Project Immortality: Using GitHub To Make Your Work Live Forever
summary
Transforms an R script into a package with DESCRIPTION + MIT license and functions so dependencies install reproducibly, trains an XGBoost model (ffopportunity) that predicts fantasy points, then automates weekly/nightly runs with a 20-line GitHub Actions workflow that installs R and runs the script. Avoids repo bloat (GitHub ~75 MB per file, 2 GB per repo; one example grew to 6 GB) by storing large outputs in GitHub Releases / piggyback where files can be ~2 GB and versioned separately.
AI-written summary · about
chapters (7)
- 0:00Project Immortality intro
- 2:25Why projects die
- 4:34Who this talk is for
- 6:43Reproducible with R packages
- 9:35Automating predictions
- 11:01Storing data efficiently
- 16:26Summary
related talks
17mProject Immortality: Using GitHub To Make Your Work Live Forever
05mMaking Awesome Automations with GitHub Actions
30mSpend More Time Coding & Automate the Boring Stuff
21mGetting the Most Out of Git
25mAutomate Maintainer Tasks With GitHub Actions
37mHandling User Feedback: A Case Study With GitHub Actions Manager
45mHow GitHub builds GitHub on GitHub
01h 06mHow GitHub delivers GitHub using GitHub