I worked to improve the data quality and helped with the automation of ML model training as well as the observability of the models. I participated in projects developing Machine Learning models to solve classification and regression problems with Scikit-Learn and deep learning problems with TensorFlow.

With the data science team, I applied and generalized statistical models to large datasets with Python, Scikit-Learn, and a lot of SQL. Other activities I got involved in were: Construction of production pipelines for ML using Airflow, Kubeflow, and SageMaker (implementing continuous training).

I was responsible for designing and migrating the top Ruby on Rails applications previously on AWS to Google Cloud, using Docker and Kubernetes (GKE). During the migration, I worked with eBPF for the first time and applied performance profiling for some applications.

The automatic deploy pipeline that did not exist before was implemented using Codeship and Google Cloud Build. As a result of this work, it was possible to guarantee an immutable and scalable infrastructure.