Building GitLab with GitLab: Why there is no MLOPs without DevSecOps

The Data Science team at GitLab is using the GitLab DevSecOps Platform to enhance experiment reproducibility, automate training of ML models, and leverage ML experiment tracking. The team is "dogfooding" the platform by building a container image with all the necessary dependencies under version control and re-pulling it whenever a new version of the code is run. This process ensures standardization and reproducibility. The team also utilizes GitLab CI/CD and GPU-enabled runners for faster model training. By automating these processes and leveraging GitLab's DevSecOps features, the team is able to address challenges in MLOps and improve their ML workflow development.