Use this page as a practical starting point for working on shared HPC systems.
The goal is to cover the basics researchers need early: connecting to clusters, organizing jobs, using schedulers responsibly, and avoiding common performance and reproducibility traps.