Featurevisor

Cloud native architecture

Build and distribute Featurevisor datafiles with your choice of CI pipeline and static file hosting. Applications evaluate the loaded configuration locally.

What is cloud native?

Cloud native applications use infrastructure that can be provisioned and scaled as demand changes. Reliability still depends on the application's design, deployment and recovery practices.

How Featurevisor fits

  • Features and variables are declared in files, with the conditions under which their values change.
  • Changes are pull requests, which your GitOps pipeline lints and tests before deployment.
  • After a merge to main, CI builds datafiles and publishes them to your hosting or CDN.
  • Applications load a datafile into an SDK, and each evaluation uses it without a network request.

Featurevisor itself runs no service. At runtime, everything an application needs is a static file on your own hosting, so there is no Featurevisor uptime to depend on and nothing between your applications and your users.

What your team still owns

Choose repository permissions, review requirements and CI checks. Featurevisor validates definitions, but it does not enforce your Git host's access policies.

Plan how datafiles reach applications:

  • Choose hosting capacity and cache settings for your traffic.
  • Decide when each application fetches and refreshes its datafile.
  • Handle failed downloads and keep a usable fallback.
  • Test how you restore an earlier configuration if a change causes a problem.

Browser and mobile datafiles are inspectable. Keep secrets out of them and enforce authorization on your server.

Start with deployment for publishing datafiles, SDK guides for loading them, and configuration rollback for recovery.