Per-Role Breakdown

Pass rates and metric means split by role template (sorted weakest first)

Role	N	Pass	Correctness	Completeness	Context	Context	Opening	Voice	Faithfulness
llm_genai_engineer	22	55%	0.79	0.65	0.81	0.75	0.40	0.62	0.89	Inspect →
backend_engineer	25	64%	0.83	0.72	0.84	0.81	0.47	0.65	0.92	Inspect →
mlops_engineer	18	67%	0.82	0.74	0.85	0.80	0.48	0.68	0.91	Inspect →
ml_engineer	28	71%	0.85	0.78	0.88	0.82	0.55	0.71	0.93	Inspect →
frontend_engineer	15	73%	0.81	0.76	0.85	0.81	0.52	0.69	0.91	Inspect →
sre	14	79%	0.85	0.78	0.86	0.83	0.55	0.72	0.92	Inspect →
data_engineer	20	80%	0.87	0.82	0.88	0.84	0.60	0.74	0.93	Inspect →