How I use AI as a Product Designer

Designers can utilize AI, specifically Large Language Models (LLMs) and Generative AI, beyond creating images, illustrations, or videos. The examples below showcase how I used AI to streamline processes, enhance product quality, and benefit stakeholders.

  • Component Library Documentation

  • Legacy Error Message Analysis

  • Product Design Job Family Framework

  • SaaS Table Design Meta-Analysis

 

Component Library Documentation

Working with MUI design components in a React-based development environment, I leveraged both LLMs and Generative AI to generate structured guidelines for each component, aligning them with our specific business needs. The model provided a high-level overview of the component’s function, five best practices (Do’s and Don’ts) for its use, and suggested possible alternatives—while reflecting the complexity of our enterprise use cases.

Outcome

  • Context-aware examples tailored to our platform and industry.

  • Cross-functional teams gained a shared vocabulary and understanding of design decisions.

  • Ensured a consistent approach to UI component usage.

  • Enabled more effective collaboration between design, development, and product teams.

⚡ How I used Generative AI to speed up design system creation—read now on Medium!

 

Dialog Messages Analysis

I used a LLM to extract dialog messages embedded in a desktop application codebase, analyze clarity, and recommend improvements. Additionally, the model suggested appropriate UI components and categorized messages by severity.

Outcome

  • Identified and extracted over 300 error messages hidden in the code.

  • Delivered structured recommendations to enhance the most critical errors.

  • The project team accelerated the process of refining error messages, completing in days not weeks.

🍾 How I used Generative AI to analyze dialog messages - read now on Medium!

 

Product Design Job Family Framework

To support career growth within our Product Design team, I utilized an LLM to create a structured job family framework. The model helped define skill progression, aligning with both company HR policies and broader industry standards.

Outcome

  • HR teams gained a clearer understanding of UX career paths.

  • Designers received well-defined career progression guidelines.

  • Leadership could communicate growth opportunities effectively.

  • Established a structured, scalable UX job framework.

💡 I used ChatGPT to rethink job families in DesignOps—more insights on Medium!

 

SaaS Table Design Meta-Analysis

I gathered best practices for data table design from the past 15 years and used Generative AI to analyze recurring themes and emerging trends. This meta-analysis informed our approach to designing complex data grids within our SaaS platform.

Outcome

  • The end-users benefited from a more intuitive and efficient data presentation.

  • The design team, gained insights into modern, best-practice table design methodologies.

  • Developed complex table designs aligned with industry best practices.

  • Reduced cognitive load and improved usability for data-heavy workflows.

🚀 I analyzed 26 expert articles on SaaS data table design—read my full breakdown on Medium!

 

Conclusion

By leveraging LLMs and Generative AI in these diverse applications, I was able to drive efficiency, improve collaboration, and enhance product quality. These AI-driven initiatives not only streamlined workflows but also empowered various teams, ensuring design decisions were well-informed and impactful.