Accessibility Tools

Language / Translation

Text & Reading

Text Size
Line Spacing

Focus & Motion

Back to Home

Our Methods

How we use play, prototyping, and iteration to build AI tools that address real-world community issues.

Play as Practice

We believe that the best civic technology emerges from a spirit of play and experimentation. Rather than treating technology development as a rigid, top-down process, we embrace curiosity, tinkering, and creative exploration.

Play allows us to ask "what if?" without the pressure of perfection. It creates space for unexpected solutions and helps us understand community needs in ways that formal requirements gathering never could.

When we play with ideas, we discover possibilities that structured processes would never reveal. Our tools are born from genuine curiosity about how technology might serve neighbors, not from product roadmaps.

1

Explore

Discover community needs through conversation and observation

2

Experiment

Build rapid prototypes to test ideas quickly

3

Evaluate

Test with real users and gather honest feedback

4

Iterate

Refine, improve, and sometimes start over

Micro-Prototyping

Instead of building monolithic applications, we create small, focused tools—what we call "micro-apps"—that solve specific problems for specific communities. These tools are designed to be:

Lightweight: They run on modest hardware and don't require expensive infrastructure.

Modular: They can be combined, adapted, or replaced as needs change.

Fast to deploy: We can get working prototypes into community hands within days, not months.

This approach lets us iterate faster than traditional civic technology development, while still maintaining the safety and inclusivity required for public-serving applications.

// Our development philosophy
const approach = {
buildTime: "days, not months",
scope: "focused, not bloated",
feedback: "continuous",
failure: "expected & valued",
ownership: "community"
};

AI as Tool, Not Author

We use AI extensively in our development process—but in a very specific way. AI helps us code, but humans generate the ideas.

This distinction matters. The challenges facing our communities are deeply human: housing, accessibility, civic engagement, democratic participation. These require human understanding, empathy, and lived experience to identify and frame correctly.

Once we've identified what to build and why, AI becomes a powerful accelerator. It helps us write code faster, debug issues, and explore technical possibilities. But the vision, the values, and the priorities remain firmly in human hands—specifically, in the hands of the communities we serve.

Our AI Coding Workflow

Human: Identifies the problem, talks to neighbors, frames the need

Human: Designs the solution approach and user experience

AI: Assists with implementation, code generation, debugging

Human: Reviews, tests, refines, and makes final decisions

"Technology is not neutral. It reflects the values of its creators. By keeping human judgment at the center of our design process, we ensure our tools serve community values, not algorithmic ones."

What AI Won't Do:

  • Decide what problems to solve
  • Replace community input
  • Make ethical decisions for us
  • Operate without human oversight

Ready to Build With Us?

We're always looking for community members, developers, and civic leaders who want to experiment with new approaches to local challenges.

Try the AI Playground