Rumo

• Develop and maintain model training, validation and inference pipelines. • Design feature and data pipelines that support machine learning applications in production. • Convert models developed in exploratory environments into reproducible, scalable pipelines. • Integrate machine learning solutions into the organization’s data ecosystem. • Collaborate with engineering, DevOps and DataOps teams to evolve the platform and the organization’s machine learning practices. • Work closely with data scientists to operationalize analytical models. • Support the transition of models from exploratory environments to production pipelines. • Build pipelines that enable recurring training, updating and model evolution. • Ensure models can run at scale and meet performance requirements for operational use. • Prepare, structure and transform the data required for model training and execution. • Develop data pipelines and feature engineering focused on machine learning applications. • Work with data engineering to ensure the availability and quality of the data used by models. • Implement monitoring mechanisms for model performance, stability and behavior in production. • Support the identification of model degradation and the need for retraining. • Contribute to the continuous improvement of machine learning solutions in production. • Ensure software development best practices are applied to machine learning solutions. • Use code versioning, code review and engineering practices for developing ML pipelines. • Contribute to the standardization and evolution of the area’s machine learning engineering practices.