Evaluation Dashboard
Deterministic screening and use-case-aware routing for LLM outputs in regulated pharmaceutical R&D.
0Total evaluations
0Auto-pass
0Awaiting review
0Auto-fail
0SME approved
0SME rejected
0%Mean overall score
Score drift (KS test)
Drift monitoring activates once at least 8 evaluations are recorded (KS two-sample test on rolling windows).