My career started as a Data Scientist, but as a necessity from the company I used to work for (and because it was a subject of my interest), I shifted to an MLOps hat. We didn't have many people in that role at the time, but it was really interesting to make this move.

Most of my time, I spent trying to better understand how to "approximate" the Data Science/ML workloads to the Development stack of the company, so I spent a lot of time learning about containerization, and multiple ways to deploy those artifacts. On top of that, I started learning about the CI/CD stack, and introducing the CT (Continuous Training), by tracking metrics of the live models that were being served, and triggering data-drift alerts.

Most of my work was done using Python, and the FastAPI library, and the containerization was done mostly using Docker, but I had to gain an understanding of how to deploy it in cloud environments, at the time it was really valuable to learn Terraform to understand how to use Infrastructure as Code.