I've worked with several teams who’ve run into the same issue: testing at the feature level becomes hard to scale when infra gets expensive and clunky to manage.

My group has addressed this by letting teams spin up test environments on Kubernetes, on demand, and run any kind of test, without committing to heavy pre-provisioned infrastructure. It’s been a practical way for larger engineering orgs to reduce test costs while speeding up delivery.

Let me know if you’d be open to a quick walkthrough or would prefer some examples from others who’ve rolled it out. Overview here: https://www.youtube.com/watch?v=ECY3ebzeCOs