What Happens When Designers, Product Thinkers, and AI Build Together?

Summary

At Knapsack Patterns Denver, our team used AI-powered tools to take an idea from concept to a deployed prototype in roughly two hours. The experience showed how AI can accelerate delivery while also reinforcing the importance of human expertise, collaboration, and systems thinking throughout the process.

I attended Knapsack Patterns Denver, an event focused on design systems, AI, and the future of digital product development.

I’ve attended plenty of conversations about AI over the past few years. Many of them focus on possibilities, predictions, or concerns about what might happen someday.

This experience was different. Instead of talking about AI, we built with it.

Over the course of a few hours, teams of designers, product thinkers, engineers, and digital practitioners were challenged to identify a problem and create a working solution. What followed was one of the most interesting collaborative exercises I’ve experienced in a long time.

The Challenge We Chose

As our group began discussing common frustrations across organizations, a familiar theme emerged.

Information is everywhere.

• Design systems contain component documentation and usage guidelines.
• Jira contains requirements, decisions, and delivery history.
• Governance documents explain policies and standards.
• Knowledge bases contain operational procedures.
• Repositories contain implementation details.
• Teams carry additional knowledge in conversations, meetings, and institutional memory.

The problem isn’t that organizations lack information.

The challenge is that finding the right information at the right moment often takes more effort than it should.

Product Owners, designers, engineers, and operations teams frequently need answers that already exist somewhere within the organization. The friction is knowing where to find them.

Our Idea: A Design System Intelligence Hub

Our team envisioned an AI-powered interface that could connect to the systems organizations already use and provide role-specific guidance based on who is asking the question.

Rather than searching through multiple tools, a team member could simply ask for what they need. The system would understand both the question and the context of the person asking it.

• For Product Owners
  • A Product Owner might be defining requirements for a new feature and want to understand whether a similar component already exists.
  • Instead of digging through documentation, the system could identify relevant components, surface previous decisions, highlight constraints, and suggest considerations for requirements, acceptance criteria, and implementation planning.
• For Designers
  • A designer exploring a new interaction pattern could ask whether similar solutions already exist within the design system.
  • The system could recommend related components, identify established patterns, and help maintain consistency across the product ecosystem.
• For Engineers
  • An engineer could ask whether existing components, services, or implementation patterns already solve part of the problem.
  • The system could surface reusable assets, technical documentation, and implementation guidance, helping reduce duplicate work and accelerate delivery.
• For Operations and Content Teams
  • Teams responsible for implementing content, configuring pages, or maintaining digital experiences could verify whether proposed changes align with design system standards.
  • The system could identify potential compliance issues before changes are published and help teams understand implementation requirements.

The goal wasn’t simply to create another search tool. The goal was to create a conversational layer that connects organizational knowledge and presents it in a way that is meaningful to the person asking the question.

Why This Matters

As organizations grow, knowledge becomes fragmented. The result is often duplicate work, inconsistent implementation, slower onboarding, and communication gaps between teams.

A system like the one we envisioned could help address several common challenges.

• Reduced Duplicate Work
  • Teams spend significant time recreating solutions that already exist because they cannot easily discover them.
  • Making existing knowledge more accessible helps teams build on prior work rather than starting from scratch.
• Better Design System Adoption
  • Design systems only create value when people can find and use them.
  • By making guidance conversational and contextual, organizations can lower the barrier to adoption.
• Faster Onboarding
  • New team members often struggle to understand where information lives.
  • Providing a single entry point for organizational knowledge could dramatically reduce ramp-up time.
• Improved Governance
  • Policies, standards, and implementation guidance become more effective when they are accessible at the moment decisions are being made.
• Better Cross-Functional Communication
  • Perhaps most importantly, a shared knowledge layer can help bridge communication gaps between product, design, engineering, and operations teams.

From Whiteboard to Prototype in Two Hours

The most fascinating part of the experience wasn’t the concept itself. It was how quickly we were able to bring it to life. Our team moved from discussion to requirements, from requirements to prototype, and from prototype to a working deployed experience in roughly two hours.

For me personally, part of the excitement came from working directly with tools that were new to me. I set up Git, created a repository, worked with Claude Code, and watched ideas become functioning software at a pace that would have felt impossible only a few years ago.

What stood out wasn’t that AI generated code. What stood out was how AI accelerated collaboration. Questions were answered faster. Ideas were tested faster. Concepts became tangible faster. The technology didn’t replace the team. It amplified the team’s ability to move forward.

What I Learned

The experience left me thinking about several larger trends.

• AI as Collaboration Layer
  • Much of today’s conversation focuses on AI as a content generation tool. What I experienced felt different. AI acted as a connector between people, systems, and information.
  • Its greatest value may not only be in creating new knowledge, but also in helping organizations access and apply the knowledge they already possess.
• Design Systems Are Evolving
  • Design systems have traditionally been viewed as collections of components, patterns, and documentation. Increasingly, they are becoming organizational knowledge systems.
  • The future involves making organizational knowledge accessible and actionable.
• Product Thinking Becomes More Important
  • When information becomes easier to access, and implementation becomes faster, understanding the problem becomes even more important.
  • The ability to define goals, identify constraints, prioritize outcomes, and align teams remains fundamentally human work.
• Context Is Becoming as Valuable as Content
  • Organizations already have enormous amounts of information. The challenge is delivering the right information to the right person at the right time.
  • Context is becoming one of the most valuable assets an organization can create.

Final Reflection

The most interesting outcome of the afternoon wasn’t the prototype. It was the realization that the future of AI may have less to do with replacing designers, Product Owners, engineers, or operations teams and more to do with helping them work together more effectively.

The organizations that benefit most from AI may not be the ones generating the most content. They may be the ones that reduce the most friction.

If AI can help people find the information they need, understand it in their own context, and collaborate more effectively across disciplines, that’s a future worth building toward.