Rubric design, moderation, and what actually breaks when you try to automate grading in a real examination cell.
Most rubrics are written for someone who already knows the subject. Making one explicit enough for a machine turns out to make it fairer for students too.
Fully automated grading keeps failing in higher education for reasons that have little to do with model accuracy. Here is what actually breaks, and the design that avoids it.
Everything we write about here, we also do. If you want it applied to your own situation, start here.