Skip to content
#

blind-grading

Star

Here are 2 public repositories matching this topic...

owieschon / Trust.Ladder

Star 0

TrustLadder: an offline, stdlib-only measurement engine for a preregistered study of whether forcing an AI coding agent through a real enforcement gate cuts its shipped-defect rate. Signs hash-chained run-records, grades them blind behind a calibration gate, and runs frozen statistics that refuse to run until validity gates pass.

python bootstrap statistics ed25519 reproducibility ai-safety experiment-design ai-agents preregistration hash-chain conflict-of-interest llm-evaluation fail-closed stdlib-only blind-grading
  • Updated Jun 25, 2026
  • Python

Improve this page

Add a description, image, and links to the blind-grading topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the blind-grading topic, visit your repo's landing page and select "manage topics."

Learn more