R Shiny - From Conception to the Cloud

on YouTube ↗14m

" summary

Team publishes a Shiny app to AWS using GitLab CI/CD pipelines and a repeatable infra framework, learns the hard way after about ten failed deploys. Development uses golem framework with one Shiny module per dashboard (five dashboards), dependency management via renv or packrat and local SQLite during prototyping before switching to Amazon Redshift. Secrets move from hardcoded keys to environment variables, packaging plus unit tests catch missing dependencies, and separate dev/staging/production environments enable safer rollouts.

AI-written summary · about

" chapters (8)

1. 0:00Start
2. 2:43How to code together?
3. 4:05Step by Step with Golem
4. 5:16Dependency Management
5. 7:13Distribution of the workload
6. 7:48Integrating the Backend
7. 8:59Keep it Secure!
8. 11:09Build, release, run

" related talks

• 14mR Shiny - From Conception to the CloudIvonne Carrillo Dominguez · rstudio::conf(2022)
• 15mA Robust Framework for Automated Shiny App TestingSydeaka Watson · rstudio::conf(2022)
• 19mThe Road to Easier Shiny App DeploymentsLiam Kalita · posit::conf(2023)
• 01h 32mCreating Production-Ready Shiny AppsMario Annau · DSC DACH 2023
• 29mProductionizing R scripts in the cloudGergely Daroczi · satRday Los Angeles 2019
• 20mWhat I learned from running Shiny in productionLuis de Sousa · satRday Johannesburg 2019
• 05mCI/CD Pipelines - Oh, the Places You'll Go!Trevor Nederlof · posit::conf(2023)
• 20mA New Era for Shiny-based Clinical Submissions using WebAssemblyEric Nantz · posit::conf(2024)