CI/CD Tools are very commonly used by developers. MLOps is widely talked about and used to make the practice of deploying, managing, and monitoring machine learning models in production easier. What if you could harness the power of MLOps easily and quickly for your projects?
In this talk, we start by showing the core processes that consist of an MLOps lifecycle and start out with an example of training and monitoring a model on changes to source code which could allow you to right off the back write GitHub Actions that could orchestrate a machine learning pipeline, accumulate the results, and report them on a pull request or check whether a model should be retrained or the impact any changes you make has to your Machine Learning models. We also talk about the role continuous deployment plays in the MLOps lifecycle and show how one could set up continuous deployment for ML models with GitHub Actions. We also show the audience how they could extend the capability of CI/CD in this context by creating and using end-to-end pipelines with Kubeflow or Argo with GitHub Actions.