hm. ml people love static evals and such, but have you considered approaches that typically appear in saas? (slow-rollouts, org/user constrained testing pools

hm. ml people love static evals and such, but have you considered approaches that typically appear in saas? (slow-rollouts, org/user constrained testing pools with staged rollouts, real-world feedback from actual usage data (where privacy policy permits)?