I'm working on a project looking to solve a similar problem, but using a very different mechanism to provide git-like semantics for any database that runs in a

I'm working on a project looking to solve a similar problem, but using a very different mechanism to provide git-like semantics for any database that runs in a container (https://titan-data.io https://titan-data.io). We support local docker containers today and will be adding k8s support in the coming weeks to run in any Kubernetes cluster using k8s snapshots (e.g. push data state from CI/CD k8s cluster to S3 bucket, and clone locally later for forensics). Each automation approach (SQL, dumpfiles, or storage) carries different tradeoffs, but we believe there's a real problem facing developers when it comes to managing data within their workflows. Would love to learn more about what some of the killer use cases are, and how these approaches can complement each other.