jcraft

We've seen customers do something very similar and we've helped them get privacy-safe data into each preview. Handwaving a lot here, but the steps are 1) anonymize production data to a destination, 2) snapshot the anonymized data, and 3) replicate the anonymized data to all of the previews.

> but the data part is hard.

I'd love to learn more about the challenges you have with data. My company essentially provides data-anonymization-as-a-service and we integrate with ephemeral environments, but we could probably improve the integrations.

Big if, but if you can use database containers it's relatively low cost to spin them up in a namespace, load data from a database snapshot, run the migration, and then tear down the container or even the entire namespace.

My company can anonymize production data while preserving the utility of the data. You get privacy-safe data that looks just like production data with the same distributions. It's real data.

The typical workflow is to anonymize production data, snapshot it, and make it available to preview and testing environments as replicas. It's pretty fast.

There are ways to anonymize production data and make it available in preview and testing environments. My company has been doing that for customers and it has been effective. You get privacy-safe data that is still very close to production data.

Depending on the application architecture, you can spin up database containers and seed them with replicas of the anonymized data.

If the data contains sensitive information about individuals (PII), you can't replicate that across all environments if you're in a regulated industry. There are ways to anonymize the data, eliminating the PII risks while preserving data utility so testing is accurate.

This is fantastic; this was always a hunch I had but you've put metrics behind the cost of unreviewed code. We just signed up two more of our repos for analysis.