity to think about life after 50 when I…

    11 Comments
  • The Prioritization Pyramid In Product Design

    I’ve found the Material Design prioritizing triangle for the Button component to be the singlemost important figure on…

    7 Comments
  • HIL = Humans In the Lifecycle

    The Post-Deployment Challenge of Enterprise AI Indeed’s Chief Economist Svenja Gudell has mapped out GenAI's true…

    7 Comments
  • Let’s Start from the Very Beginning

    A Simple Context Engineering Trick There’s a line from The Sound of Music that has always stayed with me: “Let’s start…

    11 Comments
  • When GenAI Feels More Creative Than Me

    It feels like every morning I wake up to news of another “breakthrough” in generative AI. Today, it’s a realtime video…

    13 Comments
  • AI First Means Humanz First

    I remember the grating sound of the miniature crane we used to lift the wooden tofu molds, each filled with soybean…

    9 Comments

Others also viewed

Explore content categories