How did you build the DevOps/CI/CD pipeline? How did you register new models in the catalog?

To build a DevOps/CI/CD pipeline, I usually use tools such as GitLab CI, Jenkins, or GitHub Actions. The main stages include:

1. Build — compiling and assembling artifacts (e.g., Go binaries).
2. Testing — running unit and integration tests.
3. Code quality analysis — static analysis, linters.
4. Deployment — automatic release to testing or production environments.

For registering new models in a catalog (e.g., machine learning models), I used specialized services or databases where each model is described with metadata: version, date, parameters, quality metrics. MLflow or a custom catalog with REST API for registration and search were often used.

An example of a simplified model registration step in a Go pipeline:

func RegisterModel(modelName, version string, metrics map[string]float64) error {
    // Send a POST request to the model catalog service
    payload := map[string]interface{}{
        "name": modelName,
        "version": version,
        "metrics": metrics,
    }
    // ... code to send HTTP request
    return nil
}

Thus, the pipeline ensures automation from code to deployment and registration of models in a unified process.