[Demo] AI-powered UX enhancement: Aligning GitHub documentation with USWDS at Austin Public Library
Summary
In this session, I will explore the groundbreaking approach taken by the Austin Public Library's Innovation Lab in integrating AI with GitHub for UX enhancement. Our primary challenge was to streamline and enrich the documentation process, adhering to the complex guidelines of the U.S. Web Design System (USWDS) while ensuring the content was accessible and user-friendly. Leveraging AI, we developed a system that not only automates the creation of diverse documentation formats like Markdown, but also ensures alignment with USWDS standards. This innovative method significantly improved the efficiency of our documentation process, making it more coherent and navigable. A key highlight is our recent revamp of the Breadcrumbs navigation strategy, demonstrating AI's role in enhancing user experience and interface design. The session showcases how AI can transform UX practices, offering practical insights for UX professionals looking to leverage AI in their work. Takeaways Insight into the innovative use of AI for streamlining documentation processes, adhering to USWDS guidelines in a public library setting Practical examples of how AI can transform UX practices, with a focus on creating efficient, user-friendly, and standards-compliant documentation Demonstration of the AI-assisted update of the Breadcrumbs navigation strategy, demonstrating tangible improvements in UX design and user navigation Strategies for leveraging AI to enhance collaboration among UX designers, developers, and content creators, fostering a more integrated and effective workflow Tips for utilizing tools like GitHub Pages for rapid deployment and continuous integration, highlighting the role of AI in accelerating project development
Key Insights
-
•
Long-term stewardship of data requires not just preservation but contextualization and knowledge transfer.
-
•
AI, particularly ChatGPT, can effectively assist non-expert coders in writing, debugging, and documenting Python scripts.
-
•
Using markdown for documentation makes technical details accessible and readable in web browsers, aiding future users.
-
•
Ethical data stewardship involves openness, transparency, reuse, and modularity aligned with the US Web Design System principles.
-
•
Historical terminology in data (e.g., 'lunatic asylum') needs sensitive updating to reflect modern language while preserving discovery.
-
•
GitHub serves as a collaborative and version-controlled platform that can host data, scripts, and documentation for lasting accessibility.
-
•
Combining scrapers written in Python with AI enables efficient extraction and structuring of public data from websites into CSV.
-
•
Incremental iteration on scripts and documentation over time supports maintainability and adaptability.
-
•
Creating custom GPTs with preset instructions facilitates consistent documentation and adherence to standards.
-
•
Mentorship, community groups, and AI tools together help lower barriers for people new to coding and data projects.
Notable Quotes
"We don’t just want to hand over the data as a black box; we want to provide contextual information to preserve knowledge."
"I’m not a Python developer, I’m getting a lot of assistance from ChatGPT to help me write that."
"Markdown is very concise and useful, and GitHub can render it into human-readable, nicely formatted pages."
"We’re trying to do this ethically, calling upon the US Web Design System principles: be open, transparent, reuse and improve."
"If people are looking for lunatic asylum or Austin State Hospital, they should both be findable, but presented appropriately."
"You can copy and paste error messages into ChatGPT, and it can help you fix your scripts quickly."
"Public data has a life of its own; it is sometimes one thing, sometimes another, depending on context and display."
"This project is about creating community and enabling discovery, and that’s what makes us successful ethical data stewards."
"OpenAI’s custom GPTs allow adding files and instructions, making AI a real partner in documentation and governance."
"Find a mentor—an AI, a person, or a project—and use community resources like Open Austin to learn and grow."
Or choose a question:
More Videos
"You are all very excited about the future of design ops because you are charting the course together."
Alana WashingtonTheme 3 Intro
October 1, 2021
"Working at a government institution isn’t linear. Some projects might work out. Some simply won’t."
Sofía Delsordo Kassim VeraPublic Policy for Jalisco's Designers to Make Design Matter
December 8, 2021
"The most important part of asking for feedback is not just the feedback itself, but introducing folks to the ideas and getting them on board."
Crystal PhilcoxThe Many Faces of Operations
November 6, 2017
"Accessibility is not just one person’s responsibility; it must be distributed through the whole product team."
Sam ProulxDesigning For Screen Readers: Understanding the Mental Models and Techniques of Real Users
December 10, 2021
"Our principle is that whatever we build must help researchers avoid becoming a silo."
Sofia QuinteroThe Product Philosophy Behind EnjoyHQ
March 10, 2021
"In 2023, design leaders were being restructured out and design relegated to a tactical, downstream function."
Doug PowellDesignOps and the Next Frontier: Leading Through Unpredictable Change
September 11, 2025
"Insights are unexpected discoveries about how things work and how to make them work better - Gary Klein."
Johanna KollmannInsights-Driven Product Strategy: Get your Research to Count
December 6, 2022
"ML ops offers governance, accountability, and a clear stakeholder responsibility framework borrowed from DevOps."
Joel BranchHumanizing AI: Filling the Gaps with Multi-faceted Research
March 11, 2021
"The promise of an LLM assistant requires users to build new mental models while using it."
Katie JohnsonDisrupting generative AI products with just-in-time consumer insights
June 4, 2024
