Iter8 is a metrics-driven release optimizer built for DevSecOps, MLOps, SRE and data science teams. Iter8 makes it easy to ensure that new versions of apps and ML models perform well, are secure, and maximize business value.
Iter8 experiments make it simple to collect performance, risk, and business metrics for apps and ML models, assess, compare and validate one or more app/ML model versions, promote the winning version, and maximize business value in each release.
Iter8 experiment charts are Helm charts under the covers, and enable simple, declarative, and reusable metrics-driven experiments. Iter8 combines experiment charts with user supplied values to generate runnable
experiment.yaml files or Kubernetes experiment manifests. The former is used for running experiments in the local environment, while the latter is used for running experiments inside Kubernetes.
Iter8 uses experiment charts located in the Iter8 GitHub repo by default. You can create, package and host Iter8 experiment charts in any GitHub repo and use them with Iter8 CLI, and other Iter8 components.
Iter8 is written in
go and builds on a few awesome open source projects including: