Brand your docs, apps, and ggplots using LLMs

From R+AI 2025, hosted by R Consortium Abstract A single _brand.yml file enables you to consistently use company branding across all of your projects. However, it takes a significant amount of time to develop a _brand.yml file and {ggplot2} scales & themes that follow company branding. In this talk, I’ll introduce you to the createBranding shiny app that enables you to rapidly create and iterate over _brand.yml and {ggplot2} goodies using Large Language Models (LLMs). The app takes in the screenshot(s) of company branding guidelines and user instructions, provides them to the Google Gemini model via {ellmer}, and generates an editable _brand.yml file. Subsequently, the app creates a colour palette and {ggplot2} scales & theme based on the saved _brand.yml file. This workflow is possible due to the structured output feature of {ellmer}. Bio Umair Durrani is a Data Scientist at Presage Group Inc., where he conducts scientific studies in the field of human factors and decision-making. Umair also develops web applications and data tools for optimizing data collection and reporting. With a PhD in Civil Engineering, he is passionate about modeling and mitigating human error in transportation systems. Umair also teaches data analysis workshops and frequently gives talks on topics such as R programming, Shiny apps, and open-source tools for transportation research. R Consortium Resources Main R Consortium Site: https://www.r-consortium.org/ R+AI website: https://rconsortium.github.io/RplusAI_website R Consortium webinars: https://r-consortium.org/webinars/webinars.html Blog: https://r-consortium.org/blog/ LinkedIn: https://www.linkedin.com/company/r-consortium/