A skill-quality gate that tests skills like code with planted-bad evals, held-out checks and rollback on regression; proposals only.
Share on X Original post See it on the shelf @JonnyCakeSouza on X
Not affiliated with SpaceXSI.
A skill-quality gate that tests skills like code with planted-bad evals, held-out checks and rollback on regression; proposals only.
Share on X Original post See it on the shelf @JonnyCakeSouza on X